fdo-mirrors/libinput
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/libinput/libinput.git synced 2025-12-24 22:50:05 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
libinput/test/test-touchpad-tap.c

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

5576 lines
146 KiB
C
Raw Normal View History

test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
/*
* Copyright © 2014-2015 Red Hat, Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
#include <config.h>
#include <errno.h>
#include <fcntl.h>
#include <libinput.h>
#include <unistd.h>
#include "libinput-util.h"
#include "litest.h"
START_TEST(touchpad_1fg_tap)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
test: fix some empty queue handling Use the litest_assert_empty_queue() instead of manual checking, and remove the manual checks after the function call. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-11 16:17:33 +10:00
litest_assert_empty_queue(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_doubletap)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t oldtime, curtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers_1st"),
nfingers2 = litest_test_param_get_i32(test_env->params, "fingers_2nd");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0, button2 = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers2 > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
switch (nfingers2) {
case 1:
button2 = BTN_LEFT;
break;
case 2:
button2 = BTN_RIGHT;
break;
case 3:
button2 = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers2);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
/* one to three finger down, all fingers up, repeat with possibly
different number of fingers -> two button event pairs */
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
msleep(10);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
msleep(10);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers2) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
msleep(10);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers2) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event, button, LIBINPUT_BUTTON_STATE_PRESSED);
oldtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
event = libinput_get_event(li);
ptrev = litest_is_button_event(event, button, LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_lt(oldtime, curtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event, button2, LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_lt(oldtime, curtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
oldtime = curtime;
event = libinput_get_event(li);
ptrev = litest_is_button_event(event, button2, LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_lt(oldtime, curtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_multitap)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t oldtime = 0, curtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: enable drag lock for multitap tests Without this enabled, we stay in the single/double tap part of the state machine and a triple tap is just a double tap followed by a single tap. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:34:56 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
msleep(10);
}
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
for (ntaps = 0; ntaps <= range; ntaps++) {
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_ge(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
oldtime = curtime;
}
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_multitap_n_drag_move)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t oldtime = 0, curtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: enable drag lock for multitap tests Without this enabled, we stay in the single/double tap part of the state machine and a triple tap is just a double tap followed by a single tap. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:34:56 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
msleep(10);
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 70, 50, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
touchpad: fix the number of button clicks in multitap tap-tap-down-move should emit 1 click + press, not 2 clicks + press https://bugs.freedesktop.org/show_bug.cgi?id=92016 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-09-18 09:14:43 +10:00
for (ntaps = 0; ntaps < range; ntaps++) {
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_ge(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
oldtime = curtime;
}
event = libinput_get_event(li);
ptrev = litest_is_button_event(event, button, LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_multitap_n_drag_2fg)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t oldtime = 0, curtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
touchpad: disable MT for all semi-mt devices Synaptics, Elantech and Alps semi-mt devices all have issues with reporting correct MT data, even the bounding box which semi-mt devices are supposed to report is wrong. Synaptics devices have massive jumps with two fingers down. Elantech devices may open slots without coordinate data. Alps devices may send 0/0 coordinates as initial slot position. All these may be addressable with specific quirks, but the actual benefit is largely restricted to better palm detection (though even with quirks this is unlikely to work) and support for pinch gestures (again, lack of coordinates makes supporting those hard anyway). Elantech: https://bugs.freedesktop.org/show_bug.cgi?id=93583 Alps: https://bugzilla.redhat.com/show_bug.cgi?id=1295073 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-19 09:05:31 +10:00
if (libevdev_has_property(dev->evdev, INPUT_PROP_SEMI_MT))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: pretend the jumpy semi-mt touchpad is a single-touch touchpad The first finger is accurate, it's just the second finger that is imprecise, so we can't handle it as a true touch. Instead, revert the device back to being a single-touch touchpad and use the fake touch bits for second finger handling. Two-finger scrolling thus becomes usable though we will lose out on other features like thumb detection. Useful scrolling trumps that though. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-07-30 11:54:38 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: enable drag lock for multitap tests Without this enabled, we stay in the single/double tap part of the state machine and a triple tap is just a double tap followed by a single tap. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:34:56 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
msleep(10);
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
touchpad: permit only one finger on the touchpad at the start of a tap-drag This only affects the actual dragging part of the tap-and-drag interaction; n-finger tap-and-drag is supposed to be performed with a n-finger tap followed by a 1-finger drag. Allowing a second finger in the middle of a drag is still necessary for a finger swap, which users may need in long-distance drags, especially when drag-lock is disabled. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-12 12:09:07 +01:00
litest_touch_move_to(dev, 0, 50, 50, 30, 50, 10);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 1, 70, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
touchpad: fix the number of button clicks in multitap tap-tap-down-move should emit 1 click + press, not 2 clicks + press https://bugs.freedesktop.org/show_bug.cgi?id=92016 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-09-18 09:14:43 +10:00
for (ntaps = 0; ntaps < range; ntaps++) {
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_ge(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
oldtime = curtime;
}
event = libinput_get_event(li);
ptrev = litest_is_button_event(event, button, LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 1, 70, 50, 90, 50, 10);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_multitap_n_drag_click)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t oldtime = 0, curtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: enable drag lock for multitap tests Without this enabled, we stay in the single/double tap part of the state machine and a triple tap is just a double tap followed by a single tap. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:34:56 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
msleep(10);
}
litest_touch_down(dev, 0, 50, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: restore the non-debounced litest_button_click() 8cf6893 removed it to make search/replace easier, restore it for the tests where we don't want debouncing to automatically be handled. Still left in place are the various top software button cases. Because of the button re-routing through the fallback interface we need those to be debounced. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-20 10:49:08 +10:00
litest_button_click(dev, BTN_LEFT, true);
litest_button_click(dev, BTN_LEFT, false);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
for (ntaps = 0; ntaps <= range; ntaps++) {
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_ge(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
oldtime = curtime;
}
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_multitap_timeout)
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t ptime, rtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
litest_enable_tap(dev->libinput_device);
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
msleep(10);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
msleep(10);
}
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
for (ntaps = 0; ntaps <= range; ntaps++) {
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
LIBINPUT_BUTTON_STATE_PRESSED);
ptime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
LIBINPUT_BUTTON_STATE_RELEASED);
rtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_lt(ptime, rtime);
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
}
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_multitap_n_drag_timeout)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t oldtime = 0, curtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: enable drag lock for multitap tests Without this enabled, we stay in the single/double tap part of the state machine and a triple tap is just a double tap followed by a single tap. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:34:56 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
msleep(10);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
msleep(10);
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
touchpad: fix the number of button clicks in multitap tap-tap-down-move should emit 1 click + press, not 2 clicks + press https://bugs.freedesktop.org/show_bug.cgi?id=92016 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-09-18 09:14:43 +10:00
for (ntaps = 0; ntaps < range; ntaps++) {
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
oldtime = curtime;
}
event = libinput_get_event(li);
ptrev = litest_is_button_event(event, button, LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 70, 50, 10);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_multitap_n_drag_high_delay)
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
litest_enable_tap(dev->libinput_device);
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
/* Tap timeout is 180ms after a touch or release. Make sure we
* go over 180ms for touch+release, but stay under 180ms for
* each single event. */
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
msleep(100);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
msleep(100);
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
litest_touch_down(dev, 0, 50, 50);
litest_touch_move_to(dev, 0, 50, 50, 50, 70, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
for (ntaps = 0; ntaps < range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
}
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
touchpad: multitap state transitions use the same timing used for taps Multitap sequences (more than 2 taps) had a 180ms timer set only on press, not on release. New taps within those 180ms could either trigger multitap+drag or another multitap (for N+1 taps), resetting the timer on press once again. If no new tap appears within those 180ms, the sequence was considered complete. This behavior differed from regular taps: for the very first tap of a sequence the timer was set both on touch and on release. The multitap timing caused misdetection of triple-tap-and-drag sequences as the timer was hit frequently. Some of those were correctly detected, others as tripletap only. Changing the timer to be set on press **and** release gives us a more lenient timeout. 180ms for tap-and-drag and 180ms for the next tap down after release. This was also the behavior for the xorg synaptics driver. Note that quadruple-tap-and-drag didn't suffer from this because the timeout resulted in double-tap + double-tap-and-drag. Which has the same user-visible effect.
2019-03-13 21:53:54 -06:00
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_multitap_n_drag_tap)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t oldtime = 0, curtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: move the helper functions to the litest.h file To avoid duplication Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-07-30 14:48:12 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
msleep(10);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
msleep(10);
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
touchpad: fix the number of button clicks in multitap tap-tap-down-move should emit 1 click + press, not 2 clicks + press https://bugs.freedesktop.org/show_bug.cgi?id=92016 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-09-18 09:14:43 +10:00
for (ntaps = 0; ntaps < range; ntaps++) {
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_ge(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
oldtime = curtime;
}
event = libinput_get_event(li);
ptrev = litest_is_button_event(event, button, LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 70, 50, 10);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 0);
litest_touch_down(dev, 0, 70, 50);
litest_touch_up(dev, 0);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_multitap_n_drag_tap_click)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t oldtime = 0, curtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: move the helper functions to the litest.h file To avoid duplication Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-07-30 14:48:12 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
touchpad: send multitap button events with the correct timestamps For multitap, we're one tap behind with the button clicks, i.e. we send the first full click button on the second tap, etc. Remember the timestamps of the touches so we can send the events with the right timestamps. This makes tapping more accurate because the time between taps and various timeouts matter less for double-click detection. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 13:52:07 +10:00
msleep(10);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
msleep(10);
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
touchpad: fix the number of button clicks in multitap tap-tap-down-move should emit 1 click + press, not 2 clicks + press https://bugs.freedesktop.org/show_bug.cgi?id=92016 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-09-18 09:14:43 +10:00
for (ntaps = 0; ntaps < range; ntaps++) {
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_ge(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
oldtime = curtime;
}
event = libinput_get_event(li);
ptrev = litest_is_button_event(event, button, LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 70, 50, 10);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 0);
litest_touch_down(dev, 0, 70, 50);
test: restore the non-debounced litest_button_click() 8cf6893 removed it to make search/replace easier, restore it for the tests where we don't want debouncing to automatically be handled. Still left in place are the various top software button cases. Because of the button re-routing through the fallback interface we need those to be debounced. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-20 10:49:08 +10:00
litest_button_click(dev, BTN_LEFT, true);
litest_button_click(dev, BTN_LEFT, false);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
/* the physical click */
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_n_drag)
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
litest_enable_tap(dev->libinput_device);
test: move the helper functions to the litest.h file To avoid duplication Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-07-30 14:48:12 +10:00
litest_disable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 80, 80, 20);
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 0);
/* don't use helper functions here, we expect the event be available
* immediately, not after a timeout that the helper functions may
* trigger.
*/
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
event = libinput_get_event(li);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_notnull(event);
test: drop unused variable Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-03-28 08:55:35 +10:00
litest_is_button_event(event, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: add non-draglock test and rename draglock-dependent tests Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:59:21 +10:00
libinput_event_destroy(event);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_n_drag_draglock)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: move the helper functions to the litest.h file To avoid duplication Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-07-30 14:48:12 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 80, 80, 20);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
/* lift finger, set down again, should continue dragging */
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 80, 80, 20);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_up(dev, 0);
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_n_drag_draglock_tap)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers_1st"),
nfingers2 = litest_test_param_get_i32(test_env->params, "fingers_2nd");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers2 > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: move the helper functions to the litest.h file To avoid duplication Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-07-30 14:48:12 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 80, 80, 20);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
/* lift finger, set down again, should continue dragging */
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 80, 80, 20);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 0);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers2) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers2) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_n_drag_draglock_tap_click)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: move the helper functions to the litest.h file To avoid duplication Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-07-30 14:48:12 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 80, 80, 20);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 0);
litest_touch_down(dev, 0, 50, 50);
test: restore the non-debounced litest_button_click() 8cf6893 removed it to make search/replace easier, restore it for the tests where we don't want debouncing to automatically be handled. Still left in place are the various top software button cases. Because of the button re-routing through the fallback interface we need those to be debounced. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-20 10:49:08 +10:00
litest_button_click(dev, BTN_LEFT, true);
litest_button_click(dev, BTN_LEFT, false);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
/* the physical click */
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_n_drag_draglock_timeout)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: move the helper functions to the litest.h file To avoid duplication Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-07-30 14:48:12 +10:00
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
START_TEST(touchpad_tap_n_drag_draglock_sticky)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
litest_enable_tap(dev->libinput_device);
litest_enable_drag_lock_sticky(dev->libinput_device);
litest_disable_hold_gestures(dev->libinput_device);
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
}
litest_drain_events(li);
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
_fallthrough_;
case 2:
litest_touch_down(dev, 1, 50, 30);
_fallthrough_;
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
_fallthrough_;
case 2:
litest_touch_up(dev, 1);
_fallthrough_;
case 1:
litest_touch_up(dev, 0);
break;
}
litest_touch_down(dev, 0, 50, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_empty_queue(li);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tapndrag(li);
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
litest_assert_empty_queue(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_up(dev, 0);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_n_drag_2fg)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
/* Test: tap with 1-3 fingers (multiple times), then a 1fg move
* followed by a second finger down and *both* fingers moving.
* This is a special behavior catering for the use-case when a user
* needs a second finger down to "hold" the drag while resetting the
* first finger.
* Even though it's 2fg movement, we expect it to behave like a 1fg
* drag. This behavior may change in the future.
*/
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: move the helper functions to the litest.h file To avoid duplication Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-07-30 14:48:12 +10:00
litest_disable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 30, 70);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 1, 80, 70);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 30, 70, 30, 30, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
START_TEST(touchpad_tap_n_drag_2fg_scroll)
{
/* Test: tap with 1-3 fingers, then immediate 2fg scroll.
* We expect this to be a tap followed by a scroll.
*/
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
litest_enable_2fg_scroll(dev);
litest_enable_tap(dev->libinput_device);
litest_disable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
}
litest_drain_events(li);
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
case 1:
litest_touch_up(dev, 0);
break;
}
/* Two fingers down + move to trigger scrolling */
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_touch_move_two_touches(dev, 50, 50, 70, 50, 0, 20, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
High-resolution scroll wheel support Starting with kernel v5.0 two new axes are available for high-resolution wheel scrolling: REL_WHEEL_HI_RES and REL_HWHEEL_HI_RES. Both axes send data in fractions of 120 where each multiple of 120 amounts to one logical scroll event. Fractions of 120 indicate a wheel movement less than one detent. This commit adds a new API for scroll events. Three new event types that encode the axis source in the event type name and a new API to get a normalized-to-120 value that also used by Windows and the kernel (each multiple of 120 represents a logical scroll click). This addresses a main shortcoming with the existing API - it was unreliable to calculate the click angle based on the axis value+discrete events and thus any caller using the axis value alone would be left with some ambiguity. With the v120 API it's now possible to (usually) calculate the click angle, but more importantly it provides the simplest hw-independent way of scrolling by a click or a fraction of a click. A new event type is required, the only way to integrate the v120 value otherwise was to start sending events with a discrete value of 0. This would break existing xf86-input-libinput (divide by zero, fixed in 0.28.2) and weston (general confusion). mutter, kwin are unaffected. With the new API, the old POINTER_AXIS event are deprecated - callers should use the new API where available and discard any POINTER_AXIS events. Notable: REL_WHEEL/REL_HWHEEL are emulated by the kernel but there's no guarantee that they'll come every accumulated 120 values, e.g. Logitech mice often send events that don't add up to 120 per detent. We use the kernel's wheel click emulation instead of doing our own. libinput guarantees high-resolution events even on pre-5.0 kernels. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2018-11-22 10:24:54 +10:00
litest_assert_only_axis_events(li, LIBINPUT_EVENT_POINTER_SCROLL_FINGER);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
START_TEST(touchpad_tap_n_drag_draglock_2fg_scroll)
{
/* Test: tap with 1-3 fingers, trigger drag-lock,
* then immediate 2fg scroll.
* We expect this to be a tap-and-drag followed by a scroll.
