1 Release 1.12.2 (2012-04-29 Chris Wilson <chris@chris-wilson.co.uk>)
2 ===================================================================
3 After such a long gestation period for the release of Cairo 1.12, we
4 inevitably accumulated a few bugs that were flushed out by broadening the
5 test base. Thanks to everybody who tried the release, apologies to any one
6 unfortunate enough to encounter a bug and many thanks for reporting it. As
7 a result Adrian Johnson, Alexandros Frantzis, Andrea Canciani, Kalev
8 Lember, Maarten Bosman, Marcus Meissner, Nis Martensen and Uli Schlachter
9 have squashed many more bugs and improved the documentation. I would
10 strongly recommend everyone to upgrade to cairo-1.12.2.
16 Allow applications to create 0x0 xlib surfaces, such as used by LibreOffice.
17 https://bugs.freedesktop.org/show_bug.cgi?id=49118
19 Trim composite extents for SOURCE/CLEAR operators to the mask.
21 Use fallback fonts in PDF for unhandled computed glyph widths
22 https://bugs.freedesktop.org/show_bug.cgi?id=48349
24 Handle snapshots of recording surfaces for analysing pattern extents.
25 Fixes a regression of reporting the PDF bounding box as being the page size.
27 Fix allocation size for PDF pattern ids.
28 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=49089
30 Fix emission of rectilinear dashed segments, with and without scaling, and
31 application of degenerate line joins.
33 Clamp unbounded fixup polygons to the clip extents.
35 Prevent infinite loop due to rounding errors whilst incrementing along dashes.
37 Prevent overflow for inline a8 span filling.
39 Miscellaneous build fixes for Cygwin on Windows and Solaris.
41 Release 1.12.0 (2012-03-23 Chris Wilson <chris@chris-wilson.co.uk>)
42 ===================================================================
43 It's taken over 18 months, but the wait is finally over. A new cairo release!
44 We are pleased to annouce a new stable release of Cairo that brings many
45 new features and performance improvements, all whilst maintaining
46 compatibility with cairo-1.0 and all releases since. We recommend anyone
47 using a previous release of Cairo to upgrade to 1.12.0.
49 The major feature of this release is the introduction of a new procedural
50 pattern; the mesh gradient. This, albeit complex, gradient is constructed
51 from a set of cubic Bezier patches and is a superset of all other gradient
52 surfaces which allows for the construction of incredibily detailed patterns.
53 In PDF parlance, the mesh gradient corresponds with type 7 patterns. Many
54 thanks to Andrea Canciani for bringing this to Cairo, and for his work on
55 making gradient handling robust.
57 Not content with just adding another procedural pattern, Cairo 1.12 also
58 adds new API to create a callback pattern,
59 cairo_pattern_create_raster_source, that allows the application to
60 provide the pixel data for the region of interest at the time of
61 rendering. This can be used for instance, by an application to decode
62 compressed images on demand and to keep a cache of those decompressed
63 images, independently of Cairo. When combined with the recording
64 surface, it should form a useful basis for a deferred renderer.
66 With the release of cairo-1.12, we also introduce a new supported
67 backend for interoperating with X using XCB. Uli Schlachter, also
68 maintainer of awesome and contributor to libxcb, has volunteered to
69 maintain cairo-xcb for us. Thanks Uli!
71 For cairo-1.12, we have also added some common API to address any
72 surface as an image and so allow direct modification of the raster data.
73 Previously, only the Quartz and Win32 backends supported a very narrow
74 interface to allow for efficient pixel upload. Now with
75 cairo_surface_create_similar_image, cairo_surface_map_to_image, and
76 cairo_surface_unmap_image, Cairo exports a consistent method for
77 treating those surfaces as an image and so allow modification inplace.
78 These are the same routines used internally, and should support
79 efficient transfer or direct mapping of the target surfaces as
82 Another focus over the past year has been to address many performance
83 issues, without sacrificing the composition model. To accomplish the
84 goal, once again the rasterisation pipeline was overhauled and made
85 explicit, giving the backends the freedom to implement their own
86 specific pipeline whilst also providing a library of common routines
87 from which to build the pipeline. For instance, this allows the image
88 backend and the gl backend to composite scan line primitives inplace,
89 and to then implement custom fallbacks to catch the corner cases that do
90 not map onto their fastest paths. Similarly, this allows for the Xlib
91 backend to implement trapezoidation without compromising the other
92 backends, yet still allow for the pipeline to be used elsewhere for
93 testing and fallbacks. Clipping was once again overhauled, so that the
94 common cases for the raster pipelines could be captured and processed
95 with fast paths with the emphasis on performing geometric clipping to
96 reduce the frequency of using multi-pass clipmasks. Stroking was made
97 faster, both by providing specialised fast-paths for simple, yet frequent,
98 cases (such as stroking around a rectangle) and by reducing the number
99 of edges generated by the general stroker.
101 As part of the focus on performance, Cairo 1.12 introduces some
102 antialias hints (NONE,FAST, GOOD, BEST) that are interpolated by the
103 raserisers to fine tune their performance versus quality. Cairo 1.12
104 also introduces a new observation architecture,
105 cairo_surface_observer_t, which can be used to analyse the amount of
106 time consumed by drawing commands and help identify inefficiencies in
107 both Cairo and the application.
109 Last, but by no means least, the OpenGL backend has seen significant
110 work including the port to GLESv2 and the exploitation of advanced
111 hardware features. Interesting times.
113 As always, I would like to thank everyone who contributed to Cairo,
114 not only through writing code, but also submitting documentation, bug
115 reports, suggestions and generally having fun with Cairo! In particular
116 though this release could not have happened without the efforts of
117 Adrian Johnson, Alexandros Frantiz, Andrea Canicani, Martin Robinson,
118 Nis Martensen, and Uli Schlachter. Thanks.
121 Snapshot 1.11.4 (2012-13-12)
122 ============================
123 The cairo community is pleased to finally announce the long aniticpated
124 release candidate for 1.12, 1.11.4, of the cairo graphics library. This
125 is the first major update to cairo in over a year and brings a large
126 number of new features; undoubtably a few bugs as well.
128 While many people have contributed and have helped to test the release,
129 providing feedback on 1.10 and suggesting improvements, this release
130 is the result of a few persevering souls who deserve recognition for their
131 outstanding contributions: Andrea Canciani (all round bug fixing,
132 performance tuning and master of the gradients), Adrian Johnson (PDF
133 supremo) and Uli Schlachter (who stepped forward as maintainer for the
136 Major additions since 1.11.2:
138 * cairo_surface_map_to_image API for pixel level access to any surface
140 * New antialias hints to control the trade-off between speed and quality
142 * A callback pattern, cairo_pattern_create_raster_source, for lazy
143 decoding of image data.
145 * cairo_surface_observer_t, a new type of surface to gather performance
148 * XCB as a supported backend
150 * A rewritten compositor pipeline for performance improvements for, but not
151 limited to, the xlib and image backends.
152 From ION and PineView through to SandyBridge, every machine I have shows
153 across the board performance improvement on the cairo-traces:
