On a very loaded machine, the process might not receive the quit signal
in a timely manner, and this introduce false positive results.
Add a longer timeout. This shouldn't interfere with the global time
spent in the tests, but will allow some loaded environment to pass
the tests.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
This parameter is already included by default in ci-templates, but
we also need it in freebsd
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
Gitlab supports masked tokens that get sanitized during log output but these
tokens are still in the environment. meson dumps the environment into
testlog.txt, resulting in our tokens leaking.
Avoid that leak by using a netrc file instead. The token value now refers to
the file name which is safe enough to leak into the test logs.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
In evdev_device_calibrate, the user matrix was not being stored when it
was the identity matrix. This resulted in
libinput_device_config_calibration_get_matrix not providing the correct
matrix. Instead of giving the identity matrix, the last non-identity
matrix set was given.
This just moves the storage of the user matrix in
evdev_device_calibrate to be above the identity matrix early return so
that it always get stored.
Signed-off-by: Brian Ashworth <bosrsf04@gmail.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
meson implicitly sets install to whether install_dir is nonzero. Which means
it's superfluous anyway and removing it drops the meson warning:
WARNING: Project specifies a minimum meson_version '>= 0.41.0' but uses
features which were added in newer versions:
* 0.50.0: {'install arg in configure_file'}
Fixes#334
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
We keep running into the proximity timeout for these tests, especially under
valgrind. To avoid this, manually intersperse the touch events with tablet
events.
Note that this manual loop would just work even without tablet events
because we no longer have a 10ms delay between touch events as enforced by
litest_touch_move_to. But let's do the right thing anyway.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
In order for two devices to be in the same group, they need to share
identical LIBINPUT_DEVICE_GROUP attributes. The `wacom_handle_ekr` function
overwrites the VID/PID for an ExpressKey Remote, but the 'phys' path is
left unchanged. This only works if the EKR and the device we want to pair
it with are both direct sibings in the USB tree. It isn't always possible
to actually connect the devices like this, however. The Cintiq Pro 32 and
24, for instance, have multiple internal USB hubs and place the pen sensor
and the USB port for the EKR dongle behind different ones.
By copying the 'phys' path of the device we want to pair with, it is
possible to reproduce the entire LIBINPUT_DEVICE_GROUP and ensure that
the two devices actually end up paired in libinput.
Signed-off-by: Jason Gerecke <jason.gerecke@wacom.com>
Instead of a simple yes/no/maybe for thumbs, have a more extensive state
machine that keeps track of the thumb. Since we only support one thumb anyway,
the tracking moves to the tp_dispatch struct.
Test case changes:
touchpad_clickfinger_3fg_tool_position:
with better thumb detection we can now handle this properly and expect a
right button (2fg) press for the test case
touchpad_thumb_no_doublethumb_with_timeout:
two thumbs are now always two fingers, so let's switch to axis events here
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
We don't need speed detection for non-clickpads - the only reason to ever drop
a second finger on those is to either scroll or trigger a gesture. Unlike
clickpads, where a dropped finger may be a thumb to click.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Check if there's a thumb if we have two touches. If one finger moves but
the thumb remains still, we assume that one is really a thumb. But if the
thumb moves while the finger is still, let's assume this is a 2-finger scroll.
Extracted from Matt Mayfield's thumb detection patchset
We can't detect pinch when gestures are off anyway, so we don't need to check
the finger distances.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Put some basic location checks in, if the fingers are next to each other and
vertically close, assume scroll over swipe.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>:
When a touchpad has thumb detection enabled, avoid false-positive gestures
involving a resting thumb by using two thresholds: inner and outer.
While both touches remain inside their inner thresholds, remain in UNKNOWN
state to allow for accurate gesture detection even with no timeout.
If both touches move outside their inner thresholds, start a pinch or
swipe/scroll gesture according to direction, as usual.
If one touch moves outside its outer threshold while the other has not yet
exceeded its inner threshold, and thumb detection is enabled, then if one
touch is >20mm lower, mark it as a thumb and cancel the gesture.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
No functional changes, just prep work for a later patch where the thumbs will
dynamically update their state (instead of just using yes/no/maybe).
Extracted from Matt Mayfield's thumb detection patches.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This moves the thumb state logging directly into that helper function too.
Extracted from Matt Mayfield's thumb detection patches.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Currently the same as tp_touch_active() but this will change.
No functional changes.
Extracted from Matt Mayfield's thumb detection patches.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
The previously 'scroll'-named timeout is also used for swipe, so let's rename
it. And the pinch one isn't used at all.
Extracted from Matt Mayfield's thumb detection patches.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
We can't run this in parallel so it doesn't gain us any speed advantage. If
anything, it'll be slower because it's more setup time in between. But: meson
doesn't display the result until the test suite finished, so having this
broken up into smaller chunks means we're more likely to see a general failure
early.
And the failure should be quicker to reproduce as well.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
All filter arguments currently force a -j1 unless otherwise specified. Change
this for --filter-group since that one is most likely invoked by some test
setup that can either add -j X or set the environment variable LITEST_JOBS as
well.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