*/
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_enable_2fg_scroll(dev);
litest_enable_tap(dev->libinput_device);
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
}
litest_drain_events(li);
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
case 1:
litest_touch_up(dev, 0);
break;
}
/* Drag with one finger */
litest_touch_down(dev, 0, 50, 50);
litest_touch_move_to(dev, 0, 50, 50, 50, 70, 10);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
/* Release finger to trigger drag-lock */
litest_touch_up(dev, 0);
/* Two fingers down + move to trigger scrolling */
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_touch_move_two_touches(dev, 50, 50, 70, 50, 0, 20, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
High-resolution scroll wheel support Starting with kernel v5.0 two new axes are available for high-resolution wheel scrolling: REL_WHEEL_HI_RES and REL_HWHEEL_HI_RES. Both axes send data in fractions of 120 where each multiple of 120 amounts to one logical scroll event. Fractions of 120 indicate a wheel movement less than one detent. This commit adds a new API for scroll events. Three new event types that encode the axis source in the event type name and a new API to get a normalized-to-120 value that also used by Windows and the kernel (each multiple of 120 represents a logical scroll click). This addresses a main shortcoming with the existing API - it was unreliable to calculate the click angle based on the axis value+discrete events and thus any caller using the axis value alone would be left with some ambiguity. With the v120 API it's now possible to (usually) calculate the click angle, but more importantly it provides the simplest hw-independent way of scrolling by a click or a fraction of a click. A new event type is required, the only way to integrate the v120 value otherwise was to start sending events with a discrete value of 0. This would break existing xf86-input-libinput (divide by zero, fixed in 0.28.2) and weston (general confusion). mutter, kwin are unaffected. With the new API, the old POINTER_AXIS event are deprecated - callers should use the new API where available and discard any POINTER_AXIS events. Notable: REL_WHEEL/REL_HWHEEL are emulated by the kernel but there's no guarantee that they'll come every accumulated 120 values, e.g. Logitech mice often send events that don't add up to 120 per detent. We use the kernel's wheel click emulation instead of doing our own. libinput guarantees high-resolution events even on pre-5.0 kernels. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2018-11-22 10:24:54 +10:00
litest_assert_only_axis_events(li, LIBINPUT_EVENT_POINTER_SCROLL_FINGER);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_n_drag_3fg_btntool)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) > 2 ||
test: fix a bunch of tests expecting BTN_TOOL_TRIPLETAP A device may have 1 or 2 slots without setting BTN_TOOL_TRIPLETAP, those devices will fail those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-05-08 13:59:27 +10:00
!libevdev_has_event_code(dev->evdev, EV_KEY, BTN_TOOL_TRIPLETAP))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 0, 40, 30);
litest_touch_down(dev, 1, 50, 30);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 1);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
break;
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 0);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 30, 70);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 1, 80, 90);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 30, 70, 30, 30, 5);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
/* Putting down a third finger should end the drag */
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 1);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
/* Releasing the fingers should not cause any events */
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 0);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_n_drag_3fg)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
/* Test: tap with 1-3 fingers (multiple times), then a 1fg move
* followed by a second finger down and *both* fingers moving.
* This is a special behavior catering for the use-case when a user
* needs a second finger down to "hold" the drag while resetting the
* first finger.
* Even though it's 2fg movement, we expect it to behave like a 1fg
* drag. This behavior may change in the future.
*/
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) < 3)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
/* 1fg down triggers the drag */
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 30, 70);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
/* 2fg is allowed now without cancelling the drag */
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 1, 80, 90);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 30, 70, 30, 30, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
/* Putting down a third finger should end the drag */
litest_touch_down(dev, 2, 50, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
/* Releasing the fingers should not cause any events */
litest_touch_up(dev, 2);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
START_TEST(touchpad_tap_n_drag_3fg_swipe)
{
/* Test: tap with 1-3 fingers, then immediate 3fg swipe.
* We expect this to be a tap followed by a swipe.
*/
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
unsigned int button = 0;
if (litest_slot_count(dev) < 3)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
}
litest_drain_events(li);
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
case 1:
litest_touch_up(dev, 0);
break;
}
litest_touch_down(dev, 0, 30, 50);
litest_touch_down(dev, 1, 50, 50);
litest_touch_down(dev, 2, 80, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
litest_touch_move_three_touches(dev, 30, 50, 50, 50, 80, 50, 0, 20, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after tap We have two behaviors here: - tap + 2fg -> scrolling - tap + 1fg move + 2f down -> dragging Let's document this. The 3fg case only has one situation, so let's test that. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-15 17:48:34 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_gesture_event(li, LIBINPUT_EVENT_GESTURE_SWIPE_BEGIN, 3);
litest_assert_only_typed_events(li, LIBINPUT_EVENT_GESTURE_SWIPE_UPDATE);
litest_touch_up(dev, 2);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
litest_assert_gesture_event(li, LIBINPUT_EVENT_GESTURE_SWIPE_END, 3);
litest_assert_empty_queue(li);
}
END_TEST
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
START_TEST(touchpad_tap_n_drag_draglock_3fg_swipe)
{
/* Test: tap with 1-3 fingers, trigger drag-lock,
* then immediate 3fg swipe.
* We expect this to be a tap-and-drag followed by a swipe.
*/
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
unsigned int button = 0;
if (litest_slot_count(dev) < 3)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_enable_tap(dev->libinput_device);
litest_enable_drag_lock(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
}
litest_drain_events(li);
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
case 1:
litest_touch_up(dev, 0);
break;
}
/* Drag with one finger */
litest_touch_down(dev, 0, 50, 50);
litest_touch_move_to(dev, 0, 50, 50, 50, 70, 10);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
/* Release finger to trigger drag-lock */
litest_touch_up(dev, 0);
litest_touch_down(dev, 0, 30, 50);
litest_touch_down(dev, 1, 50, 50);
litest_touch_down(dev, 2, 80, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_touch_move_three_touches(dev, 30, 50, 50, 50, 80, 50, 0, 20, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: add test cases for 2/3 finger movement after drag-lock Same as after a tap, just with a short drag between tap and 2/3 finger movement. Also fixes a finger coord typo in one of the previously added test cases. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2021-02-16 11:13:29 +01:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_gesture_event(li, LIBINPUT_EVENT_GESTURE_SWIPE_BEGIN, 3);
litest_assert_only_typed_events(li, LIBINPUT_EVENT_GESTURE_SWIPE_UPDATE);
litest_touch_up(dev, 2);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
litest_assert_gesture_event(li, LIBINPUT_EVENT_GESTURE_SWIPE_END, 3);
litest_assert_empty_queue(li);
}
END_TEST
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
START_TEST(touchpad_2fg_tap)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: fix maybe-uninitialized compiler warnings touchpad-tap.c: In function ‘touchpad_3fg_tap_btntool_inverted’: touchpad-tap.c:1548:2: warning: ‘button’ may be used uninitialized in this function [-Wmaybe-uninitialized] and similar False positive, if button isn't set by now we would've abort()-ed before we even get here. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2016-11-22 10:59:18 +10:00
unsigned int button = 0;
touchpad: fix the button timestamps for double/tripletap Both events had the same timestamp but we have the timestamp from the original event - use it. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-27 14:12:11 +10:00
struct libinput_event *ev;
struct libinput_event_pointer *ptrev;
uint64_t ptime, rtime;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: add optional value names to parametrized tests Only implemented for i32 values so far, used for enums and enum-like constants, replaces a runtime lookup from stringly-typed parameters. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1141>
2025-02-13 22:10:02 +01:00
enum libinput_config_tap_button_map map =
litest_test_param_get_i32(test_env->params, "map");
test: change tap tests to use parameters This also fixes the invocation of the touchpad_3fg_tap_btntool_pointerjump which was written as ranged test but never got invoked with a range (but _i defaults to zero). Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1119>
2025-01-08 09:21:14 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_set_tap_map(dev->libinput_device, map);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
switch (map) {
case LIBINPUT_CONFIG_TAP_MAP_LRM:
button = BTN_RIGHT;
break;
case LIBINPUT_CONFIG_TAP_MAP_LMR:
button = BTN_MIDDLE;
break;
default:
litest_abort_msg("Invalid map range %d", map);
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 70);
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
touchpad: fix the button timestamps for double/tripletap Both events had the same timestamp but we have the timestamp from the original event - use it. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-27 14:12:11 +10:00
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev, button, LIBINPUT_BUTTON_STATE_PRESSED);
ptime = libinput_event_pointer_get_time_usec(ptrev);
libinput_event_destroy(ev);
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev, button, LIBINPUT_BUTTON_STATE_RELEASED);
rtime = libinput_event_pointer_get_time_usec(ptrev);
libinput_event_destroy(ev);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_lt(ptime, rtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_2fg_tap_inverted)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: fix maybe-uninitialized compiler warnings touchpad-tap.c: In function ‘touchpad_3fg_tap_btntool_inverted’: touchpad-tap.c:1548:2: warning: ‘button’ may be used uninitialized in this function [-Wmaybe-uninitialized] and similar False positive, if button isn't set by now we would've abort()-ed before we even get here. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2016-11-22 10:59:18 +10:00
unsigned int button = 0;
touchpad: fix the button timestamps for double/tripletap Both events had the same timestamp but we have the timestamp from the original event - use it. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-27 14:12:11 +10:00
struct libinput_event *ev;
struct libinput_event_pointer *ptrev;
uint64_t ptime, rtime;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: add optional value names to parametrized tests Only implemented for i32 values so far, used for enums and enum-like constants, replaces a runtime lookup from stringly-typed parameters. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1141>
2025-02-13 22:10:02 +01:00
enum libinput_config_tap_button_map map =
litest_test_param_get_i32(test_env->params, "map");
test: change tap tests to use parameters This also fixes the invocation of the touchpad_3fg_tap_btntool_pointerjump which was written as ranged test but never got invoked with a range (but _i defaults to zero). Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1119>
2025-01-08 09:21:14 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_set_tap_map(dev->libinput_device, map);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
switch (map) {
case LIBINPUT_CONFIG_TAP_MAP_LRM:
button = BTN_RIGHT;
break;
case LIBINPUT_CONFIG_TAP_MAP_LMR:
button = BTN_MIDDLE;
break;
default:
litest_abort_msg("Invalid map range %d", map);
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 70);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
touchpad: fix the button timestamps for double/tripletap Both events had the same timestamp but we have the timestamp from the original event - use it. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-27 14:12:11 +10:00
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev, button, LIBINPUT_BUTTON_STATE_PRESSED);
ptime = libinput_event_pointer_get_time_usec(ptrev);
libinput_event_destroy(ev);
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev, button, LIBINPUT_BUTTON_STATE_RELEASED);
rtime = libinput_event_pointer_get_time_usec(ptrev);
libinput_event_destroy(ev);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_lt(ptime, rtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
touchpad: ignore any semi-mt movement in the same frame as a finger release Semi-MT devices provide a bounding box of the fingers, and internally we don't treat them as real MT device. Depending which finger currently provides ABS_X/Y we may get a large jump when the other finger is released. Basic sequence is finger 1 down, finger 2 down, finger 1 up. On the last interaction, the ABS_X/Y which was on finger 1's coordinates now jumps to finger 2's coordinates. This is interpreted as movement by the tapping code, resulting in missed two-finger taps. Ignore these movements on semi-mt devices. https://bugs.freedesktop.org/show_bug.cgi?id=105043 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-15 09:09:28 +10:00
START_TEST(touchpad_2fg_tap_move_on_release)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: ignore any semi-mt movement in the same frame as a finger release Semi-MT devices provide a bounding box of the fingers, and internally we don't treat them as real MT device. Depending which finger currently provides ABS_X/Y we may get a large jump when the other finger is released. Basic sequence is finger 1 down, finger 2 down, finger 1 up. On the last interaction, the ABS_X/Y which was on finger 1's coordinates now jumps to finger 2's coordinates. This is interpreted as movement by the tapping code, resulting in missed two-finger taps. Ignore these movements on semi-mt devices. https://bugs.freedesktop.org/show_bug.cgi?id=105043 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-15 09:09:28 +10:00
litest_drain_events(dev->libinput);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 70);
litest_push_event_frame(dev);
litest_touch_move(dev, 0, 55, 55);
litest_touch_up(dev, 1);
litest_pop_event_frame(dev);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: ignore any semi-mt movement in the same frame as a finger release Semi-MT devices provide a bounding box of the fingers, and internally we don't treat them as real MT device. Depending which finger currently provides ABS_X/Y we may get a large jump when the other finger is released. Basic sequence is finger 1 down, finger 2 down, finger 1 up. On the last interaction, the ABS_X/Y which was on finger 1's coordinates now jumps to finger 2's coordinates. This is interpreted as movement by the tapping code, resulting in missed two-finger taps. Ignore these movements on semi-mt devices. https://bugs.freedesktop.org/show_bug.cgi?id=105043 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-15 09:09:28 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
touchpad: don't tap for 2fg down, followed by a single finger up The following sequence currently generates a right-button event: finger 1 down finger 2 down finger 1 up finger 2 held down This is easily triggered with short scroll events. There are two issues here: first is that the tapping code elsewhere treats any tap with a second finger down as a left-button tap, not a right button one. So if anything, we should generate a left button click here, not a right button click. Arguably, generating a button click here is wrong though, it's not a very well defined sequence and relatively difficult to trigger intentionally. So the best solution here is to simply ignore the release event and move straight back to state HOLD - unless the second finger is released within the timeout. If the finger is set down again during the timeout, we move straight to TOUCH_2_HOLD - this could eventually be interpreted as a tap, but not for now. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-08-31 11:10:58 +10:00
START_TEST(touchpad_2fg_tap_n_hold_first)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: don't tap for 2fg down, followed by a single finger up The following sequence currently generates a right-button event: finger 1 down finger 2 down finger 1 up finger 2 held down This is easily triggered with short scroll events. There are two issues here: first is that the tapping code elsewhere treats any tap with a second finger down as a left-button tap, not a right button one. So if anything, we should generate a left button click here, not a right button click. Arguably, generating a button click here is wrong though, it's not a very well defined sequence and relatively difficult to trigger intentionally. So the best solution here is to simply ignore the release event and move straight back to state HOLD - unless the second finger is released within the timeout. If the finger is set down again during the timeout, we move straight to TOUCH_2_HOLD - this could eventually be interpreted as a tap, but not for now. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-08-31 11:10:58 +10:00
litest_drain_events(dev->libinput);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 70);
litest_touch_up(dev, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: don't tap for 2fg down, followed by a single finger up The following sequence currently generates a right-button event: finger 1 down finger 2 down finger 1 up finger 2 held down This is easily triggered with short scroll events. There are two issues here: first is that the tapping code elsewhere treats any tap with a second finger down as a left-button tap, not a right button one. So if anything, we should generate a left button click here, not a right button click. Arguably, generating a button click here is wrong though, it's not a very well defined sequence and relatively difficult to trigger intentionally. So the best solution here is to simply ignore the release event and move straight back to state HOLD - unless the second finger is released within the timeout. If the finger is set down again during the timeout, we move straight to TOUCH_2_HOLD - this could eventually be interpreted as a tap, but not for now. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-08-31 11:10:58 +10:00
litest_assert_empty_queue(li);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: don't tap for 2fg down, followed by a single finger up The following sequence currently generates a right-button event: finger 1 down finger 2 down finger 1 up finger 2 held down This is easily triggered with short scroll events. There are two issues here: first is that the tapping code elsewhere treats any tap with a second finger down as a left-button tap, not a right button one. So if anything, we should generate a left button click here, not a right button click. Arguably, generating a button click here is wrong though, it's not a very well defined sequence and relatively difficult to trigger intentionally. So the best solution here is to simply ignore the release event and move straight back to state HOLD - unless the second finger is released within the timeout. If the finger is set down again during the timeout, we move straight to TOUCH_2_HOLD - this could eventually be interpreted as a tap, but not for now. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-08-31 11:10:58 +10:00
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_2fg_tap_n_hold_second)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: don't tap for 2fg down, followed by a single finger up The following sequence currently generates a right-button event: finger 1 down finger 2 down finger 1 up finger 2 held down This is easily triggered with short scroll events. There are two issues here: first is that the tapping code elsewhere treats any tap with a second finger down as a left-button tap, not a right button one. So if anything, we should generate a left button click here, not a right button click. Arguably, generating a button click here is wrong though, it's not a very well defined sequence and relatively difficult to trigger intentionally. So the best solution here is to simply ignore the release event and move straight back to state HOLD - unless the second finger is released within the timeout. If the finger is set down again during the timeout, we move straight to TOUCH_2_HOLD - this could eventually be interpreted as a tap, but not for now. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-08-31 11:10:58 +10:00