155 i5-2520m gnome-system-monitor: 5.97x speedup
156 pnv gnome-system-monitor: 4.86x speedup
157 i5-2520m firefox-asteroids: 4.66x speedup
158 pnv firefox-asteroids: 4.43x speedup
159 image firefox-canvas: 3.82x speedup
160 i5-2520m firefox-canvas-alpha: 3.49x speedup
161 image firefox-asteroids: 2.87x speedup
162 pnv firefox-talos-svg: 2.83x speedup
163 ion grads-heat-map: 2.75x speedup
164 pnv firefox-canvas-alpha: 2.66x speedup
165 image gnome-system-monitor: 2.66x speedup
166 image swfdec-giant-steps: 2.46x speedup
167 image firefox-canvas-alpha: 2.14x speedup
168 i5-2520m firefox-talos-svg: 2.03x speedup
169 image grads-heat-map: 2.02x speedup
170 ion gnome-system-monitor: 2.00x speedup
171 pnv firefox-particles: 1.99x speedup
172 i5-2520m grads-heat-map: 1.96x speedup
173 pnv firefox-canvas: 1.92x speedup
174 ion firefox-particles: 1.80x speedup
175 image poppler-reseau: 1.77x speedup
176 pnv xfce4-terminal-a1: 1.72x speedup
177 image firefox-talos-svg: 1.65x speedup
178 pnv grads-heat-map: 1.63x speedup
179 i5-2520m firefox-canvas: 1.63x speedup
180 pnv swfdec-youtube: 1.62x speedup
181 image ocitysmap: 1.59x speedup
182 i5-2520m firefox-fishbowl: 1.56x speedup
183 i5-2520m poppler-reseau: 1.50x speedup
184 i5-2520m evolution: 1.50x speedup
185 i5-2520m midori-zoomed: 1.43x speedup
186 pnv firefox-planet-gnome: 1.42x speedup
187 i5-2520m firefox-talos-gfx: 1.41x speedup
188 i5-2520m gvim: 1.41x speedup
189 pnv ocitysmap: 1.37x speedup
190 image poppler: 1.31x speedup
191 ion firefox-canvas-alpha: 1.35x speedup
192 ion firefox-talos-svg: 1.34x speedup
193 i5-2520m ocitysmap: 1.32x speedup
194 pnv poppler-reseau: 1.31x speedup
195 i5-2520m firefox-planet-gnome: 1.31x speedup
196 pnv firefox-fishbowl: 1.30x speedup
197 pnv evolution: 1.28x speedup
198 image gvim: 1.27x speedup
199 i5-2520m swfdec-youtube: 1.25x speedup
200 pnv gnome-terminal-vim: 1.27x speedup
201 pnv gvim: 1.25x speedup
202 image firefox-planet-gnome: 1.25x speedup
203 image swfdec-youtube: 1.25x speedup
206 And a plethora of minor improvements everywhere!
209 Snapshot 1.11.2 (2011-01-23)
210 ===========================
212 In this first snapshot along the way to cairo-1.12.0, we are very excited
213 to announce the introduction of Bezier surface gradients, known as type
214 6/7 gradients in PS/PDF parlance. This is the culmination of much work by
215 the dynamic duo: Adrian Johnson and Andrea Canciani. Thanks guys!
217 Also, I want to warmly welcome Uli Schlachter who recently joined the
218 Cairo community on a mission. That mission is to make cairo-xcb a
219 supported backend for 1.12. And for this snapshot he has made great
220 strides in fixing all the bugs I had left behind. Thanks Uli!
222 And we have also seen a new contributor, Alexandros Frantzis, who has
223 begun bringing up cairo-gl for GLESv2 devices. Thanks Alex!
225 And lastly, I must also thank Adrian and Andrea for the vast numbers of
226 bugs that they have tackled between them, fixing all those little corner
227 cases that lie hidden until too late.
231 The ability to construct piece-wise Bezier surface gradients:
233 cairo_pattern_create_mesh
235 constructs a pattern of type CAIRO_PATTERN_TYPE_MESH using
237 cairo_pattern_mesh_begin_patch
238 cairo_pattern_mesh_end_patch
239 cairo_pattern_mesh_curve_to
240 cairo_pattern_mesh_line_to
241 cairo_pattern_mesh_move_to
242 cairo_pattern_mesh_set_control_point
243 cairo_pattern_mesh_set_corner_color_rgb
244 cairo_pattern_mesh_set_corner_color_rgba
245 cairo_pattern_mesh_get_patch_count
246 cairo_pattern_mesh_get_path
247 cairo_pattern_mesh_get_corner_color_rgba
248 cairo_pattern_mesh_get_control_point
250 The introduction of a unique ID accessible via the mime data type:
251 CAIRO_MIME_TYPE_UNIQUE_ID
257 Release 1.10.2 (2010-12-25 Chris Wilson <chris@chris-wilson.co.uk>)
258 ===================================================================
259 The cairo community is pleased to announce the 1.10.2 release of the
260 cairo graphics library. This is the first update to cairo's stable 1.10
261 series and contains a large number of bug fixes.
263 While many people have contributed and have help to test the release,
264 2 people deserve special recognition for their efforts in tracking down
265 and fixing bugs, Andrea Canciani and Adrian Johnson. Thanks to their
266 tremendous efforts, and of all cairo contributors, it is much
269 We recommend everyone upgrade to cairo 1.10.2 and hope that everyone
270 will continue to have lots of fun with cairo!
277 Fix embedding of grayscale jpegs in PS.
278 https://bugs.freedesktop.org/show_bug.cgi?id=31632
280 Fix the reported path of extents containing a curve.
282 Fix the compositing of unaligned boxes.
284 Reset the clipper in PDF upon finish.
286 Fix degenerates arcs to become a degenerate line.
288 Build support for autoconf 2.67
290 Fix painting of transformed patterns in PS
292 Fix the EPS bounding box for PS
293 https://bugs.freedesktop.org/show_bug.cgi?id=24688
295 Fix the missing content for EPS
296 https://bugs.freedesktop.org/show_bug.cgi?id=24688
298 Fix regression upon changing page size in PS/PDF
299 https://bugs.freedesktop.org/show_bug.cgi?id=24691
301 Only use ActualText with PDF-1.5 documents
303 Fix the bbox for type1 fallbacks.
305 Reset the color after ending the context in PDF
306 https://bugs.freedesktop.org/show_bug.cgi?id=31140
308 Fix the advance of subsetted type1 fonts
309 https://bugs.freedesktop.org/show_bug.cgi?id=31062
311 Fix handling of EXTEND_NONE gradients for PDF
313 Restrict in-place optimisation for a8 image masks with SOURCE
316 Release 1.10.0 (2010-09-06 Chris Wilson <chris@chris-wilson.co.uk>)
317 ===================================================================
318 The cairo community is astounded (and flabbergast) to finally announce
319 the 1.10.0 release of the cairo graphics library. This is a major update
320 to cairo, with new features and enhanced functionality which maintains
321 compatibility for applications written using any previous major cairo
322 release, (1.8, 1.6, 1.4, 1.2, or 1.0). We recommend that anybody using
323 a previous version of cairo upgrade to cairo 1.10.0.
325 One of the more interesting departures for cairo for this release is the
326 inclusion of a tracing utility, cairo-trace. cairo-trace generates a
327 human-readable, replayable, compact representation of the sequences of
328 drawing commands made by an application. This can be used to inspecting
329 applications to understand issues and as a means for profiling
330 real-world usage of cairo.
332 The traces generated by cairo-trace have been collected in
334 git://git.cairographics.org/git/cairo-traces
336 and have driven the performance tuning of cairo over the last couple of
337 years. In particular, the image backend is much faster with a new
338 polygon rasterisation and a complete overhaul of the tessellator. Not
339 only is this faster, but also eliminates visual artifacts from
340 self-intersecting strokes. Not only has cairo-trace been driving
341 performance improvements within cairo, but as a repeatable means of
342 driving complex graphics it has been used to tune OpenGL, DDX, and
345 Cairo's API has been extended to better support printing, notably
346 through the ability to include a single compressed representation of an
347 image for patterns used throughout a document, leading to dramatic file
348 size reductions. Also the meta-surface used to record the vector
349 commands compromising a drawing sequence is now exposed as a
350 CAIRO_SURFACE_TYPE_RECORDING, along with a new surface that is a child of a
351 larger surface, CAIRO_SURFACE_TYPE_SUBSURFACE. One typical usage of a
352 subsurface would be as a source glyph in a texture atlas, or as a
353 restricted subwindow within a canvas.
355 Cairo's API has also resurrected the RGB16 format from the past as
356 the prevalence of 16-bit framebuffers has not diminished and is a
357 fore-taste of the extended format support we anticipate in the future.
358 Increasing cairo's utility, we introduce the cairo_region_t for handling
359 sets of pixel aligned rectangles commonly used in graphics applications.
360 This is a merger of the GdkRegion and the pixman_region_t, hopefully
361 providing the utility of the former with the speed of the latter.
363 Furthermore cairo has been reworked to interoperate more closely with
364 various acceleration architectures, gaining the ability to share
365 those hardware resources through the new cairo_device_t. For instance,
366 with the new OpenGL backend that supersedes the Glitz backend, hardware
367 and rendering operations can be shared between a classic OpenGL
368 application mixing libVA for the hardware assisted video decode with
369 cairo for high quality overlays all within the same OpenGL canvas.
371 Many thanks for the hard work of Adrian Johnson, Andrea Canciani, Behdad
372 Esfahbod, Benjamin Otte, Carl Worth, Carlos Garcia Campos, Chris Wilson,
373 Eric Anholt, Jeff Muizelaar, Karl Tomlinson, M Joonas Pihlaja, Søren
374 Sandmann Pedersen and many others that have contributed over the last
375 couple of years to cairo. Thank you all!
