Vladimir Vukicevic
8ad8aa6360
Use XCopyArea when possible, for optimization and bug workaround.
...
Exercise XCopyArea, XRenderComposite, and XSetTile/XFillRectangle paths for _cairo_xlib_surface_composite
2005-07-15 13:45:19 +00:00
Carl Worth
2534057cb5
Add new check-valgrind target for running the test suite under the influence of valgrind.
...
Fix memory leak when reference image is not found.
2005-07-15 10:39:59 +00:00
Carl Worth
07d09acf2a
Fix a few memory leaks (missing cairo_pattern_destroy in 3 places).
2005-07-15 01:38:38 +00:00
Carl Worth
3a053c1e58
Fix memory leak, (missing _cairo_pattern_fini).
2005-07-15 01:05:56 +00:00
Carl Worth
89a9771b60
Remove errant line of code that was inadvertently committed recently.
2005-07-15 00:56:58 +00:00
Carl Worth
3e05319bf8
Fix leak of surface.
2005-07-14 17:55:58 +00:00
Carl Worth
6df1baa395
Call _cairo_array_fini on the several array objects to patch memory leaks.
2005-07-14 17:52:17 +00:00
Carl Worth
df181620c0
Eek. Yet _more_ fixups for _cairo_pattern_release_surface missed earlier.
2005-07-14 16:56:08 +00:00
Stuart Parmenter
174056f736
blue and green values were swapped in the DO_SOURCE case.
2005-07-14 16:50:58 +00:00
Carl Worth
40edca337e
Rewrite to use cairo_image_surface_create_from_png rather than custom read_png_argb32. In addition to being simpler, this eliminates the leak of the image data buffer.
...
Add calls to cairo_pattern_destroy to close two memory leaks.
2005-07-14 16:18:39 +00:00
Carl Worth
d8500060a9
Fix up one more call to _cairo_pattern_release_surface missed in the previous commit.
2005-07-14 15:47:18 +00:00
Carl Worth
3cd006bb2b
Fix to accept a cairo_pattern_t rather than a cairo_surface_t as the primary argument.
...
Track change in _cairo_pattern_release_surface and also pass the appropriate pattern for each acquired surface. The previous backend mismatch was causing memory leaks.
Remove stale comment.
Add missing fclose to keep valgrind happy about memory leaks.
2005-07-14 15:10:47 +00:00
Carl Worth
288f0f7f49
Make the per-backend test results print XFAIL rather than FAIL for expected failures.
2005-07-14 12:50:28 +00:00
Carl Worth
6f6c9196c8
Fix comment describing test result. It's not enough that all tested backends are successful. We also require that at least one backend is actually tested.
2005-07-14 12:25:41 +00:00
Carl Worth
456e3ffc4d
Don't consider a test to fail if it can't create a surface at all (eg. no X server is available). Instead mark this backend as untested and only consider the overall test a success if all tested backend are successful.
2005-07-14 12:20:42 +00:00
Carl Worth
0c3ee348f8
Add missing (putatively unreachable) return value.
2005-07-14 11:48:40 +00:00
Carl Worth
14ea7b583b
Remove the copying of the region. This was a workaround for a missing copy bug in libpixman that has since been fixed. So now it was just a memory leak.
2005-07-14 11:45:39 +00:00
Carl Worth
171b053347
Whitespace fix. (pixman_image_destroy): Fix leak of the clip region associated with an image. (NOTE: xserver/render doesn't have the same bug since it has FreePicture that calls into both DestroyPicture and DestroyPictureClip separately).
2005-07-14 11:41:00 +00:00
Carl Worth
513e9e6fe5
Patch memory leak of png_info object.
2005-07-14 11:24:56 +00:00
Carl Worth
574b702029
Rename create-for-png test to create-from-png to match the naming of the cairo_image_surface_create_from_png function.
2005-07-14 11:17:12 +00:00
Carl Worth
c41e9d698f
Patch a few memory leaks.
2005-07-14 11:11:15 +00:00
Kristian Høgsberg
3a46944637
Use the _cairo_surface_*() functions when replaying.
...
Fold the "locate fallbacks" pass into the postscript output pass, and add a simple, first implementation of image fallbacks.
2005-07-13 12:32:51 +00:00
Carl Worth
aafc2e7497
Rename CAIRO_FONT_BACKEND_DEFAULT to the more accurate CAIRO_SCALED_FONT_BACKEND_DEFAULT.
2005-07-13 11:34:37 +00:00
Carl Worth
6725cc9d7f
New predicates to allow checking for cairo_ft derivates of generic font type.
...
Add explicit checks for cairo_ft derivatives of generic fonts rather than just blindly assuming that's what we get.
2005-07-13 11:01:25 +00:00
Carl Worth
dda555de99
Remove destroy notifier. This simplifies the implementation a bit, and no anticipated use of cairo_hash_table_t in cairo needs the destroy notifier. Most uses will be hash-backed object create/destroy functions.
...
(_cairo_hash_table_destroy): Document that it is now a fatal error to call _cairo_hash_table_destroy on a non-empty hash table.
(_cairo_hash_table_insert): Document that it is now a fatal error to insert an entry with a key that matches an existing entry.
(_cairo_hash_table_random_entry): Add predicate function so that the user can select a random entry satisying the given predicate.
(_cairo_hash_table_remove): Change return type to void since failure is really not possible here.
2005-07-12 14:43:37 +00:00
Carl Worth
efb413c084
Add note that glyph measurement performance needs to improve.