litest_drain_events(dev->libinput);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 70);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: don't tap for 2fg down, followed by a single finger up The following sequence currently generates a right-button event: finger 1 down finger 2 down finger 1 up finger 2 held down This is easily triggered with short scroll events. There are two issues here: first is that the tapping code elsewhere treats any tap with a second finger down as a left-button tap, not a right button one. So if anything, we should generate a left button click here, not a right button click. Arguably, generating a button click here is wrong though, it's not a very well defined sequence and relatively difficult to trigger intentionally. So the best solution here is to simply ignore the release event and move straight back to state HOLD - unless the second finger is released within the timeout. If the finger is set down again during the timeout, we move straight to TOUCH_2_HOLD - this could eventually be interpreted as a tap, but not for now. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-08-31 11:10:58 +10:00
litest_assert_empty_queue(li);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: don't tap for 2fg down, followed by a single finger up The following sequence currently generates a right-button event: finger 1 down finger 2 down finger 1 up finger 2 held down This is easily triggered with short scroll events. There are two issues here: first is that the tapping code elsewhere treats any tap with a second finger down as a left-button tap, not a right button one. So if anything, we should generate a left button click here, not a right button click. Arguably, generating a button click here is wrong though, it's not a very well defined sequence and relatively difficult to trigger intentionally. So the best solution here is to simply ignore the release event and move straight back to state HOLD - unless the second finger is released within the timeout. If the finger is set down again during the timeout, we move straight to TOUCH_2_HOLD - this could eventually be interpreted as a tap, but not for now. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-08-31 11:10:58 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
START_TEST(touchpad_2fg_tap_quickrelease)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 70);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 0);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 1);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 0);
litest_event(dev, EV_KEY, BTN_TOUCH, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_1fg_tap_click)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
/* Finger down, finger up -> tap button press
* Physical button click -> no button press/release
* Tap timeout -> tap button release */
litest_touch_down(dev, 0, 50, 50);
litest_touch_up(dev, 0);
litest_event(dev, EV_KEY, BTN_LEFT, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_event(dev, EV_KEY, BTN_LEFT, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_2fg_tap_click)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
/* two fingers down, left button click, fingers up
-> one left button, one right button event pair */
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
litest_event(dev, EV_KEY, BTN_LEFT, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_event(dev, EV_KEY, BTN_LEFT, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(clickpad_2fg_tap_click)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
/* two fingers down, button click, fingers up
-> only one button left event pair */
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
litest_event(dev, EV_KEY, BTN_LEFT, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_event(dev, EV_KEY, BTN_LEFT, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_2fg_tap_click_apple)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
/* two fingers down, button click, fingers up
-> only one button right event pair
(apple have clickfinger enabled by default) */
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
litest_event(dev, EV_KEY, BTN_LEFT, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_event(dev, EV_KEY, BTN_LEFT, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_no_2fg_tap_after_move)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
/* one finger down, move past threshold,
second finger down, first finger up
-> no event
*/
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 90, 90, 10);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
litest_touch_down(dev, 1, 70, 50);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_no_2fg_tap_after_timeout)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
/* one finger down, wait past tap timeout,
second finger down, first finger up
-> no event
*/
litest_touch_down(dev, 0, 50, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
litest_touch_down(dev, 1, 70, 50);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_no_first_fg_tap_after_move)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
/* one finger down, second finger down,
second finger moves beyond threshold,
first finger up
-> no event
*/
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(dev->libinput);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 1, 70, 50, 90, 90, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(dev->libinput);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(dev->libinput);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
while ((event = libinput_get_event(li))) {
test: add litest_assert_event_type_not_one_of Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-10-14 18:58:24 +10:00
litest_assert_event_type_not_one_of(event,
LIBINPUT_EVENT_POINTER_BUTTON);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
libinput_event_destroy(event);
}
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_double_tap_click)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
/* finger(s) down, up, one finger down, button click, finger up
-> two button event pairs */
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
litest_event(dev, EV_KEY, BTN_LEFT, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_event(dev, EV_KEY, BTN_LEFT, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_n_drag_click)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
/* finger(s) down, up, one finger down, move, button click, finger up
-> two button event pairs, motion allowed */
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 80, 50, 10);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_event(dev, EV_KEY, BTN_LEFT, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_event(dev, EV_KEY, BTN_LEFT, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_3fg_tap)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: fix maybe-uninitialized compiler warnings touchpad-tap.c: In function ‘touchpad_3fg_tap_btntool_inverted’: touchpad-tap.c:1548:2: warning: ‘button’ may be used uninitialized in this function [-Wmaybe-uninitialized] and similar False positive, if button isn't set by now we would've abort()-ed before we even get here. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2016-11-22 10:59:18 +10:00
unsigned int button = 0;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
int i;
test: add optional value names to parametrized tests Only implemented for i32 values so far, used for enums and enum-like constants, replaces a runtime lookup from stringly-typed parameters. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1141>
2025-02-13 22:10:02 +01:00
enum libinput_config_tap_button_map map =
litest_test_param_get_i32(test_env->params, "map");
test: change tap tests to use parameters This also fixes the invocation of the touchpad_3fg_tap_btntool_pointerjump which was written as ranged test but never got invoked with a range (but _i defaults to zero). Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1119>
2025-01-08 09:21:14 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) < 3)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_set_tap_map(dev->libinput_device, map);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
switch (map) {
case LIBINPUT_CONFIG_TAP_MAP_LRM:
button = BTN_MIDDLE;
break;
case LIBINPUT_CONFIG_TAP_MAP_LMR:
button = BTN_RIGHT;
break;
default:
litest_abort_msg("Invalid map range %d", map);
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
for (i = 0; i < 3; i++) {
touchpad: fix the button timestamps for double/tripletap Both events had the same timestamp but we have the timestamp from the original event - use it. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-27 14:12:11 +10:00
uint64_t ptime, rtime;
struct libinput_event *ev;
struct libinput_event_pointer *ptrev;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
touchpad: for 2/3-finger tap, use the last finger down as press time This makes the tapping times shorter and hopefully more obvious. It also fixes a bug where repeated tripletap (by tapping with one finger while leaving the other two down) could cause incorrect timestamps. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 14:51:06 +10:00
msleep(5);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 1, 70, 50);
touchpad: for 2/3-finger tap, use the last finger down as press time This makes the tapping times shorter and hopefully more obvious. It also fixes a bug where repeated tripletap (by tapping with one finger while leaving the other two down) could cause incorrect timestamps. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 14:51:06 +10:00
msleep(5);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_down(dev, 2, 80, 50);
touchpad: for 2/3-finger tap, use the last finger down as press time This makes the tapping times shorter and hopefully more obvious. It also fixes a bug where repeated tripletap (by tapping with one finger while leaving the other two down) could cause incorrect timestamps. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 14:51:06 +10:00
msleep(10);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_touch_up(dev, (i + 2) % 3);
litest_touch_up(dev, (i + 1) % 3);
litest_touch_up(dev, (i + 0) % 3);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
touchpad: fix the button timestamps for double/tripletap Both events had the same timestamp but we have the timestamp from the original event - use it. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-27 14:12:11 +10:00
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev,
button,
LIBINPUT_BUTTON_STATE_PRESSED);
ptime = libinput_event_pointer_get_time_usec(ptrev);
libinput_event_destroy(ev);
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev,
button,
LIBINPUT_BUTTON_STATE_RELEASED);
rtime = libinput_event_pointer_get_time_usec(ptrev);
libinput_event_destroy(ev);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_lt(ptime, rtime);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
}
}
END_TEST
touchpad: for 2/3-finger tap, use the last finger down as press time This makes the tapping times shorter and hopefully more obvious. It also fixes a bug where repeated tripletap (by tapping with one finger while leaving the other two down) could cause incorrect timestamps. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 14:51:06 +10:00
START_TEST(touchpad_3fg_tap_tap_again)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
int i;
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) < 3)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: for 2/3-finger tap, use the last finger down as press time This makes the tapping times shorter and hopefully more obvious. It also fixes a bug where repeated tripletap (by tapping with one finger while leaving the other two down) could cause incorrect timestamps. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 14:51:06 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: for 2/3-finger tap, use the last finger down as press time This makes the tapping times shorter and hopefully more obvious. It also fixes a bug where repeated tripletap (by tapping with one finger while leaving the other two down) could cause incorrect timestamps. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 14:51:06 +10:00
uint64_t ptime, rtime;
struct libinput_event *ev;
struct libinput_event_pointer *ptrev;
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
msleep(5);
litest_touch_down(dev, 1, 70, 50);
msleep(5);
litest_touch_down(dev, 2, 80, 50);
msleep(10);
litest_touch_up(dev, 0);
msleep(10);
litest_touch_down(dev, 0, 80, 50);
msleep(10);
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
litest_touch_up(dev, 2);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: for 2/3-finger tap, use the last finger down as press time This makes the tapping times shorter and hopefully more obvious. It also fixes a bug where repeated tripletap (by tapping with one finger while leaving the other two down) could cause incorrect timestamps. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 14:51:06 +10:00
for (i = 0; i < 2; i++) {
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev,
BTN_MIDDLE,
LIBINPUT_BUTTON_STATE_PRESSED);
ptime = libinput_event_pointer_get_time_usec(ptrev);
libinput_event_destroy(ev);
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev,
BTN_MIDDLE,
LIBINPUT_BUTTON_STATE_RELEASED);
rtime = libinput_event_pointer_get_time_usec(ptrev);
libinput_event_destroy(ev);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_lt(ptime, rtime);
touchpad: for 2/3-finger tap, use the last finger down as press time This makes the tapping times shorter and hopefully more obvious. It also fixes a bug where repeated tripletap (by tapping with one finger while leaving the other two down) could cause incorrect timestamps. https://bugs.freedesktop.org/show_bug.cgi?id=100796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-28 14:51:06 +10:00
}
}
END_TEST
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
START_TEST(touchpad_3fg_tap_quickrelease)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) < 3)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
litest_touch_down(dev, 2, 80, 50);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 0);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 1);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 2);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 0);
litest_event(dev, EV_KEY, BTN_TOUCH, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_MIDDLE, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_MIDDLE, LIBINPUT_BUTTON_STATE_RELEASED);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
START_TEST(touchpad_3fg_tap_pressure_btntool)
touchpad: only begin fake touches when we have at least one finger down If a single-touch touchpad drops below the pressure threshold in the same frame where a fake finger is added, we begin a fake touch here. The subsequent loop ends this fake touch because real_fingers_down is 0. This causes the tapping code to have a mismatch of how many fingers are down because it never sees the touch begin event for that finger. https://bugs.freedesktop.org/show_bug.cgi?id=105160
2018-02-20 13:27:57 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) >= 3 ||
test: fix a bunch of tests expecting BTN_TOOL_TRIPLETAP A device may have 1 or 2 slots without setting BTN_TOOL_TRIPLETAP, those devices will fail those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-05-08 13:59:27 +10:00
!libevdev_has_event_code(dev->evdev, EV_KEY, BTN_TOOL_TRIPLETAP))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
/* libinput doesn't export when it uses pressure detection, so we
* need to reconstruct this here. Specifically, semi-mt devices are
* non-mt in libinput, so if they have ABS_PRESSURE, they'll use it.
*/
if (!libevdev_has_event_code(dev->evdev, EV_ABS, ABS_MT_PRESSURE))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: only begin fake touches when we have at least one finger down If a single-touch touchpad drops below the pressure threshold in the same frame where a fake finger is added, we begin a fake touch here. The subsequent loop ends this fake touch because real_fingers_down is 0. This causes the tapping code to have a mismatch of how many fingers are down because it never sees the touch begin event for that finger. https://bugs.freedesktop.org/show_bug.cgi?id=105160
2018-02-20 13:27:57 +10:00
litest_enable_tap(dev->libinput_device);
litest_enable_edge_scroll(dev);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: only begin fake touches when we have at least one finger down If a single-touch touchpad drops below the pressure threshold in the same frame where a fake finger is added, we begin a fake touch here. The subsequent loop ends this fake touch because real_fingers_down is 0. This causes the tapping code to have a mismatch of how many fingers are down because it never sees the touch begin event for that finger. https://bugs.freedesktop.org/show_bug.cgi?id=105160
2018-02-20 13:27:57 +10:00
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: only begin fake touches when we have at least one finger down If a single-touch touchpad drops below the pressure threshold in the same frame where a fake finger is added, we begin a fake touch here. The subsequent loop ends this fake touch because real_fingers_down is 0. This causes the tapping code to have a mismatch of how many fingers are down because it never sees the touch begin event for that finger. https://bugs.freedesktop.org/show_bug.cgi?id=105160
2018-02-20 13:27:57 +10:00
litest_drain_events(li);
test: fix an incomplete comment Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:31:35 +10:00
/* drop below the pressure threshold in the same frame as starting a
touchpad: only keep low-pressure fingers alive for 2+-slot touchpads Regression introduced by 3979b9e16a5ed141506d95f80ddfd7b94651dcfa, bug 105258. With that commit, we only ended real touches when we had less than nslots fake fingers down. i.e. tripletap on a 2 slot touchpad would not end the first/second touch even if the pressure goes below the threshold. e.g. Lenovo x270 needs this, see https://bugs.freedesktop.org/attachment.cgi?id=137672, it dips below the pressure threshold for the first slot and ends the second slot in the same frame as the third finger is detected. Fun times. Anyway, this breaks semi-mt touchpads, another fine category of devices, because some of those can detect hovering fingers at low pressure, see bug 105535. Because semi-mt devices are generally garbage, we treat them as single-touch devices instead. So whenever two fingers are down, we treat both as above the pressure threshold, even when they're physicall hovering. Fix this by making the x270 fix conditional on at least 2 slots. https://bugs.freedesktop.org/show_bug.cgi?id=105535 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-21 10:24:12 +10:00
* third touch, see
* E: 8713.954784 0001 014e 0001 # EV_KEY / BTN_TOOL_TRIPLETAP 1
* in https://bugs.freedesktop.org/attachment.cgi?id=137672
*/
touchpad: only begin fake touches when we have at least one finger down If a single-touch touchpad drops below the pressure threshold in the same frame where a fake finger is added, we begin a fake touch here. The subsequent loop ends this fake touch because real_fingers_down is 0. This causes the tapping code to have a mismatch of how many fingers are down because it never sees the touch begin event for that finger. https://bugs.freedesktop.org/show_bug.cgi?id=105160
2018-02-20 13:27:57 +10:00
litest_event(dev, EV_ABS, ABS_MT_PRESSURE, 3);
litest_event(dev, EV_ABS, ABS_PRESSURE, 3);
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 0);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
litest_push_event_frame(dev);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 1);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 0);
litest_pop_event_frame(dev);
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: don't end below-threshold pressure touches if nfake_fingers > nslots If we have more BTN_TOOL_*TAP fingers down than we have slots, ignore any below-threshold pressure changes on the slots. When a touchpad only detects two touches, guessing whether the third touch has sufficient pressure is unreliable. Instead, always assume that all touches have sufficient pressure when we exceed the slot number. Exception: if all real fingers are below the pressure threshold, the fake fingers are ignored too. Related to https://bugs.freedesktop.org/show_bug.cgi?id=105258 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 11:38:12 +10:00
litest_assert_button_event(li, BTN_MIDDLE, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_MIDDLE, LIBINPUT_BUTTON_STATE_RELEASED);
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
}
END_TEST
START_TEST(touchpad_3fg_tap_hover_btntool)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) >= 3 ||
test: fix a bunch of tests expecting BTN_TOOL_TRIPLETAP A device may have 1 or 2 slots without setting BTN_TOOL_TRIPLETAP, those devices will fail those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-05-08 13:59:27 +10:00
!libevdev_has_event_code(dev->evdev, EV_KEY, BTN_TOOL_TRIPLETAP))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
/* libinput doesn't export when it uses pressure detection, so we
* need to reconstruct this here. Specifically, semi-mt devices are
* non-mt in libinput, so if they have ABS_PRESSURE, they'll use it.