377 Snapshot 1.9.14 (2010-07-26)
378 ============================
380 A quiet couple of weeks, hopefully Cairo is seeing widescale deployment and
381 we are being to see the results of the stabilisation effort. Clipping bugs
382 seems to have been the order of the last couple of weeks, with a couple
383 reported and duly fixed. Thank you Igor Nikitin and Karl Tomlinsion for
384 finding those regressions. At this point all that seems to remain to do is
385 to fix the outstanding regressions in the PDF backend...
390 Clip doesn't work for text on the image backend
391 https://bugs.freedesktop.org/show_bug.cgi?id=29008
393 Add explicit dependency for cxx
394 https://bugs.freedesktop.org/show_bug.cgi?id=29114
396 Fix regressions in reporting clip extents
397 https://bugs.freedesktop.org/show_bug.cgi?id=29120
398 https://bugs.freedesktop.org/show_bug.cgi?id=29121
399 https://bugs.freedesktop.org/show_bug.cgi?id=29122
400 https://bugs.freedesktop.org/show_bug.cgi?id=29124
401 https://bugs.freedesktop.org/show_bug.cgi?id=29125
404 Snapshot 1.9.12 (2010-07-12)
405 ============================
407 A couple of weeks spent fixing those annoying bugs and cleaning up the build
408 system; the list of outstanding tasks to complete for the stable release is
409 finally shrinking. The chief bug fixer has been Benjamin Otte who not only
410 made sure that the public API is consistent and being tested for its
411 consistency, but also ensured that the documentation was up-to-date and
412 spent time clarifying cases where even the Cairo developers have come
413 unstuck in the past. Many thanks, Benjamin. However, he was not alone,
414 as Andrea Canciani continued his fine work in isolating broken corner cases
415 and proceeding to fix them, and tidying up the quartz backend. And last, but
416 definitely not least, M Joonas Pihlaja tried building Cairo across a
417 perverse range of systems and fixed up all the loose bits of code that came
418 unravelled. Thanks everybody!
423 cairo_surface_set_mime_data, cairo_surface_get_mime_data:
425 The length parameter is now an unsigned long (as opposed to an unsigned
426 int). The parameter is intended to be an equivalent to a size_t without
427 requiring POSIX types and be large enough to store the size of the
428 largest possible allocation.
430 cairo_gl_surface_create_for_texture:
432 This a new surface constructor for cairo-gl that explicitly enables
433 render-to-texture for foreign, i.e. application, textures.
435 cairo_region_xor, cairo_region_xor_rectangle
437 A couple of utility routines add to the region handling interface for
438 the purpose of replacing existing GdkRegion functionality.
443 https://bugs.launchpad.net/ubuntu/+source/cairo/+bug/600622
445 Inkscape was caught in the act of attempting to modify a finished surface.
446 Unfortunately, we had the ordering of our guards and assertions wrong and
447 so an ordinary application error was triggering an assert in Cairo. This
448 lead Benjamin to add a test case to ensure that the entire public API
449 could handle erroneous input and then proceeded to fix a whole slew of
453 https://bugs.freedesktop.org/show_bug.cgi?id=28888
455 A regression introduced by the special casing of uploading images to an
456 xlib surface in-place which was ignoring the translation applied to the
460 Snapshot 1.9.10 (2010-06-26)
461 ============================
463 The first "quick" snapshot in the run up to the stable release. The
464 last snapshot was picked up by the bleeding edge distributions and so the
465 bug reports have to started to roll in. The most frequent of these are the
466 introduction of rendering errors by applications that modify a surface
467 without subsequently calling cairo_surface_mark_dirty(). Make sure the
468 application developers are aware of increased reliance on strict use of the
469 Cairo API before 1.10 is released!
471 The usual slew of bugs reported and we would like to thank Zoxc for
472 contributing the WGL interface for cairo-gl, and finding more build
473 failures on win32. And it just wouldn't be a 1.9 snapshot unless
474 Benjamin Otte improved the error handling within cairo-gl, as well as
475 isolating and fixing some more errors in the test suite. The biggest bug of
476 the snapshot turned out to be a major sign extension issue that had lain
477 hidden for many years and was suddenly exposed by incorrectly rounding
478 rectangles when performing non-antialiased rendering. Also to the relief
479 of many we have included the downstream patch to honour the user's LCD
480 filtering preferences for subpixel rendering of fonts. The interface
481 remains private for the time being, whilst the proposed public API is
488 Snapshot 1.9.8 (2010-06-12)
489 ===========================
491 One major API changes since the last snapshot, and a whole slew of bugs
492 fixed and inconsistencies eliminated. Far too many bugs fixed to
493 individually identify. We need to thank Benjamin Otte for his fantastic
494 work on the cairo-gl backend making it faster and more robust, Andrea
495 Canciani for finding so many bugs and developing test cases for them, as
496 well fixing them. And last but not least we must all thank Adrian Johnson for
497 continuing to eliminate bugs and improving the PostScript and PDF backends.
499 This snapshot represents almost 4 months of bug fixing, bringing Cairo to
500 a point where we consider it almost ready to be a candidate for release.
501 There are a few known bugs left to be fixed, being tracked in
502 https://bugs.freedesktop.org/show_bug.cgi?id=24384, so please give Cairo a
503 whirl and report any regressions. The plan is to release a new snapshot
504 every other week leading to a 1.10 release with a target date of
509 CAIRO_FORMAT_RGB16_565
511 16 bit devices still remain popular, and so with great demand,
512 CAIRO_FORMAT_RGB16_565 has been restored enabling applications to create
513 and use 16 bit images as sources and render targets.
515 cairo_surface_create_for_rectangle()
517 It is common practice to cut an image up into many smaller pieces and use
518 each of those as a source - a technique called texture atlasing.
519 cairo_surface_create_for_rectangle() extends Cairo to directly support use
520 of these subregions of another cairo_surface_t both as a source and as a
523 cairo_region_create()
524 cairo_region_create_rectangle()
525 cairo_region_create_rectangles()
527 cairo_region_reference()
528 cairo_region_destroy()
530 cairo_region_status()
531 cairo_region_get_extents()
532 cairo_region_num_rectangles()
533 cairo_region_get_rectangle()
534 cairo_region_is_empty()
535 cairo_region_contains_rectangle()
536 cairo_region_contains_point()
537 cairo_region_translate()
538 cairo_region_subtract()
539 cairo_region_subtract_rectangle()
540 cairo_region_intersect()
541 cairo_region_intersect_rectangle()
543 cairo_region_union_rectangle()
545 The Cairo region API was actually added a couple of snapshots ago, but we
546 forgot to mention it at the time. A simple API for the handling of
547 rectangular pixel-aligned regions by Soeren Sandmann.
550 Backend-specific improvements
551 -----------------------------
554 Benjamin Otte made more than 200 commits in which he refactored the cairo-gl
555 backend, reducing a lot of code duplication and enabled him to begin working
556 on improving performance by reducing state changes and associated overhead.
560 Access to the underlying connection to the Display is now thread-safe
561 enabling cairo-xlib to be used in a multi-threaded application without fear
562 of random corruption. Thanks Benjamin Otte!
564 cairo-xlib will now attempt to use PolyModeImprecise when compositing
565 trapezoids (i.e. a fill or a stroke operation with a non-trivial path) which
566 should allow hardware drivers more scope for accelerating the operation at
567 the cost of potentially incurring minute rendering errors. The mode can be
568 forced back to PolyModePrecise by setting the antialias parameter to
569 CAIRO_ANTIALIAS_SUBPIXEL.
573 A notable improvement was contributed by Alexander Shulgin to enable SVG to
574 reference external image through the use an extended MIME data type.
576 Snapshot 1.9.6 (2010-02-19)
577 ===========================
582 The device is a generic method for accessing the underlying interface
583 with the native graphics subsystem, typically the X connection or
584 perhaps the GL context. By exposing a cairo_device_t on a surface and
585 its various methods we enable finer control over interoperability with
586 external interactions of the device by applications. The use case in
587 mind is, for example, a multi-threaded gstreamer which needs to serialise
588 its own direct access to the device along with Cairo's across many
591 Secondly, the cairo_device_t is a unifying API for the mismash of
592 backend specific methods for controlling creation of surfaces with
593 explicit devices and a convenient hook for debugging and introspection.
595 The principal components of the API are the memory management of:
597 cairo_device_reference(),
598 cairo_device_finish() and
599 cairo_device_destroy();
601 along with a pair of routines for serialising interaction:
603 cairo_device_acquire() and
604 cairo_device_release()
606 and a method to flush any outstanding accesses:
608 cairo_device_flush().
610 The device for a particular surface may be retrieved using:
612 cairo_surface_get_device().
