They don't support C11 anyway. Additionally, CLang defines a very
conservative __GNUC__ value (which is also configurable via the
-fgnuc-version= command-line argument)
Most GDI functions do not set the last error, so GetLastError() returns
unrelated error codes. There are some exceptions, however, like BitBlt
and CreateDIBSection.
Whether a GDI function sets the last error is stated in the reference
documentation on MSDN.
There are two check-link.c files which were "tests" in the autotools
build and are unused ever since the autotools build was removed.
Signed-off-by: Uli Schlachter <psychon@znc.in>
We should default on every platform we care about to hidden symbols, to
avoid leaking private symbols.
On Windows this is the default state of affairs with the MSVC toolchain;
with GCC and GCC-compatible toolchains, we need to opt into this
behaviour. Luckily for us, Cairo already has an annotation for public
symbols, so we can easily tweak it to include the visibility attribute.
When building ancillary libraries as part of the Cairo compilation on
Windows, we use a pre-processor symbol to ensure that we keep the
dllexport annotation. This avoids including the cairoint.h header file.
Fixes: #582
The cairo-boilerplate static library cannot use private API defined in
the main Cairo shared library, because it has no access to those
symbols.
Since the code is small, we can just dump it into the boilerplate
library.
The default RGB colorspace must be converted to the GPU's colorspace
using CGColorTransformConvertUsingCMSConverter. Profiling has shown this
function to consume as much as 48% of a redraw cycle in gdk-quartz.
There seems to be no named colorspace that matches the one stored on the
display, so we use the one associated with the main display. This has
some risks for users with multiple monitors but in testing with my own
two-monitor setup, one of which is HDR and the other not, the colorspace
was the same for both.
This is applied to quartz surfaces created with
cairo_quartz_surface_create(); surfaces created with
cairo_quartz_surface_create_for_cg_context will inherit the colorspace
from the context.
In order to generate PNGs that look right I've converted the
existing debugging functions for writing a quartz surface to png
into private functions and wired cairo-boilerplate-quartz to use
them. Using the generic cairo routine produced washed-out PNGs.
Fixes https://gitlab.freedesktop.org/cairo/cairo/-/issues/330
The GL support in Cairo has always been a prototype, and
nothing happened in the past 10+ years to make it work as
it was meant to.
GL support is not enabled by any downstream packagers of
Cairo, so nobody should notice its absence.
WINVER and _WIN32_WINNT macros were defined in each source files and
headers that were including <windows.h>. However, because DirectWrite
requires new Windows API, some files included <windows.h> without the
version macros. This inconsistency sometimes caused troubles.
Define the version macros in meson.build.
In !309 Taylor R Campbell found a number of instances of ctype
incorrectly passed a signed char. In many cases, where only ASCII
characters are used, the code should have been using the cairo version
of the ctype function to avoid locale issues.
Nobody cares about BeOS in 2022, and it cannot be built with Meson anyway.
Removing the BeOS backend also allows us to drop the dependency on a C++
compiler and simplify the Autotools build considerably by removing all
the extra CXX targets.
The FT change is because my MinGW build is using a more recent version
of FT.
Remove the disabled _cairo_win32_scaled_font_text_to_glyphs() code to
fix the defined but not used warning.
_cairo_win32_scaled_font_text_to_glyphs() was diabled in d9408041aa with
the comment:
"Currently disable the win32-font text_to_glyphs(), until that one
is updated. Or better yet, remove it and implement
ucs4_to_index(). It's the toy font API afterall."
_cairo_win32_scaled_font_ucs4_to_index() was added in d1c619bc7d.
This changes most instances of malloc() calls to use Cairo's safer
_cairo_malloc(). The malloc() call in the implementation of
boilerplate's xmalloc() is not changed since it already includes a
size=0 check.