*/
if (libevdev_has_event_code(dev->evdev, EV_ABS, ABS_MT_PRESSURE))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
if (libevdev_has_property(dev->evdev, INPUT_PROP_SEMI_MT) &&
libevdev_has_event_code(dev->evdev, EV_ABS, ABS_PRESSURE))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
litest_enable_tap(dev->libinput_device);
litest_enable_edge_scroll(dev);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 50, 70, 10);
litest_touch_move_to(dev, 1, 70, 50, 50, 70, 10);
test: don't run the MT pressure test on devices without MT pressure This test worked because on devices that don't use pressure the touches were reset when BTN_TOUCH when to 0, triggering the 'ignore fake fingers when no real fingers are down' behavior. But this is a different code path than the pressure handling, so let's separate those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-02-28 14:24:27 +10:00
litest_drain_events(li);
/* drop below the pressure threshold in the same frame as starting a
* third touch */
touchpad: only begin fake touches when we have at least one finger down If a single-touch touchpad drops below the pressure threshold in the same frame where a fake finger is added, we begin a fake touch here. The subsequent loop ends this fake touch because real_fingers_down is 0. This causes the tapping code to have a mismatch of how many fingers are down because it never sees the touch begin event for that finger. https://bugs.freedesktop.org/show_bug.cgi?id=105160
2018-02-20 13:27:57 +10:00
litest_event(dev, EV_KEY, BTN_TOUCH, 0);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 0);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: only begin fake touches when we have at least one finger down If a single-touch touchpad drops below the pressure threshold in the same frame where a fake finger is added, we begin a fake touch here. The subsequent loop ends this fake touch because real_fingers_down is 0. This causes the tapping code to have a mismatch of how many fingers are down because it never sees the touch begin event for that finger. https://bugs.freedesktop.org/show_bug.cgi?id=105160
2018-02-20 13:27:57 +10:00
litest_push_event_frame(dev);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 1);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 0);
litest_pop_event_frame(dev);
litest_assert_empty_queue(li);
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
}
END_TEST
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
START_TEST(touchpad_3fg_tap_btntool)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: fix maybe-uninitialized compiler warnings touchpad-tap.c: In function ‘touchpad_3fg_tap_btntool_inverted’: touchpad-tap.c:1548:2: warning: ‘button’ may be used uninitialized in this function [-Wmaybe-uninitialized] and similar False positive, if button isn't set by now we would've abort()-ed before we even get here. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2016-11-22 10:59:18 +10:00
unsigned int button = 0;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: add optional value names to parametrized tests Only implemented for i32 values so far, used for enums and enum-like constants, replaces a runtime lookup from stringly-typed parameters. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1141>
2025-02-13 22:10:02 +01:00
enum libinput_config_tap_button_map map =
litest_test_param_get_i32(test_env->params, "map");
test: change tap tests to use parameters This also fixes the invocation of the touchpad_3fg_tap_btntool_pointerjump which was written as ranged test but never got invoked with a range (but _i defaults to zero). Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1119>
2025-01-08 09:21:14 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) >= 3 ||
test: fix a bunch of tests expecting BTN_TOOL_TRIPLETAP A device may have 1 or 2 slots without setting BTN_TOOL_TRIPLETAP, those devices will fail those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-05-08 13:59:27 +10:00
!libevdev_has_event_code(dev->evdev, EV_KEY, BTN_TOOL_TRIPLETAP))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_set_tap_map(dev->libinput_device, map);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
switch (map) {
case LIBINPUT_CONFIG_TAP_MAP_LRM:
button = BTN_MIDDLE;
break;
case LIBINPUT_CONFIG_TAP_MAP_LMR:
button = BTN_RIGHT;
break;
default:
litest_abort_msg("Invalid map range %d", map);
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 1);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 0);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: fix some empty queue handling Use the litest_assert_empty_queue() instead of manual checking, and remove the manual checks after the function call. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-11 16:17:33 +10:00
litest_assert_empty_queue(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
}
END_TEST
START_TEST(touchpad_3fg_tap_btntool_inverted)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: fix maybe-uninitialized compiler warnings touchpad-tap.c: In function ‘touchpad_3fg_tap_btntool_inverted’: touchpad-tap.c:1548:2: warning: ‘button’ may be used uninitialized in this function [-Wmaybe-uninitialized] and similar False positive, if button isn't set by now we would've abort()-ed before we even get here. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2016-11-22 10:59:18 +10:00
unsigned int button = 0;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: add optional value names to parametrized tests Only implemented for i32 values so far, used for enums and enum-like constants, replaces a runtime lookup from stringly-typed parameters. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1141>
2025-02-13 22:10:02 +01:00
enum libinput_config_tap_button_map map =
litest_test_param_get_i32(test_env->params, "map");
test: change tap tests to use parameters This also fixes the invocation of the touchpad_3fg_tap_btntool_pointerjump which was written as ranged test but never got invoked with a range (but _i defaults to zero). Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1119>
2025-01-08 09:21:14 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) > 3 ||
test: fix a bunch of tests expecting BTN_TOOL_TRIPLETAP A device may have 1 or 2 slots without setting BTN_TOOL_TRIPLETAP, those devices will fail those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-05-08 13:59:27 +10:00
!libevdev_has_event_code(dev->evdev, EV_KEY, BTN_TOOL_TRIPLETAP))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_set_tap_map(dev->libinput_device, map);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
switch (map) {
case LIBINPUT_CONFIG_TAP_MAP_LRM:
button = BTN_MIDDLE;
break;
case LIBINPUT_CONFIG_TAP_MAP_LMR:
button = BTN_RIGHT;
break;
default:
litest_abort_msg("invalid map range %d", map);
}
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 1);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 0);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: fix some empty queue handling Use the litest_assert_empty_queue() instead of manual checking, and remove the manual checks after the function call. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-11 16:17:33 +10:00
litest_assert_empty_queue(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
}
END_TEST
touchpad: ignore the tap motion threshold if fingers > slots Do so on the synaptics serial touchpads at least, they're known to cause cursor jumps when the third finger is down. Not detecting a tap move means three-finger taps get more reliable on these touchpads. This change affects gestures who now effectively have to wait for the tap timeout to happen. It's a trade-off. https://bugs.freedesktop.org/show_bug.cgi?id=101435 https://bugzilla.redhat.com/show_bug.cgi?id=1455443 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-06-15 14:13:26 +10:00
START_TEST(touchpad_3fg_tap_btntool_pointerjump)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
unsigned int button = 0;
test: add optional value names to parametrized tests Only implemented for i32 values so far, used for enums and enum-like constants, replaces a runtime lookup from stringly-typed parameters. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1141>
2025-02-13 22:10:02 +01:00
enum libinput_config_tap_button_map map =
litest_test_param_get_i32(test_env->params, "map");
test: change tap tests to use parameters This also fixes the invocation of the touchpad_3fg_tap_btntool_pointerjump which was written as ranged test but never got invoked with a range (but _i defaults to zero). Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1119>
2025-01-08 09:21:14 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) > 3 ||
test: fix a bunch of tests expecting BTN_TOOL_TRIPLETAP A device may have 1 or 2 slots without setting BTN_TOOL_TRIPLETAP, those devices will fail those tests. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-05-08 13:59:27 +10:00
!libevdev_has_event_code(dev->evdev, EV_KEY, BTN_TOOL_TRIPLETAP))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: ignore the tap motion threshold if fingers > slots Do so on the synaptics serial touchpads at least, they're known to cause cursor jumps when the third finger is down. Not detecting a tap move means three-finger taps get more reliable on these touchpads. This change affects gestures who now effectively have to wait for the tap timeout to happen. It's a trade-off. https://bugs.freedesktop.org/show_bug.cgi?id=101435 https://bugzilla.redhat.com/show_bug.cgi?id=1455443 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-06-15 14:13:26 +10:00
litest_enable_tap(dev->libinput_device);
litest_set_tap_map(dev->libinput_device, map);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: ignore the tap motion threshold if fingers > slots Do so on the synaptics serial touchpads at least, they're known to cause cursor jumps when the third finger is down. Not detecting a tap move means three-finger taps get more reliable on these touchpads. This change affects gestures who now effectively have to wait for the tap timeout to happen. It's a trade-off. https://bugs.freedesktop.org/show_bug.cgi?id=101435 https://bugzilla.redhat.com/show_bug.cgi?id=1455443 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-06-15 14:13:26 +10:00
switch (map) {
case LIBINPUT_CONFIG_TAP_MAP_LRM:
button = BTN_MIDDLE;
break;
case LIBINPUT_CONFIG_TAP_MAP_LMR:
button = BTN_RIGHT;
break;
default:
litest_abort_msg("Invalid map range %d", map);
}
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 1);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
/* Pointer jump should be ignored */
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 20, 20, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: ignore the tap motion threshold if fingers > slots Do so on the synaptics serial touchpads at least, they're known to cause cursor jumps when the third finger is down. Not detecting a tap move means three-finger taps get more reliable on these touchpads. This change affects gestures who now effectively have to wait for the tap timeout to happen. It's a trade-off. https://bugs.freedesktop.org/show_bug.cgi?id=101435 https://bugzilla.redhat.com/show_bug.cgi?id=1455443 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-06-15 14:13:26 +10:00
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 0);
litest_event(dev, EV_KEY, BTN_TOOL_DOUBLETAP, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 1);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: ignore the tap motion threshold if fingers > slots Do so on the synaptics serial touchpads at least, they're known to cause cursor jumps when the third finger is down. Not detecting a tap move means three-finger taps get more reliable on these touchpads. This change affects gestures who now effectively have to wait for the tap timeout to happen. It's a trade-off. https://bugs.freedesktop.org/show_bug.cgi?id=101435 https://bugzilla.redhat.com/show_bug.cgi?id=1455443 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-06-15 14:13:26 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: ignore the tap motion threshold if fingers > slots Do so on the synaptics serial touchpads at least, they're known to cause cursor jumps when the third finger is down. Not detecting a tap move means three-finger taps get more reliable on these touchpads. This change affects gestures who now effectively have to wait for the tap timeout to happen. It's a trade-off. https://bugs.freedesktop.org/show_bug.cgi?id=101435 https://bugzilla.redhat.com/show_bug.cgi?id=1455443 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-06-15 14:13:26 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
touchpad: improve finger counting for synaptics serial touchpads A three-finger touch may cause slot N to end, in a frame after the BTN_TOOL_TRIPLETAP. This causes tp->nfinger_down to be decremented to 2 as the touch switches to MAYBE_END - which happens to be our num_slots. We exit early and never restore the touch correctly. Fix this by checking that the number of fake touches is equal to the slots, if it is higher then we need to check for recovery. Fixes https://gitlab.freedesktop.org/libinput/libinput/issues/99 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-02 16:58:58 +10:00
START_TEST(touchpad_3fg_tap_slot_release_btntool)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
/* Synaptics touchpads sometimes end one touch point after
* setting BTN_TOOL_TRIPLETAP.
* https://gitlab.freedesktop.org/libinput/libinput/issues/99
*/
litest_drain_events(li);
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: improve finger counting for synaptics serial touchpads A three-finger touch may cause slot N to end, in a frame after the BTN_TOOL_TRIPLETAP. This causes tp->nfinger_down to be decremented to 2 as the touch switches to MAYBE_END - which happens to be our num_slots. We exit early and never restore the touch correctly. Fix this by checking that the number of fake touches is equal to the slots, if it is higher then we need to check for recovery. Fixes https://gitlab.freedesktop.org/libinput/libinput/issues/99 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-02 16:58:58 +10:00
/* touch 1 down */
litest_event(dev, EV_ABS, ABS_MT_SLOT, 0);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, 1);
litest_event(dev, EV_ABS, ABS_MT_POSITION_X, 2200);
litest_event(dev, EV_ABS, ABS_MT_POSITION_Y, 3200);
litest_event(dev, EV_ABS, ABS_MT_PRESSURE, 78);
litest_event(dev, EV_ABS, ABS_X, 2200);
litest_event(dev, EV_ABS, ABS_Y, 3200);
litest_event(dev, EV_ABS, ABS_PRESSURE, 78);
litest_event(dev, EV_KEY, BTN_TOOL_FINGER, 1);
litest_event(dev, EV_KEY, BTN_TOUCH, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: improve finger counting for synaptics serial touchpads A three-finger touch may cause slot N to end, in a frame after the BTN_TOOL_TRIPLETAP. This causes tp->nfinger_down to be decremented to 2 as the touch switches to MAYBE_END - which happens to be our num_slots. We exit early and never restore the touch correctly. Fix this by checking that the number of fake touches is equal to the slots, if it is higher then we need to check for recovery. Fixes https://gitlab.freedesktop.org/libinput/libinput/issues/99 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-02 16:58:58 +10:00
msleep(2);
/* touch 2 and TRIPLETAP down */
litest_event(dev, EV_ABS, ABS_MT_SLOT, 1);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, 1);
litest_event(dev, EV_ABS, ABS_MT_POSITION_X, 2500);
litest_event(dev, EV_ABS, ABS_MT_POSITION_Y, 3800);
litest_event(dev, EV_ABS, ABS_MT_PRESSURE, 73);
litest_event(dev, EV_KEY, BTN_TOOL_FINGER, 0);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: improve finger counting for synaptics serial touchpads A three-finger touch may cause slot N to end, in a frame after the BTN_TOOL_TRIPLETAP. This causes tp->nfinger_down to be decremented to 2 as the touch switches to MAYBE_END - which happens to be our num_slots. We exit early and never restore the touch correctly. Fix this by checking that the number of fake touches is equal to the slots, if it is higher then we need to check for recovery. Fixes https://gitlab.freedesktop.org/libinput/libinput/issues/99 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-02 16:58:58 +10:00
msleep(2);
/* touch 2 up, coordinate jump + ends slot 1, TRIPLETAP stays */
test: abort when we detect a touch jump during the tests We never want to accidentally trigger this one. Where we trigger them on purpose, we can swap the log handler out first. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 10:00:12 +10:00
litest_disable_log_handler(li);
touchpad: improve finger counting for synaptics serial touchpads A three-finger touch may cause slot N to end, in a frame after the BTN_TOOL_TRIPLETAP. This causes tp->nfinger_down to be decremented to 2 as the touch switches to MAYBE_END - which happens to be our num_slots. We exit early and never restore the touch correctly. Fix this by checking that the number of fake touches is equal to the slots, if it is higher then we need to check for recovery. Fixes https://gitlab.freedesktop.org/libinput/libinput/issues/99 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-02 16:58:58 +10:00
litest_event(dev, EV_ABS, ABS_MT_SLOT, 0);
litest_event(dev, EV_ABS, ABS_MT_POSITION_X, 2500);
litest_event(dev, EV_ABS, ABS_MT_POSITION_Y, 3800);
litest_event(dev, EV_ABS, ABS_MT_PRESSURE, 78);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 1);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_X, 2500);
litest_event(dev, EV_ABS, ABS_Y, 3800);
litest_event(dev, EV_ABS, ABS_PRESSURE, 78);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: improve finger counting for synaptics serial touchpads A three-finger touch may cause slot N to end, in a frame after the BTN_TOOL_TRIPLETAP. This causes tp->nfinger_down to be decremented to 2 as the touch switches to MAYBE_END - which happens to be our num_slots. We exit early and never restore the touch correctly. Fix this by checking that the number of fake touches is equal to the slots, if it is higher then we need to check for recovery. Fixes https://gitlab.freedesktop.org/libinput/libinput/issues/99 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-02 16:58:58 +10:00
msleep(2);
/* slot 2 reactivated
*/
litest_event(dev, EV_ABS, ABS_MT_SLOT, 0);
litest_event(dev, EV_ABS, ABS_MT_POSITION_X, 2500);
litest_event(dev, EV_ABS, ABS_MT_POSITION_Y, 3800);
litest_event(dev, EV_ABS, ABS_MT_PRESSURE, 78);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 1);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, 3);
litest_event(dev, EV_ABS, ABS_MT_POSITION_X, 3500);
litest_event(dev, EV_ABS, ABS_MT_POSITION_Y, 3500);
litest_event(dev, EV_ABS, ABS_MT_PRESSURE, 73);
litest_event(dev, EV_ABS, ABS_X, 2200);
litest_event(dev, EV_ABS, ABS_Y, 3200);
litest_event(dev, EV_ABS, ABS_PRESSURE, 78);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: abort when we detect a touch jump during the tests We never want to accidentally trigger this one. Where we trigger them on purpose, we can swap the log handler out first. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 10:00:12 +10:00
litest_restore_log_handler(li);
touchpad: improve finger counting for synaptics serial touchpads A three-finger touch may cause slot N to end, in a frame after the BTN_TOOL_TRIPLETAP. This causes tp->nfinger_down to be decremented to 2 as the touch switches to MAYBE_END - which happens to be our num_slots. We exit early and never restore the touch correctly. Fix this by checking that the number of fake touches is equal to the slots, if it is higher then we need to check for recovery. Fixes https://gitlab.freedesktop.org/libinput/libinput/issues/99 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-02 16:58:58 +10:00
/* now end all three */
litest_event(dev, EV_ABS, ABS_MT_SLOT, 0);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 1);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_KEY, BTN_TOUCH, 0);
litest_event(dev, EV_KEY, BTN_TOOL_TRIPLETAP, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: improve finger counting for synaptics serial touchpads A three-finger touch may cause slot N to end, in a frame after the BTN_TOOL_TRIPLETAP. This causes tp->nfinger_down to be decremented to 2 as the touch switches to MAYBE_END - which happens to be our num_slots. We exit early and never restore the touch correctly. Fix this by checking that the number of fake touches is equal to the slots, if it is higher then we need to check for recovery. Fixes https://gitlab.freedesktop.org/libinput/libinput/issues/99 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-02 16:58:58 +10:00
litest_assert_button_event(li, BTN_MIDDLE, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_MIDDLE, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
START_TEST(touchpad_3fg_tap_after_scroll)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) <= 3)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
litest_enable_2fg_scroll(dev);
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
litest_touch_down(dev, 0, 40, 20);
litest_touch_down(dev, 1, 50, 20);
litest_drain_events(li);
/* 2fg scroll */
litest_touch_move_two_touches(dev, 40, 20, 50, 20, 0, 20, 10);
litest_drain_events(li);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
/* third finger tap without the other two fingers moving */
litest_touch_down(dev, 2, 60, 40);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
litest_touch_up(dev, 2);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
litest_assert_empty_queue(li);
}
END_TEST
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
START_TEST(touchpad_4fg_tap)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
int i;
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) <= 4)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
for (i = 0; i < 4; i++) {
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
litest_touch_down(dev, 2, 80, 50);
litest_touch_down(dev, 3, 90, 50);
litest_touch_up(dev, (i + 3) % 4);
litest_touch_up(dev, (i + 2) % 4);
litest_touch_up(dev, (i + 1) % 4);
litest_touch_up(dev, (i + 0) % 4);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
}
END_TEST
START_TEST(touchpad_4fg_tap_quickrelease)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) <= 4)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 50);