614 The device returned is owned by the surface.
616 API changes (to API new in the cairo 1.9.x series)
617 --------------------------------------------------
618 cairo_recording_surface_create()
619 cairo_recording_surface_ink_extents()
621 These are the replacement names for the functions previously named
622 cairo_meta_surface_create and cairo_meta_surface_ink_extents.
624 cairo_surface_set_mime_data
626 This interface is now changed such that the MIME data will be
627 detached if the surface is modified at all. This guarantees that
628 the MIME data will not become out of synch due to surface
629 modifications, and also means that for the MIME data to be useful,
630 it must be set after all modifications to the surface are
633 API removal (of experiment API)
634 -------------------------------
635 The cairo-glitz backend is removed entirely, (in favor of the new
636 cairo-gl backend). See below for more on cairo-gl.
641 Many improvements for drawing of dashed strokes
643 Fix incorrect handling of negative offset
644 Faster computation of first dash (avoids near-infinite looping)
645 Approximate extremely fine dash patterns with appropriate alpha value
647 Optimize spans-based renderers for repeated rows, (such as in a rounded rectangle)
649 Backend-specific improvements
650 -----------------------------
653 This is a new, direct-rendering backend that supports Intel graphics
654 chipsets in the i915 and i965 families. It's still experimental and
655 will likely remain that way for a while. It's already got extremely
656 good performance on the hardware it supports, so if nothing else
657 provides a working proof and performance target for the cairo-gl
658 work for Intel graphics.
662 Start using GLSL to accelerate many operations. Many thanks to Eric
663 Anholt and T. Zachary Laine for this work. For the first time, we
664 have what looks like what will be a very compelling OpenGL-based
665 backend for cairo (in terms of both quality and performance).
667 See this writeup from Eric for more details on recent progress of
668 cairo-gl (which he presented at FOSDEM 2010):
670 http://anholt.livejournal.com/42146.html
674 The image backend is made dramatically faster (3-5 times faster for
675 benchmarks consisting primarily of glyph rendering).
679 Many fixes from Robert O'Callahan and Andrea Canciani including:
681 Fixed gradient pattern painting
682 Improved A8 image handling
683 Fixes for "unbounded" and other compositing operators
687 Improvements to embedding of JPEG and JPEG2000 data.
691 Fix printing of rotated user fonts.
693 Snapshot 1.9.4 (2009-10-15)
694 ===========================
697 cairo_meta_surface_create()
698 cairo_meta_surface_ink_extents()
700 Finally exporting the internal meta-surface so that applications
701 have a method to record and replay a sequence of drawing commands.
705 Determines whether a given point is inside the current clip.
706 ??? Should this be called cairo_in_paint() instead? in-clip is the test
707 that is performed, but in-paint would be similar to in-fill and in-stroke.
713 A companion to cairo-perf-trace, this utility replays a trace against
714 multiple targets in parallel and looks for differences in the output,
715 and then records any drawing commands that cause a failure.
717 Further minimisation of the fail trace using "delta debugging".
718 More control over test/reference targets.
720 Backend improvements:
724 Server-side gradients. The theory is that we can offload computation
725 of gradients to the GPU and avoid pushing large images over the
726 connection. Even if the driver has to fallback and use pixman to render
727 a temporary source, it should be able to do so in a more efficient manner
728 than Cairo itself. However, cairo-perf suggests otherwise:
730 On tiny, Celeron/i915:
732 before: firefox-20090601 211.585
733 after: firefox-20090601 270.939
735 and on tiger, CoreDuo/nvidia:
737 before: firefox-20090601 70.143
738 after: firefox-20090601 87.326
740 In particular, looking at tiny:
742 xlib-rgba paint-with-alpha_linear-rgba_over-512 47.11 (47.16 0.05%) -> 123.42 (123.72 0.13%): 2.62x slowdown
744 xlib-rgba paint-with-alpha_linear3-rgba_over-512 47.27 (47.32 0.04%) -> 123.78 (124.04 0.13%): 2.62x slowdown
748 New experimental backends:
752 OpenVG - The initial work was done by Øyvind Kolås, and made ready for
753 inclusion by Pierre Tardy.
755 OpenGL - An advanced OpenGL compositor. The aim is to write a integrate
756 directed rendering using OpenGL at a high-level into Cairo. In
757 contrast to the previous attempt using Glitz which tried to
758 implement the RENDER protocol on top of OpenGL, using the
759 high-level interface should permit greater flexibility and
760 more offloading onto the GPU.
761 The initial work on the backend was performed by Eric Anholt.
763 Long standing bugs fixed:
765 Self-intersecting strokes.
767 A long standing bug where the coverage from overlapping semi-opaque
768 strokes (including neighbouring edges) was simply summed in lieu of
769 a costly global calculation has been fixed (by performing the costly
770 global calculation!) In order to mitigate the extra cost, the
771 tessellator has been overhauled and tune, which handles the fallback
772 for when we are unable to use the new span rasteriser on the stroke
773 (e.g. when using the current RENDER protocol). The large number of
774 pixel artefacts that implementing self-intersection elimination
775 removes is ample justification for the potential performance
776 regression. If you unfortunately do suffer a substantial performance
777 regression in your application, please consider obtaining a
778 cairo-trace and submitting it to us for analysis and inclusion into
779 our performance suite.
783 To the AuroraUX team for providing access to one of their OpenSolaris
784 machines for cairo and pixman development. http://www.auroraux.org/
786 Snapshot 1.9.2 (2009-06-12)
787 ===========================
790 cairo_surface_set_mime_data()
791 cairo_surface_get_mime_data()
793 Should this take unsigned int, unsigned long or size_t for the length
794 parameter? (Some datasets may be >4GiB in size.)
796 Associate an alternate, compressed, representation for a surface.
798 "image/jp2" (JPEG2000) is understood by PDF >= 1.5
799 "image/jpeg" is understood by PDF,PS,SVG,win32-printing.
800 "image/png" is understood by SVG.
803 cairo_pdf_surface_restrict_to_version()
804 cairo_pdf_get_versions()
805 cairo_pdf_version_to_string()
807 Similar to restrict to version and level found in SVG and PS,
808 these limit the features used in the output to comply with the PDF
809 specification for that version.
811 CAIRO_STATUS_INVALID_SIZE
812 Indicates that the request surface size is not supported by the
813 backend. This generally indicates that the request is too large.
815 CAIRO_STATUS_USER_FONT_NOT_IMPLEMENTED
816 Indicates that a required callback for a user-font was not implemented.
818 CAIRO_STATUS_LAST_STATUS
819 This is a special value to indicate the number of status values enumerated
820 at compile time. (This may differ to the number known at run-time.)
822 The built-in twin font is now called "@cairo:" and supports a limited set
823 of options like "@cairo:mono". Where are these specified?
825 cairo_in_fill() now uses HTML Canvas semantics, all edges are inside.
827 New experimental backends:
833 cairo-trace and cairo-perf-trace
835 cairo-trace generates a human-readable, replayable, compact(-ish!)
836 representation of the sequences of drawing commands made by an
839 Under the util/cairo-script directory is a library to replay traces.
841 perf/cairo-perf-trace replays traces against multiple backends
842 and makes useful benchmark reports. This is integrated with
843 'make perf'. You may collect your own traces or take advantage
844 of traces collected by the community:
846 git://git.cairographics.org/git/cairo-traces
848 (Put this into perf/cairo-traces to run these as part of "make perf".)
850 There is additional WIP in building a debugging tool for cairo applications
851 based on CairoScript (currently very preliminary, mostly serves to show
852 that GtkSourceView is too slow) :
854 people.freedesktop.org:~ickle/sphinx
858 The test suite is undergoing an overhaul, primarily to improve its speed
859 and utility. (Expect more changes in the near future to improve XFAIL
863 polygon rasterisation! Joonas implemented the Tor polygon scan converter,
864 on typical geometry is about 30% faster for the image backend.
866 Bovine Polaroids! For those not in on the joke, this is the long
867 awaited "copy-on-write snapshot" or "COW snapshot" support. The
868 user-visible feature is that including the same image multiple times
869 into a PDF file should result in only a single instance of that
870 image in the final output. This is unlike previous versions of cairo
871 which would generate very large PDF files with multiple copies of
872 the same image. Adrian says that the PDF is not quite working as
873 well as it should yet, so we hope for futher improvements before
880 Better handling of large scale-factors on image patterns.
882 Emit /Interpolate for PS,PDF images.
884 Global glyph cache - cap on the total number of inactive glyphs,
885 should prove fairer for fonts with larger glyph sets.
