lockdep is a valgrind skin which performs pthread locking correctness
validation. In particular it allows one to write assert(HOLDS_LOCK(mutex))
which both documents the preconditions for a function and enforces them
when the program is run under lockdep.
As an aide to lockdep (as it works by intercepting the pthread functions),
all the mutexes should be initialised and destroyed using
pthread_mutex_init() and pthread_mutex_destroy() rather than using static
initializers and no-ops.
Add the core support to cairo-test for running the test-suite under a
malloc fault injector. This commit contains the adjustments to
cairo_test_run() to repeat the test if it detects a failure due to fault
injection and complains if it detects unreported faults or memory leaks.
Currently we only have an example of how to call FT_Done after the last
reference to the FT cairo_font_face_t has been dropped, but do not
actually explain why this needs to be done. So add a couple of sentences
to clarify the likely lifetime of the FT_Face once exposed to cairo.
Test the decomposition of 5 different types of spline curve. Useful test
for future experiments in improving the decomposition algorithm.
Note: the vector targets all need separate reference images due to their
lack of support for cairo_set_tolerance(). Also GS strokes the Bezier
curve differently using offset curves and opposed to transcribing the
outline of a pen.
Instead of fixing the check in d36b02dc66, just
remove it. Conditionally compilation will keep cairo-quartz-image-surface.c
from being compiled in when it is not configured to be. Suggested by Behdad.
glitz currently has no mechanism to release resources upon connection
closure (ala XCloseDisplay) and so will attempt invalid accesses when
evicting old surfaces from the solid pattern cache.
POSIX allows for any of its functions to be implemented as a macro,
inflicting damage upon innocent function pointers that happen to share
the same name.
face->stream->read() is one such example.
As it's outside of our control (being a FT_Face) we cannot simply rename
the data member, so we need to explicitly dereference it as a function
pointer in order to disambiguate it for the compiler.
Bug 9102 cairo doesn't support 24 bits per pixel mode on X11
(https://bugs.freedesktop.org/show_bug.cgi?id=9102)
is a reminder that that we need to support many obscure XImage formats.
With Carl's and Behdad's work to support psuedocolor we have a mechanism
in place to handle any format that is not natively handled by pixman. The
only piece we were missing was extending the swapper to handle all-known
formats and putting in defensive checks that pixels were correctly aligned
in accordance with pixman's requirements.
Emmanuel Pacaud did some research into adding DOCTYPE after we had a
complaint that our SVG output failed to be validated by batik. The
conclusion he came to was that the validation of SVG was incomplete and
misleading, i.e. our output might generate false negatives leading to more
confusion.
Adrian Johnson proposed a hack that fixed the current test by simply
scaling the dashed stroke by device_transform.xx. Obviously this exposes a
deficiency in the test case. So add an outer loop to iterate over several
common ppi for both x and y and in doing so reveal more ugly artifacts in
the current fallback code.
func_regexp was incorrectly complaining that cairo_pure was a function without
parenthesis. The simplest solution appeared to be allow quoting of
'cairo_pure'.
As is so often the case, reading the commit log gives you fresh insight in
the problem - often called confessional debugging...
We can simplify the problem by ignoring attr->[xy]_offset, for the time
being, and focus on computing the correct matrix. This is comparatively
simple as all we need to do is perform the appropriate rounding on the
translation vector.
A complication I realised after pushing 3eb4bc3 was handling larger
sampled areas. Extending the test case revealed that the optimization
was broken for anything but the identity transform (after removing the
translation). Correctness first, leaving the "pixel-exact" solution for
interested reader...
Carl suggested that cairo_pure and cairo_const are pretty opaque, even to
the developer who added them, so it is extremely important that they have
a good description so that they are used correctly and perhaps ported to
other compilers.
With the addition of cairo_show_text_glyphs() came a couple of functions
to query whether the target supported the extended attributes. However,
at Carl's request cairo_has_show_text_glyphs() was removed - but the
documentation was not updated to reflect that.
As identified in bug 15479,
Unpredictable performance of cairo-xlib with non-integer translations of a
source surface pattern
(https://bugs.freedesktop.org/show_bug.cgi?id=15479),
source surfaces with a fractional translation hit slow paths for some
drivers, causing seemingly random performance variations. As a work-around
Owen Taylor proposed that cairo could convert non-integer translations on
NEAREST sources patterns to their integer equivalents.
The messy detail involved here is replicating the rounding mode used by
pixman for the sample offset, but otherwise the conversion is fairly
trivial.
As proof-of-principle, compute a scale factor to avoid overflow when
converting a linear pattern to pixman_fixed_t. Fixes test/huge-pattern,
but the principle should be extended to handle more cases of overflow.
Since CAN_TEST_PS_SURFACE does not currently require spectre, we were
attempting to compile in spectre support for any2ppm even on systems
without libspectre installed. Fix that by adding a separate flag for
CAIRO_HAS_SPECTRE.
From bug 18010 (https://bugs.freedesktop.org/show_bug.cgi?id=18010),
in order to make flockfile() available we need to set _POSIX_C_SOURCE and
according to the man page, the appropriate feature check is
_POSIX_THREAD_SAFE_FUNCTIONS.
Bulia Byak reported a bug where cairo was crashing with a particular
font. The font had an incorrect entry in the cmap table that caused
cairo to read from outside of the buffer allocated for the cmap.
Mention that building the configure script requires at least version 0.16
of pkg-config.
See the old bug:
Bug 4702 PKG_PROG_PKG_CONFIG: command not found
(https://bugs.freedesktop.org/show_bug.cgi?id=4702)
Bug 11734:
XRender crashes due to NULL pointer from Cairo on SGI O2
(https://bugs.freedesktop.org/show_bug.cgi?id=1173)
is an example of a case where we try to perform an XRender operation on a
surface with a format that was not supported by the XRender extension. By
marking the extension version as -1 on those surfaces, the current checks
for SURFACE_RENDER_HAS_* always correctly return false and prevent us try
to create a Picture with a NULL xrender_format.
Add a new test case to Cairo for checking the performance of Cairo's
equivalent to GDK's gdk_pixbuf_composite_color() operation. That is an
operation that happens to be extremely useful when viewing or editing
transparent images so I think it is important that it is as fast as
possible.
Add a test case to capture the current behaviour when a segment ends on
an off/on dash transition.
Originally filed as bug:
Miter artifacts for dashed strokes
https://bugs.freedesktop.org/show_bug.cgi?id=17973
Since cairo.h is a public header file, we need to be careful so that it
can be compiled by random compilers and even users specifying "-Werror
-Wundef" (mentioning no names, Company). So replace the bare (and legal)
#if _MSC_VER
with
#if defined (_MSC_VER)
just in case.
Missed updating win32 when implementing the new paginated API to query
support for the fine-grained fallbacks.
(Thanks to Adrian for spotting this oversight.)
The use of fine-grained fallbacks requires the native support of the
SOURCE operator applied to an image on the target surface. SVG 1.2
introduces the "comp-op:src" mode fulfilling this criteria - so we can
enable fine-grained fallbacks for 1.2+.
Update test/fine-grained-fallbacks to exercise this pathway in SVG 1.2 -
as SVG natively supported all the current operations within that test.
This reveals yet another librsvg bug in handling SVG 1.2.
Avoid passing a 32bit integer as a cairo_int64_t in case we do not have a
64bit native integral type. As a side-effect this means we can also use a
narrower multiply.