litest_touch_down(dev, 2, 80, 50);
litest_touch_down(dev, 3, 90, 50);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 0);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 1);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 2);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 3);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_KEY, BTN_TOOL_QUADTAP, 0);
litest_event(dev, EV_KEY, BTN_TOUCH, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
START_TEST(touchpad_move_after_touch)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
litest_drain_events(li);
/* respective number of fingers down */
switch (nfingers) {
case 5:
test: fix the touchpad finger positions for a test The 5th finger was placed in the same position as the 4th finger which doesn't have an effect in libinput but it looks wrong. And the first finger was put down at 40/30 but then moved from 70/30 to the new position, causing pointer jumps on some touchpads. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1145>
2025-02-18 15:09:45 +10:00
litest_touch_down(dev, 4, 80, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
case 4:
litest_touch_down(dev, 3, 70, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
}
/* move finger 1 */
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: fix the touchpad finger positions for a test The 5th finger was placed in the same position as the 4th finger which doesn't have an effect in libinput but it looks wrong. And the first finger was put down at 40/30 but then moved from 70/30 to the new position, causing pointer jumps on some touchpads. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1145>
2025-02-18 15:09:45 +10:00
litest_touch_move_to(dev, 0, 40, 30, 70, 60, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
/* lift finger 1, put it back */
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
litest_touch_down(dev, 0, 40, 30);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
/* lift fingers up */
switch (nfingers) {
case 5:
litest_touch_up(dev, 4);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
case 4:
litest_touch_up(dev, 3);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: don't allow for multifinger tapping after a move In the current implementation, movements > threshold and timeouts usually move to HOLD state and continue from there. Where a finger is lifted, we go back up the diagram into the previous finger count's HOLD state. The side-effect of this is that a tap of a finger can be counted as tap even after a movement: - two fingers down, move to scroll, hold down - third finger down, third finger up This sequence triggers an erroneous three-finger tap. Once the motion threshold is hit by any touch, no finger must trigger 2/3 finger tap events while any touch is down. The false tap is only triggered where the new finger can execute a tap without any other finger changing any property. This can be triggered on the reporter's Dell Precision 5520 but on most other touchpads, a new finger down will trigger slight movement, pressure or touch size updates and thus the bug cannot be triggered. Fixes #382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2019-11-08 15:16:42 +10:00
litest_assert_no_typed_events(li, LIBINPUT_EVENT_POINTER_BUTTON);
}
END_TEST
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
START_TEST(touchpad_5fg_tap)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
int i;
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) < 5)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
for (i = 0; i < 5; i++) {
litest_drain_events(li);
litest_touch_down(dev, 0, 20, 50);
litest_touch_down(dev, 1, 30, 50);
litest_touch_down(dev, 2, 40, 50);
litest_touch_down(dev, 3, 50, 50);
litest_touch_down(dev, 4, 60, 50);
litest_touch_up(dev, (i + 4) % 5);
litest_touch_up(dev, (i + 3) % 5);
litest_touch_up(dev, (i + 2) % 5);
litest_touch_up(dev, (i + 1) % 5);
litest_touch_up(dev, (i + 0) % 5);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
}
END_TEST
START_TEST(touchpad_5fg_tap_quickrelease)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) < 5)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(li);
litest_touch_down(dev, 0, 20, 50);
litest_touch_down(dev, 1, 30, 50);
litest_touch_down(dev, 2, 40, 50);
litest_touch_down(dev, 3, 70, 50);
litest_touch_down(dev, 4, 90, 50);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 0);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 1);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 2);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 3);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_ABS, ABS_MT_SLOT, 4);
litest_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
litest_event(dev, EV_KEY, BTN_TOOL_QUINTTAP, 0);
litest_event(dev, EV_KEY, BTN_TOUCH, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(clickpad_1fg_tap_click)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: replace tap config with helper function No functional change, other than that we check for status codes now too. In tests that don't specifically check the interface itself, a short enable_tap() or disable_tap() is a lot more obvious to parse for the reader. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-17 10:02:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_drain_events(dev->libinput);
/* finger down, button click, finger up
-> only one button left event pair */
litest_touch_down(dev, 0, 50, 50);
litest_event(dev, EV_KEY, BTN_LEFT, 1);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_event(dev, EV_KEY, BTN_LEFT, 0);
litest_event(dev, EV_SYN, SYN_REPORT, 0);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_is_available)
{
struct litest_device *dev = litest_current_device();
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_ge(
libinput_device_config_tap_get_finger_count(dev->libinput_device),
1);
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
}
END_TEST
START_TEST(touchpad_tap_is_not_available)
{
struct litest_device *dev = litest_current_device();
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_eq(
libinput_device_config_tap_get_finger_count(dev->libinput_device),
0);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_enabled(dev->libinput_device),
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_CONFIG_TAP_DISABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_set_enabled(dev->libinput_device,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_CONFIG_TAP_ENABLED),
LIBINPUT_CONFIG_STATUS_UNSUPPORTED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_set_enabled(dev->libinput_device,
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_CONFIG_TAP_DISABLED),
LIBINPUT_CONFIG_STATUS_SUCCESS);
}
END_TEST
START_TEST(touchpad_tap_default_disabled)
{
struct litest_device *dev = litest_current_device();
/* this test is only run on specific devices */
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_enabled(dev->libinput_device),
test: add missing checks for tap enabled/disabled by default Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-02-06 10:51:25 +10:00
LIBINPUT_CONFIG_TAP_DISABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_default_enabled(dev->libinput_device),
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_CONFIG_TAP_DISABLED);
}
END_TEST
START_TEST(touchpad_tap_default_enabled)
{
struct litest_device *dev = litest_current_device();
/* this test is only run on specific devices */
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_enabled(dev->libinput_device),
test: add missing checks for tap enabled/disabled by default Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-02-06 10:51:25 +10:00
LIBINPUT_CONFIG_TAP_ENABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_default_enabled(dev->libinput_device),
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_CONFIG_TAP_ENABLED);
}
END_TEST
START_TEST(touchpad_tap_invalid)
{
struct litest_device *dev = litest_current_device();
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_set_enabled(dev->libinput_device, 2),
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_CONFIG_STATUS_INVALID);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_set_enabled(dev->libinput_device, -1),
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
LIBINPUT_CONFIG_STATUS_INVALID);
}
END_TEST
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
START_TEST(touchpad_tap_default_map)
{
struct litest_device *dev = litest_current_device();
enum libinput_config_tap_button_map map;
map = libinput_device_config_tap_get_button_map(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(map, LIBINPUT_CONFIG_TAP_MAP_LRM);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
map = libinput_device_config_tap_get_default_button_map(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(map, LIBINPUT_CONFIG_TAP_MAP_LRM);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
}
END_TEST
test: add test to set tap map on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 15:00:00 +10:00
START_TEST(touchpad_tap_map_unsupported)
{
struct litest_device *dev = litest_current_device();
enum libinput_config_tap_button_map map;
enum libinput_config_status status;
map = libinput_device_config_tap_get_button_map(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(map, LIBINPUT_CONFIG_TAP_MAP_LRM);
test: add test to set tap map on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 15:00:00 +10:00
map = libinput_device_config_tap_get_default_button_map(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(map, LIBINPUT_CONFIG_TAP_MAP_LRM);
test: add test to set tap map on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 15:00:00 +10:00
status = libinput_device_config_tap_set_button_map(dev->libinput_device,
LIBINPUT_CONFIG_TAP_MAP_LMR);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_UNSUPPORTED);
test: add test to set tap map on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 15:00:00 +10:00
status = libinput_device_config_tap_set_button_map(dev->libinput_device,
LIBINPUT_CONFIG_TAP_MAP_LRM);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_UNSUPPORTED);
test: add test to set tap map on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 15:00:00 +10:00
}
END_TEST
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
START_TEST(touchpad_tap_set_map)
{
struct litest_device *dev = litest_current_device();
struct libinput_device *device = dev->libinput_device;
enum libinput_config_tap_button_map map;
enum libinput_config_status status;
map = LIBINPUT_CONFIG_TAP_MAP_LRM;
status = libinput_device_config_tap_set_button_map(device, map);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_SUCCESS);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
map = libinput_device_config_tap_get_button_map(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(map, LIBINPUT_CONFIG_TAP_MAP_LRM);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
map = LIBINPUT_CONFIG_TAP_MAP_LMR;
status = libinput_device_config_tap_set_button_map(device, map);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_SUCCESS);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
map = libinput_device_config_tap_get_button_map(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(map, LIBINPUT_CONFIG_TAP_MAP_LMR);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
map = LIBINPUT_CONFIG_TAP_MAP_LRM - 1;
status = libinput_device_config_tap_set_button_map(device, map);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_INVALID);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
map = LIBINPUT_CONFIG_TAP_MAP_LMR + 1;
status = libinput_device_config_tap_set_button_map(device, map);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_INVALID);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
}
END_TEST
START_TEST(touchpad_tap_set_map_no_tapping)
{
struct litest_device *dev = litest_current_device();
struct libinput_device *device = dev->libinput_device;
enum libinput_config_tap_button_map map;
enum libinput_config_status status;
map = LIBINPUT_CONFIG_TAP_MAP_LRM;
status = libinput_device_config_tap_set_button_map(device, map);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_UNSUPPORTED);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
map = LIBINPUT_CONFIG_TAP_MAP_LMR;
status = libinput_device_config_tap_set_button_map(device, map);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_UNSUPPORTED);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
map = LIBINPUT_CONFIG_TAP_MAP_LRM - 1;
status = libinput_device_config_tap_set_button_map(device, map);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_INVALID);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
map = LIBINPUT_CONFIG_TAP_MAP_LMR + 1;
status = libinput_device_config_tap_set_button_map(device, map);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_INVALID);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
}
END_TEST
test: add test for default button tap map (on non-tapping devices) Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-02-06 10:45:21 +10:00
START_TEST(touchpad_tap_get_map_no_tapping)
{
struct litest_device *dev = litest_current_device();
struct libinput_device *device = dev->libinput_device;
enum libinput_config_tap_button_map map;
map = libinput_device_config_tap_get_button_map(device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(map, LIBINPUT_CONFIG_TAP_MAP_LRM);
test: add test for default button tap map (on non-tapping devices) Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-02-06 10:45:21 +10:00
map = libinput_device_config_tap_get_default_button_map(device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(map, LIBINPUT_CONFIG_TAP_MAP_LRM);
test: add test for default button tap map (on non-tapping devices) Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-02-06 10:45:21 +10:00
}
END_TEST
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
START_TEST(touchpad_tap_map_delayed)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
enum libinput_config_tap_button_map map;
litest_enable_tap(dev->libinput_device);
litest_set_tap_map(dev->libinput_device, LIBINPUT_CONFIG_TAP_MAP_LRM);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_drain_events(dev->libinput);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 70, 70);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_set_tap_map(dev->libinput_device, LIBINPUT_CONFIG_TAP_MAP_LMR);
map = libinput_device_config_tap_get_button_map(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(map, LIBINPUT_CONFIG_TAP_MAP_LMR);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: implement tests for configurable tap button mappings Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-07-21 14:50:40 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
START_TEST(touchpad_drag_default_disabled)
{
struct litest_device *dev = litest_current_device();
/* this test is only run on specific devices */
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(libinput_device_config_tap_get_default_drag_enabled(
dev->libinput_device),
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
LIBINPUT_CONFIG_DRAG_DISABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_drag_enabled(dev->libinput_device),
test: add more tests for tap-drag config Including one to get/set it on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 14:34:57 +10:00
LIBINPUT_CONFIG_DRAG_DISABLED);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
}
END_TEST
START_TEST(touchpad_drag_default_enabled)
{
struct litest_device *dev = litest_current_device();
/* this test is only run on specific devices */
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(libinput_device_config_tap_get_default_drag_enabled(
dev->libinput_device),
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
LIBINPUT_CONFIG_DRAG_ENABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_drag_enabled(dev->libinput_device),
test: add more tests for tap-drag config Including one to get/set it on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 14:34:57 +10:00
LIBINPUT_CONFIG_DRAG_ENABLED);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
}
END_TEST
START_TEST(touchpad_drag_config_invalid)
{
struct litest_device *dev = litest_current_device();
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_set_drag_enabled(dev->libinput_device, 2),
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
LIBINPUT_CONFIG_STATUS_INVALID);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_set_drag_enabled(dev->libinput_device, -1),
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
LIBINPUT_CONFIG_STATUS_INVALID);
}
END_TEST
test: add more tests for tap-drag config Including one to get/set it on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 14:34:57 +10:00
START_TEST(touchpad_drag_config_unsupported)
{
struct litest_device *dev = litest_current_device();
enum libinput_config_status status;
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(libinput_device_config_tap_get_default_drag_enabled(
dev->libinput_device),
test: add more tests for tap-drag config Including one to get/set it on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 14:34:57 +10:00
LIBINPUT_CONFIG_DRAG_DISABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_drag_enabled(dev->libinput_device),
test: add more tests for tap-drag config Including one to get/set it on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 14:34:57 +10:00
LIBINPUT_CONFIG_DRAG_DISABLED);
status = libinput_device_config_tap_set_drag_enabled(
dev->libinput_device,
LIBINPUT_CONFIG_DRAG_ENABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_UNSUPPORTED);
test: add more tests for tap-drag config Including one to get/set it on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 14:34:57 +10:00
status = libinput_device_config_tap_set_drag_enabled(
dev->libinput_device,
LIBINPUT_CONFIG_DRAG_DISABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_SUCCESS);
test: add more tests for tap-drag config Including one to get/set it on unsupported devices Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-01-16 14:34:57 +10:00
}
END_TEST
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
START_TEST(touchpad_drag_config_enabledisable)
{
struct litest_device *dev = litest_current_device();
enum libinput_config_drag_state state;
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_enable_tap(dev->libinput_device);
litest_disable_tap_drag(dev->libinput_device);
state = libinput_device_config_tap_get_drag_enabled(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(state, LIBINPUT_CONFIG_DRAG_DISABLED);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_enable_tap_drag(dev->libinput_device);
state = libinput_device_config_tap_get_drag_enabled(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(state, LIBINPUT_CONFIG_DRAG_ENABLED);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
/* same thing with tapping disabled */
litest_enable_tap(dev->libinput_device);
litest_disable_tap_drag(dev->libinput_device);
state = libinput_device_config_tap_get_drag_enabled(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(state, LIBINPUT_CONFIG_DRAG_DISABLED);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_enable_tap_drag(dev->libinput_device);
state = libinput_device_config_tap_get_drag_enabled(dev->libinput_device);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(state, LIBINPUT_CONFIG_DRAG_ENABLED);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
}
END_TEST
START_TEST(touchpad_drag_disabled)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_enable_tap(dev->libinput_device);
litest_disable_tap_drag(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
/* lift fingers up */
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 90, 90, 10);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
}
END_TEST
START_TEST(touchpad_drag_disabled_immediate)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *ev;
struct libinput_event_pointer *ptrev;
uint64_t press_time, release_time;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_enable_tap(dev->libinput_device);
litest_disable_tap_drag(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
msleep(10); /* to force a time difference */
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev, button, LIBINPUT_BUTTON_STATE_PRESSED);
press_time = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(ev);
ev = libinput_get_event(li);
ptrev = litest_is_button_event(ev, button, LIBINPUT_BUTTON_STATE_RELEASED);
release_time = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(ev);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(release_time, press_time);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
}
END_TEST
START_TEST(touchpad_drag_disabled_multitap_no_drag)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct libinput_event *event;
struct libinput_event_pointer *ptrev;
uint32_t oldtime = 0, curtime;
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_enable_tap(dev->libinput_device);
litest_disable_tap_drag(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
msleep(10);
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 70, 50, 10);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
for (ntaps = 0; ntaps <= range; ntaps++) {
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
LIBINPUT_BUTTON_STATE_PRESSED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_gt(curtime, oldtime);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
event = libinput_get_event(li);
ptrev = litest_is_button_event(event,
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
button,
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
LIBINPUT_BUTTON_STATE_RELEASED);
curtime = libinput_event_pointer_get_time(ptrev);
libinput_event_destroy(event);
test: switch ck_assert over to litest_assert This is the first step in switching away from the check framework. Our litest macros already do almost exactly the same anyway so most of this is a simple sed with a few compiler fixes where things mismatch (nonnull -> notnull) and (_tol -> _epsilon). This now generates a whole bunch of integer mismatch warnings: check casts everything to intmax_t whereas we use typeof, so lots of warnings especially for enums. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:20:26 +10:00
litest_assert_int_ge(curtime, oldtime);