887 Compilation without fontconfig
889 Improved handling of low-bitdepth sources (e.g. copying the contents
890 of 16-bit xserver windows)
894 cairo_traps_extract_region >10x slower. Fix pending.
898 Region tracking API (ssp) for damage tracking, hit testing etc
901 An expiremental OpenGL backend?
903 Tweaks to tessellator, allocations of patterns, delayed
904 initialisation of the xlib backend (reduce the cairo overhead of
905 render_bench by ~80%).
908 Release 1.8.8 (2009-06-16 Chris Wilson <chris@chris-wilson.co.uk>)
909 ==================================================================
910 The cairo community is pleased to announce the 1.8.8 release of the
911 cairo graphics library. This is the fourth update to cairo's stable
912 1.8 series and contains a small number of bug fixes (in particular a
913 few corrections to the documentation and a few fixes in the FreeType font
914 backend). This is being released just over six months after cairo 1.8.6.
916 We recommend that everyone using cairo upgrade to 1.8.8.
922 There were reports of incompatibilities with the autotools bundled in with
923 the 1.8.6 tarball. This release has been built with automake-1.10.2 and
926 The configure check for FreeType has been improved:
928 typo in check for version of freetype in configure script
929 https://bugs.freedesktop.org/show_bug.cgi?id=19283
931 Compilation on 64-bit MacOS/X fixes:
933 Cannot build cairo_quartz_font_face_create_for_atsu_font_id on 64-bit Mac OS X
934 https://bugs.freedesktop.org/show_bug.cgi?id=15702
938 Uninitialised status return within _cairo_clip_intersect_mask(). This caused
939 random crashes and general mayhem as an error could be generated causing all
940 rendering to the context to stop.
942 Avoid transforming nearly-degenerate matrices into degenerate matrices:
944 Painting stops in this case, using -moz-transform: scale, rotate and video
945 https://bugzilla.mozilla.org/show_bug.cgi?id=467423
947 A few FreeType font handling bugs were fixed:
949 Rendering with PANGO_GRAVITY_EAST leads to different results with image and pdf
950 https://bugs.freedesktop.org/show_bug.cgi?id=21985
952 Don't call FT_Done_Face() on faces we did not create
954 zombie ft_font_face / ft_unscaled_font mutual referencing problems
955 http://bugs.freedesktop.org/show_bug.cgi?id=21706
957 Ensure win32 font backend sets the return value to -1 (indicating the absent
958 glyph) if the font index lookup for the unicode character fails. And
959 similarly fix a bug where a fatal error was raised for an invalid glyph.
961 cairo_scaled_font_glyph_extents breaks with invalid glyph id
962 http://bugs.freedesktop.org/show_bug.cgi?id=20255
964 Various improvements to the documentation, reported by Truc Troung:
966 https://bugs.freedesktop.org/show_bug.cgi?id=20095
967 https://bugs.freedesktop.org/show_bug.cgi?id=20154
968 https://bugs.freedesktop.org/show_bug.cgi?id=20180
969 https://bugs.freedesktop.org/show_bug.cgi?id=20183
970 https://bugs.freedesktop.org/show_bug.cgi?id=20182
971 https://bugs.freedesktop.org/show_bug.cgi?id=20441
974 Release 1.8.6 (2008-12-13 Chris Wilson <chris@chris-wilson.co.uk>)
975 ==================================================================
976 The cairo community is pleased to announce the 1.8.6 release of the
977 cairo graphics library. This is the third update to cairo's stable
978 1.8 series and contains a small number of bug fixes (in particular a
979 few fixes for failures of cairo 1.8.4 on Quartz and PDF, and build fixes for
980 a couple of backends). This is being released just under a month after
983 We recommend that everyone using cairo upgrade to 1.8.6.
989 Fix build of DirectFB backend with debugging enabled:
991 Bug in _cairo_directfb_surface_release_source_image function
992 http://bugs.freedesktop.org/show_bug.cgi?id=18322
998 Workaround a mis-compilation of cairo_matrix_invert() that generated invalid
999 matrices and triggered assertion failures later. The issue was reported by
1002 Invalid computation of the modulus:
1004 https://bugzilla.mozilla.org/show_bug.cgi?id=466258
1006 Invalid referencing of patterns in the Quartz backend:
1008 Failed assertion `CAIRO_REFERENCE_COUNT_HAS_REFERENCE
1009 (&pattern->ref_count)' when using cairo quartz backend
1010 http://bugs.freedesktop.org/show_bug.cgi?id=18632
1012 Invalid references to glyphs after early culling, causing segmentation faults
1015 http://lists.cairographics.org/archives/cairo/2008-December/015976.html
1017 Check for XRender in the XCB backend, or else we may attempt an invalid memory
1020 XCB backend fails with missing render.
1021 https://bugs.freedesktop.org/show_bug.cgi?id=18588
1023 Release 1.8.4 (2008-11-14 Carl Worth <cworth@cworth.org>)
1024 =========================================================
1025 The cairo community is pleased to announce the 1.8.4 release of the
1026 cairo graphics library. This is the second update to cairo's stable
1027 1.8 series and contains a small number of bug fixes, (in particular a
1028 few fixes for build failures of cairo 1.8.2 on various systems). This
1029 is being released just over two weeks after cairo 1.8.2.
1031 We recommend that everyone using cairo upgrade to 1.8.4.
1037 Fix build with older XRender that doesn't define RepeatNone:
1039 Build of xlib backend fails against old XRender (RepeatNone undeclared)
1040 https://bugs.freedesktop.org/show_bug.cgi?id=18385
1042 Fix build with bash version <= 3.0:
1044 doltlibtool broken on linux with bash 3.00.0
1045 https://bugs.freedesktop.org/show_bug.cgi?id=18363
1049 Avoid triggering a bug in X.org server 6.9 resulting in a hung machine
1052 https://bugs.freedesktop.org/show_bug.cgi?id=15628#c2
1054 Fix display of user fonts as exercised by proposed support for type3
1055 fonts in poppler (unsigned promotion fixes):
1057 Use cairo user-font for Type 3 fonts
1058 http://lists.freedesktop.org/archives/poppler/2008-October/004181.html
1060 Avoid miscomputing size of fallback images required when rendering
1061 with CLEAR, IN, or SOURCE operator to vector surfaces, (PS, PDF, SVG,
1064 Be more tolerant of broken fonts when subsetting type1 fonts:
1066 Error handling in cairo_type1_font_subset_get_glyph_names_and_widths
1067 http://lists.cairographics.org/archives/cairo/2008-October/015569.html
1069 Fix cairo_fill_extents, cairo_stroke_extents, cairo_path_extents, to
1070 correctly allow NULL parameters as documented.
1072 Fix potential crash on emitting a type3 glyph after having drawn text
1073 paths from the same font, (for example with cairo_text_path).
1075 Release 1.8.2 (2008-10-29 Carl Worth <cworth@cworth.org>)
1076 =========================================================
1077 The cairo community is pleased to announce the 1.8.2 release of the
1078 cairo graphics library. This is the first update to cairo's stable 1.8
1079 series and contains a large number of bug fixes. It is being released
1080 just over one month since cairo 1.8.0.
1082 This release consists primarily of bug fixes, but there is one notable
1083 new feature, (the ability to build cairo without an external font
1084 backend), and there are a few optimizations as well. See below for
1085 details on these changes and the most important bug fixes.
1087 While many people have contributed to this release, Chris Wilson
1088 deserves particular mention. He has contributed well over twice as
1089 many changes to cairo since 1.8.0 than everyone else combined. We
1090 greatly appreciate the tremendous efforts of Chris and all cairo
1093 We recommend everyone upgrade to cairo 1.8.2 and hope that everyone
1094 will have lots of fun with cairo!
1100 It is now possible to build cairo without any font backend, (such as
1101 freetype, win32 or quartz). This is most useful when the application
1102 provides custom font rendering through the user-font API. But in the
1103 case where no external font backend is available, and no user-font is
1104 provided, cairo will render with a failsafe font, (a stroked font
1105 covering visible ASCII character). (Behdad Esfahbod)
1109 Dramatically speed up compilation with dolt (removes much of the
1110 libtool overhead) (Behdad Esfahbod with thanks to Josh Triplett).
1112 Several minor optimizations to tessellator (special-cased comparisons,
1113 faster insert for skiplist, etc.) (Chris Wilson).