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
When the tests run ghostscipt on ps2 files it crashes with:
Error: /undefined in .setlanguagelevel
Operand stack:
2
Execution stack:
%interp_exit .runexec2 --nostringval-- --nostringval-- --nostringval-- 2 %stopped_push --nostringval-- --nostringval-- --nostringval-- false 1 %stopped_push .runexec2 --nostringval-- --nostringval-- --nostringval-- 2 %stopped_push --nostringval--
Dictionary stack:
--dict:732/1123(ro)(G)-- --dict:0/20(G)-- --dict:75/200(L)--
Current allocation mode is local
GPL Ghostscript 9.53.3: Unrecoverable error, exit code 1
The problem is the "-c 2 .setlanguagelevel -f" on the gs command line
to set ghostscript to language level 2. I'm not sure why this no longer works.
.setlanguagelevel is not mentioned any where in the gs custom operators listed in
https://www.ghostscript.com/doc/current/Language.htm
I assume this operator was meant for ghostscript internal use and
is no longer exposed in recent versions. Remove it from the test.
For example, to depend on cairo-script, inccairoscript was added to
"include_directories:" and libcairoscript was added to "link_with:".
This commit instead uses the libcairoscript_dep dependency everywhere.
Signed-off-by: Uli Schlachter <psychon@znc.in>
We *always* generate this file, and we depend on its existence.
The idea behind HAVE_CONFIG_H was being able to include random files
from different projects, back in a time where "libraries" were literally
just random files instead of actual shared objects.
Since we're not in the '80s any more, and our build system(s) define
HAVE_CONFIG_H *and* generate the config.h header file, we don't need a
conditional guard around its inclusion.
We define _GNU_SOURCE globally in both the Autotools build, through the
use of the AC_USE_SYSTEM_EXTENSIONS macro; and in the Meson build, with
add_project_arguments().
Precisely what Autotools does, instead of adding it as per-target C
argument.
Once we remove HAVE_CONFIG_H checks in every source file, we'll be able
to drop it.
Fixes errors such as
Traceback (most recent call last):
File "C:\Users\...\cairo\test\make-cairo-test-constructors.py", line 19, in <module>
for l in f.readlines():
File "c:\python39\lib\encodings\cp1253.py", line 23, in decode
return codecs.charmap_decode(input,self.errors,decoding_table)[0]
UnicodeDecodeError: 'charmap' codec can't decode byte 0x98 in position 6694: character maps to <undefined>
on non-English-language Windows locales/installations.
This gets rid of the following two leaks reported by valgrind when successfully
running a test with CAIRO_TEST_TARGET=xcb:
8,000 bytes in 2 blocks are definitely lost in loss record 9 of 10
at 0x483877F: malloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
by 0x4B7E135: read_packet (xcb_in.c:259)
by 0x4B7E135: _xcb_in_read (xcb_in.c:1031)
by 0x4B7BF8E: _xcb_conn_wait (xcb_conn.c:516)
by 0x4B7D6AE: wait_for_reply (xcb_in.c:516)
by 0x4B7D8C8: xcb_request_check (xcb_in.c:745)
by 0x18D747: _cairo_boilerplate_xcb_create_surface (cairo-boilerplate-xcb.c:310)
by 0x12906F: cairo_test_for_target (cairo-test.c:819)
by 0x12AAB5: _cairo_test_context_run_for_target (cairo-test.c:1555)
by 0x126921: _cairo_test_runner_draw (cairo-test-runner.c:250)
by 0x126921: main (cairo-test-runner.c:932)
8,000 bytes in 2 blocks are definitely lost in loss record 10 of 10
at 0x483877F: malloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
by 0x4B7E135: read_packet (xcb_in.c:259)
by 0x4B7E135: _xcb_in_read (xcb_in.c:1031)
by 0x4B7BF8E: _xcb_conn_wait (xcb_conn.c:516)
by 0x4B7D6AE: wait_for_reply (xcb_in.c:516)
by 0x4B7D7C0: xcb_wait_for_reply (xcb_in.c:546)
by 0x18D45D: find_depth (cairo-boilerplate-xcb.c:154)
by 0x18D45D: _cairo_boilerplate_xcb_create_render_0_0 (cairo-boilerplate-xcb.c:621)
by 0x12906F: cairo_test_for_target (cairo-test.c:819)
by 0x12AAB5: _cairo_test_context_run_for_target (cairo-test.c:1555)
by 0x126921: _cairo_test_runner_draw (cairo-test-runner.c:250)
by 0x126921: main (cairo-test-runner.c:932)
Signed-off-by: Uli Schlachter <psychon@znc.in>
Before this commit, running the test suite against a non-existing
display under valgrind resulted in:
$ ( cd test; DISPLAY=:2 CAIRO_TEST_TARGET=xcb valgrind --leak-check=full .libs/cairo-test-suite -f random-clip )
[...]