touchpad: add a config option to disable tap-and-drag There are a number of use-cases where tapping may be desirable, but tap-and-drag is not, e.g. where tapping is used to select multiple items in a list. Having tap-and-drag on hinders this, and the nature of the interaction means it cannot be detected based on timeouts, movement thresholds, etc. Provide an option instead to turn tap-an-drag off. Tap-and-drag remains enabled by default (though tapping is disabled by default). For the touchpad tap state diagram, the new option disables the transition from state TOUCH to state TAPPED and releases the button immediately instead. This means that multitap-and-drag is disabled too since we now just loop around in the single-tap state for multitap. It also makes tapping more responsive - we don't have to wait for the timeout before we know whether it's a tap event. The first touch time is noted, we now send the button press with the time of the first touch and the release with the time of the release. This ensures a realistic time diff between the two events. https://bugs.freedesktop.org/show_bug.cgi?id=93502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.netto> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2016-01-22 17:59:19 +10:00
oldtime = curtime;
}
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_assert_empty_queue(li);
}
END_TEST
touchpad: disable tap drag lock by default Similar to tapping, it's a feature that is useful but confusing if a user doesn't know it exists. It makes the touchpad appear laggy and slow to react in the best case, or appear like a stuck button in the worst case. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-26 16:18:29 +10:00
START_TEST(touchpad_drag_lock_default_disabled)
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput_device *device = dev->libinput_device;
enum libinput_config_status status;
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(libinput_device_config_tap_get_drag_lock_enabled(device),
touchpad: disable tap drag lock by default Similar to tapping, it's a feature that is useful but confusing if a user doesn't know it exists. It makes the touchpad appear laggy and slow to react in the best case, or appear like a stuck button in the worst case. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-26 16:18:29 +10:00
LIBINPUT_CONFIG_DRAG_LOCK_DISABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_default_drag_lock_enabled(device),
touchpad: disable tap drag lock by default Similar to tapping, it's a feature that is useful but confusing if a user doesn't know it exists. It makes the touchpad appear laggy and slow to react in the best case, or appear like a stuck button in the worst case. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-26 16:18:29 +10:00
LIBINPUT_CONFIG_DRAG_LOCK_DISABLED);
Run clang-format over the code This uses the .clang-format file in the follow-up commit, but committed prior to that to ease review of said file and various integrations. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1246>
2025-07-01 16:30:11 +10:00
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
status = libinput_device_config_tap_set_drag_lock_enabled(
device,
LIBINPUT_CONFIG_DRAG_LOCK_ENABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_SUCCESS);
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
/* ENABLED is a legacy spelling for ENABLED_TIMEOUT */
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(libinput_device_config_tap_get_drag_lock_enabled(device),
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
LIBINPUT_CONFIG_DRAG_LOCK_ENABLED_TIMEOUT);
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
status = libinput_device_config_tap_set_drag_lock_enabled(
device,
LIBINPUT_CONFIG_DRAG_LOCK_DISABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_SUCCESS);
litest_assert_enum_eq(libinput_device_config_tap_get_drag_lock_enabled(device),
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
LIBINPUT_CONFIG_DRAG_LOCK_DISABLED);
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
status = libinput_device_config_tap_set_drag_lock_enabled(
device,
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
LIBINPUT_CONFIG_DRAG_LOCK_ENABLED_TIMEOUT);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_SUCCESS);
litest_assert_enum_eq(libinput_device_config_tap_get_drag_lock_enabled(device),
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
LIBINPUT_CONFIG_DRAG_LOCK_ENABLED_TIMEOUT);
status = libinput_device_config_tap_set_drag_lock_enabled(
device,
LIBINPUT_CONFIG_DRAG_LOCK_ENABLED_STICKY);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_SUCCESS);
litest_assert_enum_eq(libinput_device_config_tap_get_drag_lock_enabled(device),
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
LIBINPUT_CONFIG_DRAG_LOCK_ENABLED_STICKY);
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
status = libinput_device_config_tap_set_drag_lock_enabled(device, 3);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_INVALID);
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
}
END_TEST
touchpad: disable tap drag lock by default Similar to tapping, it's a feature that is useful but confusing if a user doesn't know it exists. It makes the touchpad appear laggy and slow to react in the best case, or appear like a stuck button in the worst case. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-26 16:18:29 +10:00
START_TEST(touchpad_drag_lock_default_unavailable)
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput_device *device = dev->libinput_device;
enum libinput_config_status status;
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(libinput_device_config_tap_get_drag_lock_enabled(device),
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
LIBINPUT_CONFIG_DRAG_LOCK_DISABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(
libinput_device_config_tap_get_default_drag_lock_enabled(device),
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
LIBINPUT_CONFIG_DRAG_LOCK_DISABLED);
Run clang-format over the code This uses the .clang-format file in the follow-up commit, but committed prior to that to ease review of said file and various integrations. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1246>
2025-07-01 16:30:11 +10:00
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
status = libinput_device_config_tap_set_drag_lock_enabled(
device,
LIBINPUT_CONFIG_DRAG_LOCK_ENABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_UNSUPPORTED);
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
status = libinput_device_config_tap_set_drag_lock_enabled(
device,
LIBINPUT_CONFIG_DRAG_LOCK_ENABLED_TIMEOUT);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_UNSUPPORTED);
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
status = libinput_device_config_tap_set_drag_lock_enabled(
device,
LIBINPUT_CONFIG_DRAG_LOCK_ENABLED_STICKY);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_UNSUPPORTED);
touchpad: add sticky mode to drag-lock Sticky mode removes the timeout from drag-lock, only a tap ends a drag. Timeout mode remains available without changes. Sticky mode is exposed as a new value for the existing drag-lock setting. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1037>
2024-09-02 16:29:02 +02:00
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
status = libinput_device_config_tap_set_drag_lock_enabled(
device,
LIBINPUT_CONFIG_DRAG_LOCK_DISABLED);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_SUCCESS);
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
status = libinput_device_config_tap_set_drag_lock_enabled(device, 3);
test: switch the tests to use enum comparisons Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
2024-09-16 16:49:38 +10:00
litest_assert_enum_eq(status, LIBINPUT_CONFIG_STATUS_INVALID);
touchpad: hook up drag lock configuration Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-22 11:06:25 +10:00
}
END_TEST
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
static inline bool
touchpad_has_palm_pressure(struct litest_device *dev)
{
struct libevdev *evdev = dev->evdev;
quirks: disable pressure on the Lenovo Yoga 9i touchpad This touchpad is a true pressurepad and the pressure axis gives us physical pressure down. Using it as contact size gives flaky touch detection, so let's just disable the axis until we do something with that value. Fixes #562 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-01-14 12:55:30 +10:00
if (dev->which == LITEST_SYNAPTICS_PRESSUREPAD)
return false;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (libevdev_has_event_code(evdev, EV_ABS, ABS_MT_PRESSURE))
touchpad: disable the pressure axes wherever the resolution is nonzero The kernel/udev set the pressure resolution to nonzero to indicate the value is in a known scale (units/g). We use that information to disable the pressure axis on such devices - real pressure cannot be translated to contact size. For the kernel patch see: https://www.spinics.net/lists/linux-input/msg71237.html Fixes #569 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-05 08:35:10 +10:00
return libevdev_get_abs_resolution(evdev, ABS_MT_PRESSURE) == 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
return false;
}
START_TEST(touchpad_tap_palm_on_idle)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* Finger down is immediately palm */
litest_touch_down_extended(dev, 0, 50, 50, axes);
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_touch)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* Finger down is palm after touch begin */
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_touch_hold_timeout)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* Finger down is palm after tap timeout */
litest_touch_down(dev, 0, 50, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_touch_hold_move)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* Finger down is palm after tap move threshold */
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, 0, 50, 50, 60, 60, 10);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 60, 60, 60, 60, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_tapped)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* tap + palm down */
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
touchpad: handle a palm down in the tapped state We expected the first event after TAPPED to be a finger down. If that finger has been recognised as palm, the finger state isn't TOUCH_BEGIN so we get an invalid state in our FSM. libinput bug: 0: invalid tap event TAP_EVENT_PALM in state TAP_STATE_TAPPED https://bugs.freedesktop.org/show_bug.cgi?id=105370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-07 10:58:17 +10:00
START_TEST(touchpad_tap_palm_on_tapped_palm_down)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
touchpad: handle a palm down in the tapped state We expected the first event after TAPPED to be a finger down. If that finger has been recognised as palm, the finger state isn't TOUCH_BEGIN so we get an invalid state in our FSM. libinput bug: 0: invalid tap event TAP_EVENT_PALM in state TAP_STATE_TAPPED https://bugs.freedesktop.org/show_bug.cgi?id=105370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-07 10:58:17 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: handle a palm down in the tapped state We expected the first event after TAPPED to be a finger down. If that finger has been recognised as palm, the finger state isn't TOUCH_BEGIN so we get an invalid state in our FSM. libinput bug: 0: invalid tap event TAP_EVENT_PALM in state TAP_STATE_TAPPED https://bugs.freedesktop.org/show_bug.cgi?id=105370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-07 10:58:17 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: handle a palm down in the tapped state We expected the first event after TAPPED to be a finger down. If that finger has been recognised as palm, the finger state isn't TOUCH_BEGIN so we get an invalid state in our FSM. libinput bug: 0: invalid tap event TAP_EVENT_PALM in state TAP_STATE_TAPPED https://bugs.freedesktop.org/show_bug.cgi?id=105370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-07 10:58:17 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: handle a palm down in the tapped state We expected the first event after TAPPED to be a finger down. If that finger has been recognised as palm, the finger state isn't TOUCH_BEGIN so we get an invalid state in our FSM. libinput bug: 0: invalid tap event TAP_EVENT_PALM in state TAP_STATE_TAPPED https://bugs.freedesktop.org/show_bug.cgi?id=105370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-07 10:58:17 +10:00
litest_drain_events(li);
/* tap + palm down */
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: handle a palm down in the tapped state We expected the first event after TAPPED to be a finger down. If that finger has been recognised as palm, the finger state isn't TOUCH_BEGIN so we get an invalid state in our FSM. libinput bug: 0: invalid tap event TAP_EVENT_PALM in state TAP_STATE_TAPPED https://bugs.freedesktop.org/show_bug.cgi?id=105370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-07 10:58:17 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_touch_down_extended(dev, 0, 50, 50, axes);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
touchpad: handle a palm down in the tapped state We expected the first event after TAPPED to be a finger down. If that finger has been recognised as palm, the finger state isn't TOUCH_BEGIN so we get an invalid state in our FSM. libinput bug: 0: invalid tap event TAP_EVENT_PALM in state TAP_STATE_TAPPED https://bugs.freedesktop.org/show_bug.cgi?id=105370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-07 10:58:17 +10:00
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: handle a palm down in the tapped state We expected the first event after TAPPED to be a finger down. If that finger has been recognised as palm, the finger state isn't TOUCH_BEGIN so we get an invalid state in our FSM. libinput bug: 0: invalid tap event TAP_EVENT_PALM in state TAP_STATE_TAPPED https://bugs.freedesktop.org/show_bug.cgi?id=105370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-07 10:58:17 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
START_TEST(touchpad_tap_palm_on_tapped_doubletap)
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers_1st"),
nfingers2 = litest_test_param_get_i32(test_env->params, "fingers_2nd");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0, button2 = 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers2 + 1 > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
switch (nfingers2) {
case 1:
button2 = BTN_LEFT;
break;
case 2:
button2 = BTN_RIGHT;
break;
case 3:
button2 = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers2);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
/* tap + palm down + tap with additional finger(s) */
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers2) {
case 3:
litest_touch_down(dev, 3, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 2, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 1, 40, 30);
break;
}
switch (nfingers2) {
case 3:
litest_touch_up(dev, 3);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 1);
break;
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_button_event(li, button2, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button2, LIBINPUT_BUTTON_STATE_RELEASED);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_drag)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* tap + finger down (->drag), finger turns into palm */
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_touch_down(dev, 0, 50, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_drag_2fg)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int which = litest_test_param_get_i32(test_env->params, "which"),
this = which % 2, other = (which + 1) % 2,
nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* tap + finger down, 2nd finger down, finger turns to palm */
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_touch_down(dev, this, 50, 50);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_down(dev, other, 60, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, this, 50, 50, 50, 50, axes, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to(dev, other, 60, 50, 65, 50, 10);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_only_typed_events(li, LIBINPUT_EVENT_POINTER_MOTION);
litest_touch_up(dev, other);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_touch_up(dev, this);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_touch_2)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int which = litest_test_param_get_i32(test_env->params, "which"),
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
this = which % 2, other = (which + 1) % 2;
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* 2fg tap with one finger detected as palm */
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 60, 60);
litest_drain_events(li);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, this, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, this);
litest_touch_up(dev, other);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_touch_2_retouch)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int which = litest_test_param_get_i32(test_env->params, "which"),
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
this = which % 2, other = (which + 1) % 2;
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* 2fg tap with one finger detected as palm, that finger is lifted
* and taps again as not-palm */
litest_touch_down(dev, this, 50, 50);
litest_touch_down(dev, other, 60, 60);
litest_drain_events(li);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, this, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, this);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_down(dev, this, 70, 70);
litest_touch_up(dev, this);
litest_touch_up(dev, other);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_touch_3)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int which = litest_test_param_get_i32(test_env->params, "which"),
this = which % 3;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) < 3)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* 3fg tap with one finger detected as palm, that finger is lifted,
other two fingers lifted cause 2fg tap */
litest_touch_down(dev, this, 50, 50);
litest_touch_down(dev, (this + 1) % 3, 60, 50);
litest_touch_down(dev, (this + 2) % 3, 70, 50);
litest_drain_events(li);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, this, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, this);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, (this + 1) % 3);
litest_touch_up(dev, (this + 2) % 3);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_touch_3_retouch)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int which = litest_test_param_get_i32(test_env->params, "which"),
this = which % 3;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) < 3)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* 3fg tap with one finger detected as palm, that finger is lifted,
then put down again as normal finger -> 3fg tap */
litest_touch_down(dev, this, 50, 50);
litest_touch_down(dev, (this + 1) % 3, 60, 50);
litest_touch_down(dev, (this + 2) % 3, 70, 50);
litest_drain_events(li);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, this, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, this);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_down(dev, this, 50, 50);
litest_touch_up(dev, this);
litest_touch_up(dev, (this + 1) % 3);
litest_touch_up(dev, (this + 2) % 3);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_MIDDLE, LIBINPUT_BUTTON_STATE_PRESSED);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_MIDDLE, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_on_touch_4)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int which = litest_test_param_get_i32(test_env->params, "which"),
this = which % 4;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: wrap slot counting into a helper function This is prep work for future devices that announce a wrong slot count. For the tests this can be a problem if we rely on the correct slot count to decided whether to run a test or not. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-01-09 11:13:18 +10:00
if (litest_slot_count(dev) < 4)
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
/* 3fg tap with one finger detected as palm, that finger is lifted,
other two fingers lifted cause 2fg tap */
litest_touch_down(dev, this, 50, 50);
litest_touch_down(dev, (this + 1) % 4, 60, 50);
litest_touch_down(dev, (this + 2) % 4, 70, 50);
litest_touch_down(dev, (this + 3) % 4, 80, 50);
litest_drain_events(li);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, this, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, this);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, (this + 1) % 4);
litest_touch_up(dev, (this + 2) % 4);
litest_touch_up(dev, (this + 3) % 4);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_after_tap)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers");
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_multitap)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
msleep(10);
}
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
for (ntaps = 0; ntaps <= range; ntaps++) {
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
}
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_multitap_timeout)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
msleep(10);
}
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
for (ntaps = 0; ntaps <= range; ntaps++) {
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
}
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_multitap_down_again)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers + 1 > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
msleep(10);
}
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
/* keep palm finger down */
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 3, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 2, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 1, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 3);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 1);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
msleep(10);
}
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
test: fix a multitap test expecting one tap too few The problem was masked by a missing timeout, causing one up / down pair to not yet be enqueued before reading the queue.