1115 Optimize away fractional translation component when doing
1116 EXTEND_NEAREST filtering, (for better performance).
1120 Allow cloning sub-regions of similar surfaces to fix this bug
1123 Crafted gif file will crash firefox
1124 [XError: 'BadAlloc (insufficient resources for operation)']
1125 https://bugzilla.mozilla.org/show_bug.cgi?id=424333
1127 Fix some matrix confusion to fix this regression (Chris Wilson):
1129 Translucent star exports in a wrong way to PDF
1130 https://bugs.launchpad.net/inkscape/+bug/234546
1132 Fix some long-standing bugs with respect to properly computing the
1133 extents of transformed, filtered surfaces (Owen Taylor, Carl Worth,
1136 Bad clipping with EXTEND_NONE
1137 http://bugs.freedesktop.org/show_bug.cgi?id=15349
1139 Improve filtering handling in cairo-pattern.c
1140 http://bugs.freedesktop.org/show_bug.cgi?id=15367
1142 Many thanks to Chris Wilson for digging out and cleaning up
1145 Fix compilation on Solaris 10 (Chris Wilson):
1147 Cairo requires -DREENTRANT (along with -D_POSIX_THREAD_SEMANTICS)
1148 to compile on Solaris 10 with pthreads
1149 https://bugs.freedesktop.org/show_bug.cgi?id=18010
1151 Fix very old bug causing dashes to be rendered at the wrong length in
1152 fallback images (Adrian Johnson)
1154 Dashed strokes too long in fallback images
1155 https://bugs.freedesktop.org/show_bug.cgi?id=9189
1157 Fix broken dashing when a dashed path starts outside the clip region
1160 Avoid range overflow when computing large patterns (Benjamin Otte and
1163 Avoid crashing due to an invalid font with an incorrect entry in its
1164 CMAP table (Adrian Johnson).
1166 Fix bugs in computing maximum size of text requests that can be sent
1167 with the Render extension, (avoiding potential crashes when rendering
1168 large amounts of text) (Behdad Esfahbod and Chris Wilson).
1170 Fix rendering of operators unbounded by the mask (Chris Wilson).
1172 Fix compilation on systems without compiler support for a native
1173 64-bit type (Chris Wilson).
1175 Fix several cases of missing error-status propagation. (Chris Wilson,
1176 doing the work he seems to never tire of).
1178 Fix several locking issues found with the lockdep valgrind skin (Chris
1181 Backend-specific bug fixes
1182 --------------------------
1183 xlib: Avoid crash due to attempting XRender calls on pixmaps with
1184 formats not supported by the Render extension (Chris Wilson):
1186 XRender crashes due to NULL pointer from Cairo on SGI O2
1187 https://bugs.freedesktop.org/show_bug.cgi?id=11734
1189 xlib: Add support for XImages with depth of 4, 20, 24, or 28 bits
1192 cairo doesn't support 24 bits per pixel mode on X11
1193 https://bugs.freedesktop.org/show_bug.cgi?id=9102
1195 xlib: Avoid mistakenly considering two surfaces as similar just
1196 because their depths match (while their Render formats do not) (Karl
1199 ps: Fix slight mis-scaling of bitmapped fonts (Adrian Johnson)
1201 svg: Correctly emit comp-op for paint, mask, and show_glyphs
1202 operations (Emmanuel Pacaud).
1204 svg: Use finer-grained fallbacks for SVG 1.2 (as PS and PDF backends
1205 have been doing since 1.6.0) (Chris Wilson).
1207 win32: Fallback to DIB if DDB create fails for
1208 cairo_surface_create_similar (Vladimir Vukicevic).
1210 win32: Fix compatibility with Windows Mobile (Vladimir Vukicevic).
1212 win32: Fix static builds to not do __declspec(dllimport) on public
1213 functions. This requires the user to set a CAIRO_WIN32_STATIC_BUILD
1214 environment variable when compiling (Behdad Esfahbod).
1216 Release 1.8.0 (2008-09-25 Carl Worth <cworth@cworth.org>)
1217 =========================================================
1218 The cairo community is happy (and relieved) to announce the 1.8.0
1219 release of the cairo graphics library. This is a major update to
1220 cairo, with new features and enhanced functionality which maintains
1221 compatibility for applications written using any previous major cairo
1222 release, (1.6, 1.4, 1.2, or 1.0). We recommend that anybody using a
1223 previous version of cairo upgrade to cairo 1.8.0.
1225 The dominant theme of this release is improvements to cairo's ability
1226 to handle text. The highlights include a new "user fonts" feature as
1227 well as a new cairo_show_text_glyphs API which allows glyphs to be
1228 embedded in PDF output along with their original text, (for searching,
1229 selection, and copy-and-paste). Another major feature is a revamp of
1230 cairo's build system making it much easier to build cairo on various
1233 See below for more details.
1237 This new API allows the user of cairo API to provide drawings for
1238 glyphs in a font. A common use for this is implementing fonts in
1239 non-standard formats, like SVG fonts and Flash fonts. This API can
1240 also be used by applications to provide custom glyph shapes for fonts
1241 while still getting access to cairo's glyph caches. See
1242 test/user-font.c and test/user-font-proxy.c for usage examples. This
1243 is based on early work by Kristian Høgsberg. Thanks Kristian!
1245 This new API consists of the following functions (and corresponding
1248 cairo_user_font_face_create
1250 cairo_user_font_face_set_init_func
1251 cairo_user_font_face_set_render_glyph_func
1252 cairo_user_font_face_set_text_to_glyphs_func
1253 cairo_user_font_face_set_unicode_to_glyph_func
1255 An additional, new API is
1257 cairo_scaled_font_text_to_glyphs
1259 We were previously reluctant to provide this function as
1260 text-to-glyphs support in cairo was limited to "toy" font
1261 functionality, not really interesting for real-world text
1262 processing. However, with user fonts landing, this API is needed to
1263 expose full access to how user fonts convert text to glyphs. This is
1264 expected to be used by text toolkits like Pango, as well as "proxy"
1265 user-font implementations.
1267 cairo_show_text_glyphs
1268 ----------------------
1269 This new API allows the caller of cairo to provide text data
1270 corresponding to glyphs being drawn. The PDF backend implements this
1271 new API so that complex text can be copied out of cairo's PDF output
1272 correctly and reliably, (assuming the user of cairo calls
1273 cairo_show_text_glyphs). The cairo_show_text_glyphs API is definitely
1274 the most daunting API to debut in cairo. It is anticipated that pango
1275 (and similar high-level text libraries) will be the primary users of
1276 this API. In fact, pango 1.22 already uses cairo_show_text_glyphs.
1277 Behdad was the architect and implementor of this effort. Thanks,
1280 The cairo_show_text_glyphs API includes the following new functions:
1282 cairo_show_text_glyphs
1284 cairo_glyph_allocate
1287 cairo_text_cluster_allocate
1288 cairo_text_cluster_free
1290 cairo_surface_has_show_text_glyphs
1294 The primary goal of the revamp is to make the build system less
1295 fragile, (particularly for non-Linux platforms). For example, now
1296 people building on win32 will no longer need to maintain a
1297 platform-specific list of files to be built. See the new README.win32
1298 for details. Also, the .so file will now be installed with a different
1299 naming scheme, (for example, 1.7.6 will install with a .10800
1300 suffix). Many thanks to Behdad and his small army of helpers!
1302 Assorted API additions
1303 ----------------------
1304 For API completeness, several missing "getter" functions were added:
1306 cairo_scaled_font_get_scale_matrix
1308 cairo_surface_get_fallback_resolution
1310 cairo_toy_font_face_create
1311 cairo_toy_font_face_get_family
1312 cairo_toy_font_face_get_slant
1313 cairo_toy_font_face_get_weight
1315 The new cairo_toy_font_face functions provide access to functionality
1316 and settings provided by cairo_select_font_face(). Thanks Behdad!
1318 cairo-ps/cairo-pdf: More efficient output
1319 -----------------------------------------
1320 Adrian Johnson has been busy fixing all kinds of bugs in PS and PDF
1321 backends, as well making them generate much more compact output by
1322 avoiding things like re-emitting the color or linestyle on every
1323 drawing operation. Thanks Adrian!
1325 cairo-xlib: dithering
1326 ---------------------
1327 Dithering: Cairo now does simple dithering when rendering to legacy X
1328 servers. This is most visible with 8-bit visuals. Thanks Behdad!
