As Behdad suggested, we can dramatically speed up the test suite by
short-circuiting the write to a png file, only to then immediately read it
back in. So for the raster based surfaces, we avoid the round-trip through
libpng by implementing a new boilerplate method to directly extract the image
buffer from the test result. A secondary speedup is achieved by caching the
most recent reference image.
In order to run under memfault, the framework is first extended to handle
running concurrent tests - i.e. multi-threading. (Not that this is a
requirement for memfault, instead it shares a common goal of storing
per-test data). To that end all the global data is moved into a per-test
context and the targets are adjusted to avoid overlap on shared, global
resources (such as output files and frame buffers). In order to preserve
the simplicity of the standard draw routines, the context is not passed
explicitly as a parameter to the routines, but is instead attached to the
cairo_t via the user_data.
For the masochist, to enable the tests to be run across multiple threads
simply set the environment variable CAIRO_TEST_NUM_THREADS to the desired
number.
In the long run, we can hope the need for memfault (runtime testing of
error paths) will be mitigated by static analysis. A promising candidate
for this task would appear to be http://hal.cs.berkeley.edu/cil/.
This is a slightly kinder interface that accepts cairo_image_surface_t
pointers rather than pointers to the raw image data and width, height,
stride. This brings us closer to hooking up the pdiff code.
In a manner similar to flattening in advance, we now extract the sub-
surface of interest (when testing with device offsets) before calling
into the buffer_diff functions. This allows these functions to accept
a single stride value once again instead of one for each of the three
images.
The interface of the various buffer/image_diff functions is improved to
provide the maximum pixel difference in addition to the number of pixels
that differ. This value can then be used to compare against a per-backend
tolerance.
Currently I've set the SVG backend's tolerance to 1 to handle some issues
we're currently seeing of single-bit differences on different systems, (but
we're not exactly sure why yet).
Also I improved the image_diff routines to properly report a status value
on failure rather than the bogus value of -1 for pixels_changed.
Previously it was using the equation 128+diff/3, which results in
lots of gray and de-emphasized difference. Now it's using
MIN(255,diff*4) which more emphasizes the real difference.
This patch was produced by running git-stripspace on all *.[ch] files
within cairo. Note that this script would have also created all the changes
from the previous commits to remove trailing whitespace.
This patch was produced with the following (GNU) sed script:
sed -i -r -e '/^[ \t]*\/?\*/ s/[ \t]+$//'
run on all *.[ch] files within cairo, (though I manually excluded
src/cairo-atsui-font.c which has a code line that appears as a comment
to this script).
This is a mega-patch that has the advantage that the entire test suite
passes both immediately before and immediately after this commit.
The disadvantage of the mega-patch is that it does not reflect the
development history of the device-offset branch, (with its various
fumblings and flailings). To capture that history, we will next merge
in that branch.
[With this change, two tests with gradients exhibit subtle differences
under device offsets. I don't think we actually care about them
though. -cworth]
The PS/PDF backends don't allow a content to be passed in right now, so they fail against the rgb24 tests, but the trivial addition to the constructors will allow them to pass all tests with both content values.
And new constructors (currently internal only) to create an image surface with a cairo_content_t rather than a cairo_format_t.
Add a cairo_content_t argument to the constructor.
Add a cairo_content_t to the constructor and use this content value when constructing intermediate image surfaces in acquire_source, show_page, copy_page, and snapshot.
Add image flattening by compositing over white, as is done in cairo-ps-surface.c.
Track changes to cairo-paginates-surface which now requires a cairo_content_t value (no change to public PS/PDF constructors yet).
Track change in meta-surface and paginated-surface interfaces by now accepting a cairo_content_t rather than a cairo_format_t.
Ignore new output files (argb32 from pdf and ps as well as rgb24 from test-fallback, test-meta, and test-paginated).
Add new utility for flattening PNG images in order to generate the -argbf-ref.png images.
Add image_diff_flattened for comparing flattened output from PS and PDF backend with ARGB reference images by first blending the reference images over white.
Get rid of conditional, format-specific background-color initialization before running tests. Now uses ARGB(0,0,0,0) in all cases. Switch from specifying tests with a format value to specifying tests with a content value. Add support for a 'fake' COLOR_ALPHA_FLATTENED content for testing the PS and PDF output against a flattened version of the argb32 reference images (first blended over white).
Track change in cairo_ps_surface_create (now requires cairo_content_t value).
Adjust tests that draw in default (black) to first paint white so that the results are visible.
Adjust ARGB32 reference images for new white background for changed tests.
Adjust RGB24 reference images for new black background due to changed initialization (and the tests themselves being unchanged).
Define snprintf as _snprintf when under the influence of _MSC_VER.
Define int32_t and friends as __int32 and friends when under the influence of _MSC_VER.
Make include of unistd.h conditional on HAVE_UNISTD_H.
Propagate the unsigned char* change down the stack.
Add cast since XImage uses char* rather than unsigned char*.
Fix memory leak of image data.
Switch to use cairo_surface_write_png rather than a custom write_png_argb32.
Add test to exercise the cairo_image_surface_create_for_png function.
Add new testsvg script and accompanying imagediff program, (for interim SVG-based test suites while we wait for the standard cairo test suite to mature).
Split buffer_diff out into its own file for the purpose of imagediff.