2019-11-22 15:36:24 +01:00
for (ntaps = 0; ntaps <= 2 * range + 1; ntaps++) {
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
}
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_multitap_click)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
int nfingers = litest_test_param_get_i32(test_env->params, "fingers"),
range = litest_test_param_get_i32(test_env->params, "taps"), ntaps;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
unsigned int button = 0;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
if (nfingers > litest_slot_count(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 1:
button = BTN_LEFT;
break;
case 2:
button = BTN_RIGHT;
break;
case 3:
button = BTN_MIDDLE;
break;
default:
test: switch a bunch of tests from abort() to one with a message Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1062>
2024-10-14 21:35:36 +10:00
litest_abort_msg("Unexpected number of fingers (%d)", nfingers);
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
}
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
for (ntaps = 0; ntaps <= range; ntaps++) {
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
switch (nfingers) {
case 3:
litest_touch_down(dev, 2, 60, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_down(dev, 1, 50, 30);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_down(dev, 0, 40, 30);
break;
}
switch (nfingers) {
case 3:
litest_touch_up(dev, 2);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 2:
litest_touch_up(dev, 1);
Replace fallthrough comments with __attribute__((fallthrough)) This has recently been endorsed by the linux kernel, it should be good enough for us. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-07-22 15:00:32 +10:00
_fallthrough_;
test: expand drag and multitap tests to tap with up to 3 fingers This covers the addition of two- and three-finger tap-and-drag, as well as the fix for multitaps with more than one finger in later taps and the multifinger enhancement to the tap ending drag-lock. Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
2020-07-01 21:02:09 +02:00
case 1:
litest_touch_up(dev, 0);
break;
}
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
msleep(10);
}
litest_touch_down(dev, 0, 50, 50);
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
/* keep palm finger down */
litest_button_click(dev, BTN_LEFT, true);
litest_button_click(dev, BTN_LEFT, false);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
for (ntaps = 0; ntaps <= range; ntaps++) {
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, button, LIBINPUT_BUTTON_STATE_RELEASED);
}
/* the click */
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_click_then_tap)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
litest_touch_down_extended(dev, 0, 50, 50, axes);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_button_click(dev, BTN_LEFT, true);
litest_button_click(dev, BTN_LEFT, false);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_touch_up(dev, 0);
litest_assert_empty_queue(li);
litest_touch_down(dev, 0, 50, 50);
litest_touch_up(dev, 0);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_tap(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_LEFT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
START_TEST(touchpad_tap_palm_dwt_tap)
{
struct litest_device *dev = litest_current_device();
struct litest_device *keyboard;
struct libinput *li = dev->libinput;
struct axis_replacement axes[] = {
{ ABS_MT_PRESSURE, 75 },
{ -1, 0 },
};
if (!touchpad_has_palm_pressure(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
keyboard = litest_add_device(li, LITEST_KEYBOARD);
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_drain_events(li);
litest_keyboard_key(keyboard, KEY_A, true);
litest_keyboard_key(keyboard, KEY_B, true);
litest_keyboard_key(keyboard, KEY_A, false);
litest_drain_events(li);
litest_touch_down(dev, 0, 50, 50);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_keyboard_key(keyboard, KEY_B, false);
litest_drain_events(li);
test: make the timeout functions auto-dispatching The overwhelmingly vast majority of invocations want to have a dispatch before/after, so let's automate that. In case NULL is passed, that dispatch is skipped. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1169>
2025-04-01 09:23:10 +10:00
litest_timeout_dwt_long(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
/* Changes to palm after dwt timeout */
test: drop the sleep_ms argument This forces events for every ~10ms now. If we want a slower movement, we need more steps - just like a real touchpad does it. Cocinelle spatch files were variants of: @@ expression A, B, C, D, E, F, G, H, I, J, K; @@ - litest_touch_move_two_touches(A, B, C, D, E, F, G, H, I) + litest_touch_move_two_touches(A, B, C, D, E, F, G, H) The only test that needed a real fix was touchpad_no_palm_detect_2fg_scroll, it used 12ms before, now it's using 10ms so on the bcm5974 touchpad the second finger was a speed-thumb. Increasing the events and thus slowing down the pointer means it's a normal finger and the test succeeds again. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-08-28 09:16:52 +10:00
litest_touch_move_to_extended(dev, 0, 50, 50, 50, 50, axes, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_touch_up(dev, 0);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
litest_assert_empty_queue(li);
test: properly release a few leaking litest device Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-07-26 10:38:16 +10:00
test: rename litest_delete_device to litest_device_destroy To be closer to the common free/destroy/unref pattern Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1184>
2025-04-07 15:12:51 +10:00
litest_device_destroy(keyboard);
touchpad: work palm detection into the tap state machine Unlike the already-existing thumb detection, a touch may be labelled palm at any time, not just during the initial touch down. This requires full integration into the tap state machine to unwind properly. For most states, a palm detection simply ignores the finger and reverts to the most recent state. One exception is the case of two fingers down, one finger up followed by the remaining finger detected as a palm finger. This triggers a single-finger tap but with timestamps that may be from the wrong finger. Since we're within a short tap timeout anyway this should not matter too much. The special state PALM_UP is only handled in one condition (DEAD). Once a touch is a palm we basically skip over it from then on. If we end up in the DEAD state after a button press we still need to handle the palm up events accordingly to be able to return to IDLE. That transition also requires us to have an accurate count of the real fingers down (palms don't count) so we need a separate nfingers_down counter for tapping. https://bugs.freedesktop.org/show_bug.cgi?id=103210 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-11-03 10:59:36 +10:00
}
END_TEST
touchpad: correct the tap state transitions for a palm on TOUCH_BEGIN Where a touch is labelled as palm on TOUCH_BEGIN (edge palms) we would still feed the touch through the tap state machine. This would trigger the PALM transition for each state, usually reducing the touch count. When the touches were later released, the touch count was out of sync, resulting in an error message. In the case of #488, the trigger was a single evdev frame with three fingers down, the third of which was an edge palm: - touch 1 transitions from IDLE to TOUCH - touch 2 transitions from TOUCH to TOUCH_2 - touch 3 (the palm) transitioned from TOUCH_2 back to TOUCH That third transition is invalid, the palm hasn't been seen by the tap state machine so it should just be ignored. Fix this by moving making the tap state processing conditional on a touch state other than TOUCH_BEGIN. Fixes #488 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-05-28 10:58:06 +10:00
START_TEST(touchpad_tap_palm_3fg_start)
{
struct litest_device *dev = litest_current_device();
struct libinput *li = dev->libinput;
if (litest_slot_count(dev) < 3 || !litest_has_palm_detect_size(dev))
test: add litest-runner as test suite runner This replaces check. The code is a copy of pwtest which I wrote years ago for pipewire but adjusted for us here the last few days. There are a few advantages over check: - Ability to SKIP tests or mark them as NOT_APPLICABLE, the latter of which is used for early checks if a device doesn't meet requirements. - it captures stdout/stderr separately - colors! - YAML output format makes it a lot easier to read the results and eventually parse them for e.g. "restart failed tests" Less abstraction: we set up the tests, pass them to the runner and run them with the given number of forks. This is an improvement over before where we forked into N test suites which each called check which then forked again. Since we're now keeping track of those processes ourselves we can also write tests that are expected to fail with signals. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1067>
2024-10-12 10:31:42 +10:00
return LITEST_NOT_APPLICABLE;
touchpad: correct the tap state transitions for a palm on TOUCH_BEGIN Where a touch is labelled as palm on TOUCH_BEGIN (edge palms) we would still feed the touch through the tap state machine. This would trigger the PALM transition for each state, usually reducing the touch count. When the touches were later released, the touch count was out of sync, resulting in an error message. In the case of #488, the trigger was a single evdev frame with three fingers down, the third of which was an edge palm: - touch 1 transitions from IDLE to TOUCH - touch 2 transitions from TOUCH to TOUCH_2 - touch 3 (the palm) transitioned from TOUCH_2 back to TOUCH That third transition is invalid, the palm hasn't been seen by the tap state machine so it should just be ignored. Fix this by moving making the tap state processing conditional on a touch state other than TOUCH_BEGIN. Fixes #488 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-05-28 10:58:06 +10:00
litest_enable_tap(dev->libinput_device);
gestures: add hold gesture implementation Hold gestures are notifications about fingers on the touchpad. There is no coordinate attached to a hold gesture, merely the number of fingers. A hold gesture starts when the user places a finger on the touchpad and ends when all fingers are lifted. It is cancelled when the finger(s) move past applicable thresholds and trigger some other interaction like pointer movement or scrolling. Signed-off-by: José Expósito <jose.exposito89@gmail.com>
2021-05-27 19:20:03 +02:00
litest_disable_hold_gestures(dev->libinput_device);
touchpad: correct the tap state transitions for a palm on TOUCH_BEGIN Where a touch is labelled as palm on TOUCH_BEGIN (edge palms) we would still feed the touch through the tap state machine. This would trigger the PALM transition for each state, usually reducing the touch count. When the touches were later released, the touch count was out of sync, resulting in an error message. In the case of #488, the trigger was a single evdev frame with three fingers down, the third of which was an edge palm: - touch 1 transitions from IDLE to TOUCH - touch 2 transitions from TOUCH to TOUCH_2 - touch 3 (the palm) transitioned from TOUCH_2 back to TOUCH That third transition is invalid, the palm hasn't been seen by the tap state machine so it should just be ignored. Fix this by moving making the tap state processing conditional on a touch state other than TOUCH_BEGIN. Fixes #488 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-05-28 10:58:06 +10:00
litest_drain_events(li);
litest_push_event_frame(dev);
litest_touch_down(dev, 0, 50, 50);
litest_touch_down(dev, 1, 55, 55);
litest_touch_down(dev, 2, 99, 55); /* edge palm */
litest_pop_event_frame(dev);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: correct the tap state transitions for a palm on TOUCH_BEGIN Where a touch is labelled as palm on TOUCH_BEGIN (edge palms) we would still feed the touch through the tap state machine. This would trigger the PALM transition for each state, usually reducing the touch count. When the touches were later released, the touch count was out of sync, resulting in an error message. In the case of #488, the trigger was a single evdev frame with three fingers down, the third of which was an edge palm: - touch 1 transitions from IDLE to TOUCH - touch 2 transitions from TOUCH to TOUCH_2 - touch 3 (the palm) transitioned from TOUCH_2 back to TOUCH That third transition is invalid, the palm hasn't been seen by the tap state machine so it should just be ignored. Fix this by moving making the tap state processing conditional on a touch state other than TOUCH_BEGIN. Fixes #488 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-05-28 10:58:06 +10:00
litest_touch_up(dev, 2); /* release the palm */
litest_assert_empty_queue(li);
litest_touch_up(dev, 0);
litest_touch_up(dev, 1);
test: add litest_dispatch() for better test debugging Wraps libinput_dispatch() with a location which will make things a bit easier to track. Output (in --verbose) is something like: gestures_swipe_3fg_unaccel_fn():1346 - dispatching Which makes it easier to associate the various calls to libinput dispatch with the other output from libinput. This patch switches all uses of libinput_dispatch() in test cases over but not the litest functions that may call dispatch too. Remains to be seen if that is necessary. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
2024-09-13 15:13:38 +10:00
litest_dispatch(li);
touchpad: correct the tap state transitions for a palm on TOUCH_BEGIN Where a touch is labelled as palm on TOUCH_BEGIN (edge palms) we would still feed the touch through the tap state machine. This would trigger the PALM transition for each state, usually reducing the touch count. When the touches were later released, the touch count was out of sync, resulting in an error message. In the case of #488, the trigger was a single evdev frame with three fingers down, the third of which was an edge palm: - touch 1 transitions from IDLE to TOUCH - touch 2 transitions from TOUCH to TOUCH_2 - touch 3 (the palm) transitioned from TOUCH_2 back to TOUCH That third transition is invalid, the palm hasn't been seen by the tap state machine so it should just be ignored. Fix this by moving making the tap state processing conditional on a touch state other than TOUCH_BEGIN. Fixes #488 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-05-28 10:58:06 +10:00
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_PRESSED);
litest_assert_button_event(li, BTN_RIGHT, LIBINPUT_BUTTON_STATE_RELEASED);
litest_assert_empty_queue(li);
}
END_TEST
test: move all the _setup() functions into a special section This way we can loop through them instead of having to add them manually. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2018-03-21 12:54:10 +10:00
TEST_COLLECTION(touchpad_tap)
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
{
test: disable clang-format in test collections It's too much effort fighting clang-format for these snippets which all don't really do much anyway but are important to be read easily. Let's categorically disable all formatting in the test collections and move on. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1246>
2025-07-01 11:30:59 +10:00
/* clang-format off */
test: drop the custom group names The group names are forced by check (they are called suite names there) but for our test suite they provide very little benefit. Much easier to just use the filename a test is in as group name. This removes the pure substring match for --filter-group, it's now fnmatch only. group names are short enough that the typing isn't an issue and we don't want to run tests twice (e.g. 'pad' is also in 'touchpad'). This patch caused #574 until it got fixed in d838e3a3a4c Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-05 14:51:02 +10:00
litest_add(touchpad_1fg_tap, LITEST_TOUCHPAD, LITEST_ANY);
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
litest_with_parameters(params, "fingers_1st", 'i', 3, 1, 2, 3,
"fingers_2nd", 'i', 3, 1, 2, 3) {
litest_add_parametrized(touchpad_doubletap, LITEST_TOUCHPAD, LITEST_ANY, params);
}
litest_with_parameters(params, "fingers", 'i', 3, 1, 2, 3,
test: fix off-by-ones in two ranges, prevent the same in remaining ranged tests touchpad_move_after_touch contains support for up to 5 fingers but was only run with 2..4 fingers. touchpad_multitap and several others using the same parameter definition were long ago tested with 3..7 taps. In 8f92b091 this was reduced to 3..4 for CI performance, while the commit message indicates 3..5 were intended. The common theme is that the upper bound of `struct range` is interpreted as exclusive, while some uses assumed it would be inclusive. There are relatively recent helper functions range_init_inclusive and range_init_exclusive (since 817dc423) to avoid this trap. Use them on the remaining two ranged tests. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-16 17:05:39 +01:00
"taps", 'i', 3, 3, 4, 5) {
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
litest_add_parametrized(touchpad_multitap, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_multitap_timeout, LITEST_TOUCHPAD, LITEST_ANY, params);
}
test: drop the custom group names The group names are forced by check (they are called suite names there) but for our test suite they provide very little benefit. Much easier to just use the filename a test is in as group name. This removes the pure substring match for --filter-group, it's now fnmatch only. group names are short enough that the typing isn't an issue and we don't want to run tests twice (e.g. 'pad' is also in 'touchpad'). This patch caused #574 until it got fixed in d838e3a3a4c Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-05 14:51:02 +10:00
litest_add(touchpad_2fg_tap_move_on_release, LITEST_TOUCHPAD|LITEST_SEMI_MT, LITEST_SINGLE_TOUCH);
litest_add(touchpad_2fg_tap_n_hold_first, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH);
litest_add(touchpad_2fg_tap_n_hold_second, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH);
litest_add(touchpad_2fg_tap_quickrelease, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT);
litest_add(touchpad_1fg_tap_click, LITEST_TOUCHPAD|LITEST_BUTTON, LITEST_CLICKPAD);
litest_add(touchpad_2fg_tap_click, LITEST_TOUCHPAD|LITEST_BUTTON, LITEST_SINGLE_TOUCH|LITEST_CLICKPAD);
litest_add(touchpad_2fg_tap_click_apple, LITEST_APPLE_CLICKPAD, LITEST_ANY);
litest_add(touchpad_no_2fg_tap_after_move, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT);
litest_add(touchpad_no_2fg_tap_after_timeout, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT);
litest_add(touchpad_no_first_fg_tap_after_move, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT);
litest_add(touchpad_3fg_tap_tap_again, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH);
litest_add(touchpad_3fg_tap_quickrelease, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH);
litest_add(touchpad_3fg_tap_hover_btntool, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH);