1330 cairo-xlib: Avoid rendering glyphs out of surface bounds
1331 --------------------------------------------------------
1332 This seemingly harmless optimization exposed a bug in OpenOffice.org 3
1333 versions where OO.o was passing bogus surface extents to cairo,
1334 resulting in no text rendered in OO.o. Please contact your
1335 distribution's OO.o maintainers if you see this bug and point them to
1338 https://bugs.freedesktop.org/show_bug.cgi?id=16209
1340 cairo-xlib: Improved performance with X server without Render
1341 -------------------------------------------------------------
1342 Cairo now performs better on remote X servers that lack the Render
1343 extension by being smarter about using X core protocol facilities
1344 instead of falling back to doing all rendering on the client side.
1346 cairo-ft: respecting FC_FT_FACE
1347 -------------------------------
1348 Previously it was impossible to instruct cairo to do emboldening on a
1349 font face object created from an FT_Face. Cairo now respects and uses
1350 the FC_FT_FACE fontconfig pattern element, so emboldening can be
1351 achieved by using cairo_ft_font_face_create_for_pattern() and a
1352 carefully crafted pattern using FC_FT_FACE and FC_EMBOLDEN. Thanks
1355 cairo-directfb: backend improvements
1356 ------------------------------------
1357 The directfb backend, though still unsupported, has seen a good deal
1358 of improvements. Thanks Vlad!
1360 Bug fixing and optimizations
1361 ----------------------------
1362 xlib: Faster bookkeeping (Karl Tomlinson)
1363 https://bugzilla.mozilla.org/show_bug.cgi?id=453199#c5
1365 PS: Fix gradients with non-constant alpha (Chris Wilson)
1367 Fix deadlock in user-font code (Richard Hughes and Behdad Esfahbod)
1368 http://bugs.freedesktop.org/show_bug.cgi?id=16819
1370 Countless other bugs have been fixed and optimizations made, many of
1371 them thanks to Chris Wilson. Thanks Chris and others!
1373 Note also that the code that had been in cairo 1.7.x calling into
1374 freetype's optional lcd_filter function was removed from cairo before
1375 the 1.8.0 release. We do expect this code to come back in some form in
1378 Snapshot 1.7.6 (2008-09-17 Carl Worth <cworth@cworth.org>)
1379 ==========================================================
1380 The cairo community is happy to announce the 1.7.6 snapshot of the
1381 cairo graphics library. This is a "release candidate" for the upcoming
1382 1.8.0 release, so we will greatly appreciate any reports of problems
1383 in this release, and no major changes are currently planned before
1386 Notable changes in 1.7.6
1387 ------------------------
1388 The largest number of changes since 1.7.4 did not change the
1389 implementation of cairo itself, but instead revamped cairo's build
1390 system. The primary goal of the revamp is to make the build system
1391 less fragile, (particularly for non-Linux platforms). For example, now
1392 people building on win32 will no longer need to maintain a
1393 platform-specific list of files to be built. Also, the .so file will
1394 now be installed with a different naming scheme, (for example, 1.7.6
1395 will install with a .10706 suffix). Much thanks, Behdad!
1397 And, as usual, Chris Wilson has made another large round of robustness
1398 improvements, (eliminating dead code, fixing propagation of error
1399 status values, test suite improvements, etc. etc.). Thanks as always,
1402 API changes since 1.7.4
1403 -----------------------
1404 There have been a few changes of API that was new during the 1.7
1407 * Remove cairo_font_options_set_lcd_filter
1408 and cairo_font_options_get_lcd_filter
1410 Motivation: At the Cairo Summit, this API was determined to be too
1411 specific to the freetype font backend to be in the general
1412 API. A similar API with a cairo_ft prefix might be introduced
1413 in the future. Note that cairo will still respect the
1414 corresponding fontconfig settings for these options.
1416 * Replace cairo_has_show_glyphs
1417 with cairo_surface_has_show_glyphs
1419 Motivation: This really is a surface-specific interface, and the
1420 convenience function on the cairo_t is not obviously
1421 necessary. An application can easily call:
1423 cairo_surface_has_show_glyphs (cairo_get_target (cr));
1427 * Add cairo_text_cluster_flags_t
1428 to cairo_show_text_glyphs
1429 cairo_scaled_font_text_to_glyphs
1430 cairo_user_scaled_font_text_to_glyphs_func_t
1432 Motivation: This flag, (and specifically the
1433 CAIRO_TEXT_CLUSTER_FLAG_BACKWARD value), replaces the
1434 cairo_bool_t backward argument in each of the above
1435 interfaces. This leads to more readable user code, and also
1436 allows future extensibility.
1438 As always, there are no changes to any API from any major cairo
1439 release, (1.0.x, 1.2.x, 1.4.x, 1.6.x). Cairo maintains the same
1440 compatibility promise it always has.
1442 Bug fixes since 1.7.4
1443 ---------------------
1444 xlib: Faster bookkeeping (Karl Tomlinson)
1445 https://bugzilla.mozilla.org/show_bug.cgi?id=453199#c5
1447 PS: Fix gradients with non-constant alpha (Chris Wilson)
1449 Fix deadlock in user-font code (Richard Hughes and Behdad Esfahbod)
1450 http://bugs.freedesktop.org/show_bug.cgi?id=16819
1452 Several other minor fixes.
1454 Snapshot 1.7.4 (2008-08-11 Behdad Esfahbod <behdad@behdad.org>)
1455 ===============================================================
1456 The cairo community is embarrassed to announce availability of the 1.7.4
1457 snapshot of the cairo graphics library. This is a followup release to the
1458 1.7.2 snapshot to ship a tarball that can actually be built. The only
1459 change since 1.7.4 is including the missing header file
1460 cairo-user-font-private.h in the distribution.
1462 Snapshot 1.7.2 (2008-08-11 Behdad Esfahbod <behdad@behdad.org>)
1463 ===============================================================
1464 The cairo community is finally ready to announce availability of the 1.7.2
1465 snapshot of the cairo graphics library. This is embarrassingly the first
1466 snapshot in the 1.7 unstable series of cairo, leading to the eventual release
1467 of cairo 1.8, currently planned for late September.
1469 This snapshot comes four months after the 1.6.4 release. We have done a
1470 really bad job on getting development snapshots out this cycle, but
1471 hopefully all the API changes for 1.8 are now finished and the remaining
1472 weeks will be spent on bug-fixing. There is more than 400 commits worth
1473 of changes in this snapshot, and those can use some testing. Read on!
1475 Text, text, and more text!
1476 --------------------------
1477 The dominant theme of this release, and 1.8 in general, is improvements
1478 around cairo text API. Here is a high-level list of changes with text
1483 This is new API allowing the user of cairo API to provide drawings for glyphs
1484 in a font. This is most useful in implementing fonts in non-standard formats,
1485 like SVG fonts and Flash fonts, but can also be used by games and other
1486 applications to draw "funky" fonts. See test/user-font.c and
1487 test/user-font-proxy.c for usage examples. This is based on early work by
1488 Kristian Høgsberg. Thanks Kristian!
1492 This new API allows the caller of cairo to mark text glyphs with their
1493 original text. The PDF backend implements this new API and latest Pango
1494 master uses it. The result is (when bugs are fixed) that complex text can be
1495 copied out of pangocairo's PDF output correctly and reliably. There are bugs
1496 to fix though. A few poppler bugs, and some more in cairo and pango.
1498 To test show_text_glyph, just grab pango master and this cairo snapshot and
1499 print text in gedit. Open in acroread or evince, select all, copy, paste
1500 in gedit and compare. The Arabic text with diacritic marks is particularly
1501 showing bad. Try with pango/pango-view/HELLO.txt if you are brave
1502 enough. The Indic text is showing improvements, but is still coming out
1505 LCD subpixel filtering using FreeType
1506 -------------------------------------
1507 FreeType 2.3.5 added support for various LCD subpixel filtering, and
1508 fontconfig 2.6.0 added support for configuring LCD filter on a font by font
1509 basis. Cairo now relies on FreeType and fontconfig for subpixel filtering.
1510 This work is based on David Turner's original patch to cairo, maintained
1511 and tested by Sylvain Pasche and others. Thanks all!
1513 Toy font face constructor and getter
1514 ------------------------------------
1515 Mostly for API completion, but also useful for higher level (like Pango) to
1516 hook into what the user has set using cairo_select_font_face(), making that
1517 toy API a bit more useful.
1519 FreeType: respecting FC_FT_FACE
1520 -------------------------------
1521 Previously it was impossible to instruct cairo to do emboldening on a font
1522 face object created from an FT_Face. Cairo now respects and uses the
1523 FC_FT_FACE fontconfig pattern element, so emboldening can be achieved by
1524 using cairo_ft_font_face_create_for_pattern() and a carefully crafted pattern
1525 using FC_FT_FACE and FC_EMBOLDEN.