==47359== 64 bytes in 2 blocks are definitely lost in loss record 1 of 7
==47359== at 0x483AB65: calloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==47359== by 0x18A272: cairo_boilerplate_xcalloc (cairo-boilerplate-system.c:65)
==47359== by 0x18D652: _cairo_boilerplate_xcb_create_surface (cairo-boilerplate-xcb.c:269)
==47359== by 0x12906F: cairo_test_for_target (cairo-test.c:819)
==47359== by 0x12AAB5: _cairo_test_context_run_for_target (cairo-test.c:1555)
==47359== by 0x126921: _cairo_test_runner_draw (cairo-test-runner.c:250)
==47359== by 0x126921: main (cairo-test-runner.c:932)
This commit fixes that by freeing the allocated memory in the error
path.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Fix how offset, scale and transparency are handled.
Also do the same change in the "win32-printing" backend as it has a copy of the code from PDS, PS and SVG backends.
It was originally added to make bisecting easier,
but has outlived its usefuleness now.
Going forward we'll have just a single cairo-version.h
header file, the one with the real version numbers.
This is needed to fix the case where cairo is being
built as a Meson subproject, but also simplifies
things in general.
Fixes#421
This redesigns the path cache so that it does not mess with the
context functions, thereby hopefully making it much more resilient
to changes in the rest of cairo that change the way the default
context works. It is also much simpler, and it is anticipated that
it will be more maintainable. Performance in contrast to the old
cache design speeds up most traces in cairo-perf-trace, and slows
down only a lesser few by <20%.
Signed-off-by: George Matsumura <gmmatsumura01@bvsd.org>
Prior to this change, cogl often queried GLX so that it returned
a framebuffer that could not support an alpha component.
Signed-off-by: George Matsumura <gmmatsumura01@bvsd.org>
In order to be more consistent with backends such as gl, support for
creating a surface from content, width, and height parameters was
moved into the backend itself. The option to pass cairo-cogl a
framebuffer to create a texture from still exists, just now it is
not the only option.
Signed-off-by: George Matsumura <gmmatsumura01@bvsd.org>
Prior to this change, the boilerplate code crashed when given surface
dimensions less than 1. This fixes such behaviour by rounding the
dimensions up to 1, which is also done by the boilerplate code for
several other backends.
Signed-off-by: George Matsumura <gmmatsumura01@bvsd.org>
The buffer stack was moved from the surface to the device, as
mapped buffers are shared across all surfaces using the device
and more than one surface may be doing so. Stemming from this, if
a surface tries to map a buffer without first unmapping one that
has been mapped by another surface, it will trigger an error.
The parent backend functions were moved from the device to the
context, as it is possible that the context creation function could
be passed a non-cogl device. Prior to this change, many of the
subsurface tests segfaulted.
Signed-off-by: George Matsumura <gmmatsumura01@bvsd.org>
With the new cogl API, casting the framebuffer into offscreen
and onscreen types with the provided macros is a deprecated
behavior.
Signed-off-by: George Matsumura <gmmatsumura01@bvsd.org>
Without this, a buffer-swapping function only usable on onscreen
framebuffers would have been used on offscreen framebuffers.
Signed-off-by: George Matsumura <gmmatsumura01@bvsd.org>