2005-07-11 16:24:19 +00:00
Carl Worth
4329b70963
Fix typo (cairo_ft_options_t -> cairo_font_options_t).
2005-07-11 16:09:55 +00:00
Carl Worth
abdaf7555f
Add cache lock deadlock problem to roadmap for 0.5.2.
...
Disable mutex locks, (making the caches non-thread-safe again, just like they were in the last snapshot and before).
2005-07-11 16:07:08 +00:00
Carl Worth
e9d42ab9f9
ROADMAP: Add recent API change proposals (cairo_ft_options_t, cairo_xlib_surface_create needs to be screen aware, and cairo_xlib_surface_set_drawable).
2005-07-11 14:03:51 +00:00
Carl Worth
9cddbaf8e6
ROADMAP: Note progress on cairo_meta_surface_t, ARGB text, and group support.
2005-07-11 13:37:43 +00:00
Carl Worth
48e5806f1e
Note that the cairo_content_t work is done.
...
Note that this function is computing an incorrect result.
2005-07-11 13:29:45 +00:00
Jeff Muizelaar
2ff04df852
Begin merging in newer fb code.
...
The code comes from keithp: Initial import of Composite extension along with name change from Apportion. Added some accelerated code to fbpict to make this initial hack a bit faster. "real" extension bits to follow.
2005-07-09 19:00:36 +00:00
Jeff Muizelaar
1d48dc6cdb
More misc merging. Mostly s/pixman_composite/fbComposite/.
2005-07-09 18:13:14 +00:00
Jeff Muizelaar
36a57bcdcb
More misc merging. Mostly stuff like s/uint32_t/CARD32/.
2005-07-09 15:47:43 +00:00
Carl Worth
9a2ba48b29
Give enum tags an underscore prefix to match the style of the struct tags. Add new cairo_content_t and change cairo_surface_create_similar to accept a cairo_content_t rather than a cairo_format_t.
...
Change surface backend create_similar call to accept a cairo_content_t rather than a cairo_format_t.
Fix all calls into create_similar to pass a cairo_content_t rather than a cairo_format_t.
2005-07-08 10:12:28 +00:00
Carl Worth
9fd2f0a25c
Add a missing word.
...
Note that the BadMatch bug has been resolved.
2005-07-07 01:15:21 +00:00
Carl Worth
f766e06d25
Remove CAIRO_STATUS_NO_TARGET_SURFAC and add CAIRO_STATUS_INVALID_STATUS.
...
Don't check for gstate->target == NULL anymore as the API now guarantees it never occurs.
Check that path->status is a valid status value and cause an INVALID_STATUS error otherwise.
Test the new CAIRO_STATUS_INVALID_STATUS error case in cairo_append_path.
2005-07-06 14:52:01 +00:00
Carl Worth
eb12e49308
Require libpixman >= 0.1.5, (since 0.1.4 crashes on some X servers).
2005-07-06 11:23:39 +00:00
Carl Worth
ad0e040fb6
Disable the compilation of the _ps_locate_fallbacks code which, inaddition to being broken, appears to not currently be used.
2005-07-05 22:30:59 +00:00
David Reveman
e314e86cc1
Use frame buffer objects for accelerated offscreen drawing in glitz backend
2005-07-04 03:58:43 +00:00
Kristian Høgsberg
7a923e6ddd
Rewrite postscript backend to generate more interesting output than the current big-image implementation, using meta surfaces for font subsetting and image fallbacks.
...
Remove obsolete comment.
Make a couple of stylistic changes and add _cairo_output_stream_write_hex_string.
Add _cairo_surface_intersect_clip_path so we can replay path clipping.
2005-07-01 12:45:35 +00:00
Kristian Høgsberg
7ab5dbfa75
Add meta surface implementation.
...
Add cairo-meta-surface.c and cairo-meta-surface-private.h
2005-07-01 12:22:23 +00:00
Carl Worth
4e01c4aa2f
Remove debugging printfs.
2005-06-29 17:05:31 +00:00
Carl Worth
aad6dce6bc
Add _cairo_hash_table_random_entry.
...
Fix to update live_entries.
style changes.
Add code to shrink table as well as to grow it.
Call new version of resize so that table will grow or shrink as needed on insert and remove.
2005-06-29 17:02:37 +00:00
Carl Worth
e7ffbb4179
Rewrite hash table to use a single cairo_hash_entry_t* rather than void *key and void *value. This is slightly more painful to use, but lends itself to a more memory-efficient implementation. Add documentation.
2005-06-29 15:02:10 +00:00
Carl Worth
82f244eca8
Rework the cache code as a hast table with a much simpler interface, (no object derviation is required to use it).
...
Remove extraneous prototype for non-existent _cairo_cache_reference.
2005-06-29 07:04:34 +00:00
Kristian Høgsberg
5ab8a3085e
New functions to create a cairo_pattern_t corresponding to a solid color and a translucent color respectively. Document a few pattern functions.
2005-06-28 15:58:42 +00:00
Tor Lillqvist
e7442a4523
Implement.
...
reviewed by: otaylor
2005-06-28 15:03:53 +00:00
Owen Taylor
6d65ee6fc4
Also flag older XFree86 servers as buggy-repeat.
2005-06-28 13:28:53 +00:00
Tor Lillqvist
e0982300ca
Prevent crash on empty string.
2005-06-28 11:52:42 +00:00