litest_add(touchpad_3fg_tap_pressure_btntool, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH);
litest_add_for_device(touchpad_3fg_tap_slot_release_btntool, LITEST_SYNAPTICS_TOPBUTTONPAD);
litest_add(touchpad_3fg_tap_after_scroll, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH);
litest_add(touchpad_4fg_tap, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT);
litest_add(touchpad_4fg_tap_quickrelease, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT);
litest_add(touchpad_5fg_tap, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT);
litest_add(touchpad_5fg_tap_quickrelease, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT);
test: add optional value names to parametrized tests Only implemented for i32 values so far, used for enums and enum-like constants, replaces a runtime lookup from stringly-typed parameters. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1141>
2025-02-13 22:10:02 +01:00
litest_with_parameters(params, "map", 'I', 2, litest_named_i32(LIBINPUT_CONFIG_TAP_MAP_LRM, "LRM"),
litest_named_i32(LIBINPUT_CONFIG_TAP_MAP_LMR, "LMR")) {
test: change tap tests to use parameters This also fixes the invocation of the touchpad_3fg_tap_btntool_pointerjump which was written as ranged test but never got invoked with a range (but _i defaults to zero). Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1119>
2025-01-08 09:21:14 +10:00
litest_add_parametrized(touchpad_2fg_tap, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT, params);
litest_add_parametrized(touchpad_2fg_tap_inverted, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_3fg_tap_btntool, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_3fg_tap_btntool_inverted, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized_for_device(touchpad_3fg_tap_btntool_pointerjump, LITEST_SYNAPTICS_TOPBUTTONPAD, params);
litest_add_parametrized(touchpad_3fg_tap, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
}
test: fix off-by-ones in two ranges, prevent the same in remaining ranged tests touchpad_move_after_touch contains support for up to 5 fingers but was only run with 2..4 fingers. touchpad_multitap and several others using the same parameter definition were long ago tested with 3..7 taps. In 8f92b091 this was reduced to 3..4 for CI performance, while the commit message indicates 3..5 were intended. The common theme is that the upper bound of `struct range` is interpreted as exclusive, while some uses assumed it would be inclusive. There are relatively recent helper functions range_init_inclusive and range_init_exclusive (since 817dc423) to avoid this trap. Use them on the remaining two ranged tests. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-16 17:05:39 +01:00
litest_with_parameters(params, "fingers", 'i', 4, 2, 3, 4, 5) {
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
litest_add_parametrized(touchpad_move_after_touch, LITEST_TOUCHPAD, LITEST_ANY, params);
}
test: drop the custom group names The group names are forced by check (they are called suite names there) but for our test suite they provide very little benefit. Much easier to just use the filename a test is in as group name. This removes the pure substring match for --filter-group, it's now fnmatch only. group names are short enough that the typing isn't an issue and we don't want to run tests twice (e.g. 'pad' is also in 'touchpad'). This patch caused #574 until it got fixed in d838e3a3a4c Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-05 14:51:02 +10:00
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
/* Real buttons don't interfere with tapping, so don't run those for
pads with buttons */
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
litest_with_parameters(params, "fingers", 'i', 3, 1, 2, 3) {
litest_add_parametrized(touchpad_double_tap_click, LITEST_CLICKPAD, LITEST_ANY, params);
}
test: drop the custom group names The group names are forced by check (they are called suite names there) but for our test suite they provide very little benefit. Much easier to just use the filename a test is in as group name. This removes the pure substring match for --filter-group, it's now fnmatch only. group names are short enough that the typing isn't an issue and we don't want to run tests twice (e.g. 'pad' is also in 'touchpad'). This patch caused #574 until it got fixed in d838e3a3a4c Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-05 14:51:02 +10:00
litest_add(touchpad_tap_default_disabled, LITEST_TOUCHPAD|LITEST_BUTTON, LITEST_ANY);
litest_add(touchpad_tap_default_enabled, LITEST_TOUCHPAD, LITEST_BUTTON);
litest_add(touchpad_tap_invalid, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_tap_is_available, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_tap_is_not_available, LITEST_ANY, LITEST_TOUCHPAD);
litest_add(touchpad_tap_default_map, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_tap_map_unsupported, LITEST_ANY, LITEST_TOUCHPAD);
litest_add(touchpad_tap_set_map, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_tap_set_map_no_tapping, LITEST_ANY, LITEST_TOUCHPAD);
litest_add(touchpad_tap_get_map_no_tapping, LITEST_ANY, LITEST_TOUCHPAD);
litest_add(touchpad_tap_map_delayed, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_SEMI_MT);
litest_add(clickpad_1fg_tap_click, LITEST_CLICKPAD, LITEST_ANY);
litest_add(clickpad_2fg_tap_click, LITEST_CLICKPAD, LITEST_SINGLE_TOUCH|LITEST_APPLE_CLICKPAD);
test: disable clang-format in test collections It's too much effort fighting clang-format for these snippets which all don't really do much anyway but are important to be read easily. Let's categorically disable all formatting in the test collections and move on. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1246>
2025-07-01 11:30:59 +10:00
/* clang-format on */
test: split out the touchpad tap tests into multiple collections These take a long time and have a reasonable high chance of failure due to the timing constraints. Let's split them up so they don't hog the runners for that long and in case they fail, we only need to re-run a short test. Before: one test running approx 21 min, now 3 tests running approx 7 + 11 + 4 min. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1065>
2024-10-17 19:35:16 +10:00
}
TEST_COLLECTION(touchpad_tap_drag)
{
test: disable clang-format in test collections It's too much effort fighting clang-format for these snippets which all don't really do much anyway but are important to be read easily. Let's categorically disable all formatting in the test collections and move on. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1246>
2025-07-01 11:30:59 +10:00
/* clang-format off */
test: drop the custom group names The group names are forced by check (they are called suite names there) but for our test suite they provide very little benefit. Much easier to just use the filename a test is in as group name. This removes the pure substring match for --filter-group, it's now fnmatch only. group names are short enough that the typing isn't an issue and we don't want to run tests twice (e.g. 'pad' is also in 'touchpad'). This patch caused #574 until it got fixed in d838e3a3a4c Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-05 14:51:02 +10:00
litest_add(touchpad_drag_lock_default_disabled, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_drag_lock_default_unavailable, LITEST_ANY, LITEST_TOUCHPAD);
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
litest_with_parameters(params, "fingers", 'i', 3, 1, 2, 3,
test: fix off-by-ones in two ranges, prevent the same in remaining ranged tests touchpad_move_after_touch contains support for up to 5 fingers but was only run with 2..4 fingers. touchpad_multitap and several others using the same parameter definition were long ago tested with 3..7 taps. In 8f92b091 this was reduced to 3..4 for CI performance, while the commit message indicates 3..5 were intended. The common theme is that the upper bound of `struct range` is interpreted as exclusive, while some uses assumed it would be inclusive. There are relatively recent helper functions range_init_inclusive and range_init_exclusive (since 817dc423) to avoid this trap. Use them on the remaining two ranged tests. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-16 17:05:39 +01:00
"taps", 'i', 3, 3, 4, 5) {
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
litest_add_parametrized(touchpad_multitap_n_drag_tap_click, LITEST_CLICKPAD, LITEST_ANY, params);
test: split out the touchpad tap tests into multiple collections These take a long time and have a reasonable high chance of failure due to the timing constraints. Let's split them up so they don't hog the runners for that long and in case they fail, we only need to re-run a short test. Before: one test running approx 21 min, now 3 tests running approx 7 + 11 + 4 min. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1065>
2024-10-17 19:35:16 +10:00
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
litest_add_parametrized(touchpad_multitap_n_drag_timeout, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_multitap_n_drag_tap, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_multitap_n_drag_move, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_multitap_n_drag_2fg, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_multitap_n_drag_click, LITEST_CLICKPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_drag_disabled_multitap_no_drag, LITEST_TOUCHPAD, LITEST_ANY, params);
}
litest_with_parameters(params, "fingers", 'i', 3, 1, 2, 3,
"taps", 'i', 3, 1, 2, 3) {
litest_add_parametrized(touchpad_multitap_n_drag_high_delay, LITEST_TOUCHPAD, LITEST_ANY, params);
}
litest_with_parameters(params, "fingers", 'i', 3, 1, 2, 3) {
litest_add_parametrized(touchpad_tap_n_drag_click, LITEST_CLICKPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_n_drag_draglock_tap_click, LITEST_CLICKPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_n_drag, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_n_drag_2fg, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_tap_n_drag_2fg_scroll, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_tap_n_drag_draglock_2fg_scroll, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_tap_n_drag_3fg_btntool, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH|LITEST_APPLE_CLICKPAD, params);
litest_add_parametrized(touchpad_tap_n_drag_3fg, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_tap_n_drag_3fg_swipe, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_tap_n_drag_draglock_3fg_swipe, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_tap_n_drag_draglock, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_n_drag_draglock_timeout, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_n_drag_draglock_sticky, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_drag_disabled, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_drag_disabled_immediate, LITEST_TOUCHPAD, LITEST_ANY, params);
}
litest_with_parameters(params, "fingers_1st", 'i', 3, 1, 2, 3,
"fingers_2nd", 'i', 3, 1, 2, 3) {
litest_add_parametrized(touchpad_tap_n_drag_draglock_tap, LITEST_TOUCHPAD, LITEST_ANY, params);
}
test: split out the touchpad tap tests into multiple collections These take a long time and have a reasonable high chance of failure due to the timing constraints. Let's split them up so they don't hog the runners for that long and in case they fail, we only need to re-run a short test. Before: one test running approx 21 min, now 3 tests running approx 7 + 11 + 4 min. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1065>
2024-10-17 19:35:16 +10:00
test: drop the custom group names The group names are forced by check (they are called suite names there) but for our test suite they provide very little benefit. Much easier to just use the filename a test is in as group name. This removes the pure substring match for --filter-group, it's now fnmatch only. group names are short enough that the typing isn't an issue and we don't want to run tests twice (e.g. 'pad' is also in 'touchpad'). This patch caused #574 until it got fixed in d838e3a3a4c Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-05 14:51:02 +10:00
litest_add(touchpad_drag_default_disabled, LITEST_ANY, LITEST_TOUCHPAD);
litest_add(touchpad_drag_default_enabled, LITEST_TOUCHPAD, LITEST_BUTTON);
litest_add(touchpad_drag_config_invalid, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_drag_config_unsupported, LITEST_ANY, LITEST_TOUCHPAD);
litest_add(touchpad_drag_config_enabledisable, LITEST_TOUCHPAD, LITEST_ANY);
test: disable clang-format in test collections It's too much effort fighting clang-format for these snippets which all don't really do much anyway but are important to be read easily. Let's categorically disable all formatting in the test collections and move on. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1246>
2025-07-01 11:30:59 +10:00
/* clang-format on */
test: split out the touchpad tap tests into multiple collections These take a long time and have a reasonable high chance of failure due to the timing constraints. Let's split them up so they don't hog the runners for that long and in case they fail, we only need to re-run a short test. Before: one test running approx 21 min, now 3 tests running approx 7 + 11 + 4 min. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1065>
2024-10-17 19:35:16 +10:00
}
TEST_COLLECTION(touchpad_tap_palm)
{
test: disable clang-format in test collections It's too much effort fighting clang-format for these snippets which all don't really do much anyway but are important to be read easily. Let's categorically disable all formatting in the test collections and move on. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1246>
2025-07-01 11:30:59 +10:00
/* clang-format off */
test: drop the custom group names The group names are forced by check (they are called suite names there) but for our test suite they provide very little benefit. Much easier to just use the filename a test is in as group name. This removes the pure substring match for --filter-group, it's now fnmatch only. group names are short enough that the typing isn't an issue and we don't want to run tests twice (e.g. 'pad' is also in 'touchpad'). This patch caused #574 until it got fixed in d838e3a3a4c Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-05 14:51:02 +10:00
litest_add(touchpad_tap_palm_on_idle, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_tap_palm_on_touch, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_tap_palm_on_touch_hold_timeout, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_tap_palm_on_touch_hold_move, LITEST_TOUCHPAD, LITEST_ANY);
test: convert most ranged tests to parametrized ones The two remaining ranged tests (abs_device_no_range, abs_mt_device_no_range) are better served staying with ranges because parametrized tests need to explicitly list all members of the range, which for these tests is not only pretty big, but also contains abs axes reserved for future use. Those axes have no names yet, making a future-proof conversion pretty much impossible. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1142>
2025-02-06 20:14:57 +01:00
litest_with_parameters(params, "fingers", 'i', 3, 1, 2, 3) {
litest_add_parametrized(touchpad_tap_palm_on_tapped, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_palm_on_tapped_palm_down, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_palm_on_drag, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_palm_after_tap, LITEST_TOUCHPAD, LITEST_ANY, params);
}
litest_with_parameters(params, "fingers_1st", 'i', 3, 1, 2, 3,
"fingers_2nd", 'i', 3, 1, 2, 3) {
litest_add_parametrized(touchpad_tap_palm_on_tapped_doubletap, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
}
litest_with_parameters(params, "fingers", 'i', 3, 1, 2, 3,
"which", 'I', 2, litest_named_i32(0, "first"), litest_named_i32(1, "second")) {
litest_add_parametrized(touchpad_tap_palm_on_drag_2fg, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
}
litest_with_parameters(params, "which", 'I', 2, litest_named_i32(0, "first"), litest_named_i32(1, "second")) {
litest_add_parametrized(touchpad_tap_palm_on_touch_2, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_tap_palm_on_touch_2_retouch, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
}
litest_with_parameters(params, "which", 'I', 3, litest_named_i32(0, "first"), litest_named_i32(1, "second"),
litest_named_i32(2, "third")) {
litest_add_parametrized(touchpad_tap_palm_on_touch_3, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_tap_palm_on_touch_3_retouch, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
}
litest_with_parameters(params, "which", 'I', 4, litest_named_i32(0, "first"), litest_named_i32(1, "second"),
litest_named_i32(2, "third"), litest_named_i32(3, "fourth")) {
litest_add_parametrized(touchpad_tap_palm_on_touch_4, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
}
litest_with_parameters(params, "fingers", 'i', 3, 1, 2, 3,
"taps", 'i', 3, 3, 4, 5) {
litest_add_parametrized(touchpad_tap_palm_multitap, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_palm_multitap_timeout, LITEST_TOUCHPAD, LITEST_ANY, params);
litest_add_parametrized(touchpad_tap_palm_multitap_down_again, LITEST_TOUCHPAD, LITEST_SINGLE_TOUCH, params);
litest_add_parametrized(touchpad_tap_palm_multitap_click, LITEST_CLICKPAD, LITEST_ANY, params);
}
test: drop the custom group names The group names are forced by check (they are called suite names there) but for our test suite they provide very little benefit. Much easier to just use the filename a test is in as group name. This removes the pure substring match for --filter-group, it's now fnmatch only. group names are short enough that the typing isn't an issue and we don't want to run tests twice (e.g. 'pad' is also in 'touchpad'). This patch caused #574 until it got fixed in d838e3a3a4c Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-02-05 14:51:02 +10:00
litest_add(touchpad_tap_palm_click_then_tap, LITEST_CLICKPAD, LITEST_ANY);
litest_add(touchpad_tap_palm_dwt_tap, LITEST_TOUCHPAD, LITEST_ANY);
litest_add(touchpad_tap_palm_3fg_start, LITEST_TOUCHPAD, LITEST_ANY);
test: disable clang-format in test collections It's too much effort fighting clang-format for these snippets which all don't really do much anyway but are important to be read easily. Let's categorically disable all formatting in the test collections and move on. Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1246>
2025-07-01 11:30:59 +10:00
/* clang-format on */
test: move tapping tests into a separate binary The previous set hit _some_ sort of limit, but no idea what or why. When adding one more test, the touchpad test case would reliably fail with a udev timeout in litest_wait_for_udev(). This only happened in the valgrind case, the normal run succeeded. Reproduced on three different installations (2 vms on two different hosts). Move the tapping tests into a separate binary, this unwedges whatever was unhappy and sunshine, lollipops and rainbows are distributed generously. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2015-06-19 14:58:16 +10:00
}
Reference in a new issue Copy permalink