1528 PS/PDF: More efficient output
1529 -----------------------------
1530 Adrian Johnson has been busy fixing all kinds of bugs in PS and PDF
1531 backends, as well making them generate much more compact output by avoiding
1532 things like re-emitting the color or linestyle on every drawing operation.
1538 Cairo now does simple dithering when rendering to legacy X servers. This is
1539 mostly visible with 8-bit visuals.
1541 Xlib: Avoid rendering glyphs out of surface bounds
1542 --------------------------------------------------
1543 This seemingly harmless change manifested a bug with OpenOffice.org 3 versions
1544 where OO.o was passing bogus surface extents to cairo, resulting in no text
1545 rendered in OO.o. Please contact your distro's OO.o maintainers if you see
1546 this bug and point them to the following URL:
1548 https://bugs.freedesktop.org/show_bug.cgi?id=16209
1550 Xlib: Improved performance with Xrender-less X servers
1551 ------------------------------------------------------
1552 Cairo now performs better on remote, Xrender-less X servers by being smarter
1553 about using X core protocol facilities instead of falling back to doing all
1554 rendering on the client side.
1557 Directfb: backend improvements
1558 ------------------------------
1559 The directfb backend, though still unsupported, has seen a good deal of
1560 improvements. Thanks Vlad!
1563 Bug fixing and optimizations
1564 ----------------------------
1565 Countless bugs have been fixed and optimizations made, many of them thanks to
1566 Chris Wilson. Thanks Chris!
1572 cairo_show_text_glyphs
1574 This is a new text rendering API. Being a more advanced version of
1575 cairo_show_glyphs(), it is aimed for use by higher-level text toolkits like
1576 Pango, and enables better text extraction from output generated by backends
1577 like PDF and SVG. The PDF backend already implements it, and the upcoming
1578 Pango release will use it.
1580 To make that API work, a bunch of other additions were made:
1582 cairo_glyph_allocate
1584 cairo_text_cluster_t
1585 cairo_text_cluster_allocate
1586 cairo_text_cluster_free
1587 cairo_surface_has_show_text_glyphs
1590 cairo_user_font_face_create
1592 This is the "user" font face constructor, accompanied by a variety of method
1593 signatures, getters, and setters for a callback-based font backend:
1595 CAIRO_FONT_TYPE_USER
1596 cairo_user_scaled_font_init_func_t
1597 cairo_user_scaled_font_render_glyph_func_t
1598 cairo_user_scaled_font_text_to_glyphs_func_t
1599 cairo_user_scaled_font_unicode_to_glyph_func_t
1600 cairo_user_font_face_set_init_func
1601 cairo_user_font_face_set_render_glyph_func
1602 cairo_user_font_face_set_text_to_glyphs_func
1603 cairo_user_font_face_set_unicode_to_glyph_func
1604 cairo_user_font_face_get_init_func
1605 cairo_user_font_face_get_render_glyph_func
1606 cairo_user_font_face_get_text_to_glyphs_func
1607 cairo_user_font_face_get_unicode_to_glyph_func
1610 cairo_scaled_font_text_to_glyphs
1612 We were previously reluctant to provide this function as text-to-glyphs
1613 support in cairo was limited to "toy" font functionality, not really
1614 interesting for real-world text processing. However, with user-fonts
1615 landing, this API is needed to expose full access to how user-fonts
1616 convert text to glyphs. This is expected to be used by text toolkits like
1617 Pango, as well as "proxy" user-font implementations.
1621 cairo_font_options_set_lcd_filter
1622 cairo_font_options_get_lcd_filter
1624 These add the possibility to choose between various available LCD subpixel
1625 filters. The available filter values are modeled after what FreeType
1629 cairo_toy_font_face_create
1630 cairo_toy_font_face_get_family
1631 cairo_toy_font_face_get_slant
1632 cairo_toy_font_face_get_weight
1634 These provide access to functionality and settings provided by
1635 cairo_select_font_face().
1638 cairo_scaled_font_get_scale_matrix
1639 cairo_surface_get_fallback_resolution
1641 For API completeness.
1644 Various new values for cairo_status_t enum
1649 - Type3 fonts generated by cairo's PDF backend may show up in poppler/Evince
1650 in a different color than expected. This is fixed in poppler master branch.
1651 This mostly affects cairo user fonts. The test case test/user-font.c
1654 - User fonts using other fonts in their rendering are currently embedded in
1655 PDF as fallback bitmap glyphs. This will be (hopefully) fixed before 1.8.
1656 The test case test/user-font-proxy.c demonstrates this.
1659 Release 1.6.4 (2008-04-11 Carl Worth <cworth@cworth.org>)
1660 =========================================================
1661 The cairo community is wildly embarrassed to announce the 1.6.4
1662 release of the cairo graphics library. This release reverts the xlib
1663 locking change introduced in 1.6.4, (and the application crashes that
1664 it caused). The community would be glad to sack its current release
1665 manager and is accepting applications for someone who could do the job
1666 with more discipline.
1668 Revert 'add missing locking in cairo-xlib'
1669 ------------------------------------------
1670 This change was introduced in cairo 1.6.2, but also introduced a bug
1671 which causes many cairo-xlib applications to crash, (with a
1672 segmentation fault inside of XSetClipMask). Instead of attempting
1673 another fix for the broken fix, the change in 1.6.2 has been
1674 reverted. The original bug which the change was addressing has been
1675 present since at least cairo 1.4, so it is not expected that leaving
1676 this bug unfixed will cause any new problems for applications moving
1677 from cairo 1.4 to cairo 1.6.
1679 At this point, the code of cairo 1.6.4 differs from cairo 1.6.0 only
1680 in the fix for the PostScript-printer crashes.
1682 Tweak build to avoid linking with g++
1683 -------------------------------------
1684 Cairo 1.6.4 avoids a quirk in automake that was causing the cairo
1685 library to be linked with g++ and linked against libstdc++ even when
1686 only C source files were compiled for the library.
1688 Release 1.6.2 (2008-04-11 Carl Worth <cworth@cworth.org>)
1689 =========================================================
1690 The cairo community is pleased (but somewhat sheepish) to announce the
1691 1.6.2 release of the cairo graphics library. This is an update to
1692 yesterday's 1.6.0 release with an important fix to prevent cairo's
1693 PostScript output from crashing some printers. This release also
1694 includes a locking fix for cairo's xlib backend to improve thread
1695 safety. There are no changes beyond these two fixes.
1697 Fix for PostScript printer crash
1698 --------------------------------
1699 Adrian Johnson discovered that cairo 1.6.0 was being a bit hard on
1700 PostScript printers, by changing the font matrix very frequently. This
1701 causes some PostScript interpreters to allocate new font objects every
1702 few glyphs, eventually exhausting available resources. The fix
1703 involves leaving translational components of the font matrix as zero,
1704 so that the PostScript interpreter sees an identical font matrix
1705 repeatedly, and can more easily share internal font object resources.
1707 This fix has been tested to resolve the bugs posted here, (for both
1708 Xerox and Dell printers):
1710 Printing some PDFs from evince is crashing our Xerox printer
1711 http://bugs.freedesktop.org/show_bug.cgi?id=15348
1713 Cairo-generated postscript blocks Dell 5100cn
1714 http://bugs.freedesktop.org/show_bug.cgi?id=15445
1716 Add missing locking in cairo-xlib
1717 ---------------------------------
1718 Chris Wilson noticed that cairo 1.6.0 was manipulating an internal
1719 cache of GC object within cairo's Xlib backend without proper
1720 locking. The missing locking could cause failures for multi-threaded
1721 applications. He fixed this in 1.6.2 by adding the missing locks.
1723 Release 1.6.0 (2008-04-10 Carl Worth <cworth@cworth.org>)
1724 =========================================================
1725 The cairo community is quite pleased to announce the 1.6.0 release of
1726 the cairo graphics library. This is a major update to cairo, with new
1727 features and enhanced functionality which maintains compatibility for
1728 applications written using cairo 1.4, 1.2, or 1.0. We recommend that
1729 anybody using a previous version of cairo upgrade to cairo 1.6.0.
1731 The most significant new features in this release are dramatically
1732 improved PDF and PostScript[*] output, support for arbitrary X server
1733 visuals (including PseudoColor), a new Quartz backend, and and a new
1734 "win32 printing" backend. See below for more details on these and
1737 New dependency on external pixman library (Thanks, Søren!)
1738 ----------------------------------------------------------
1739 As of cairo 1.6, cairo now depends on the pixman library, for which
1740 the latest release can be obtained alongside cairo:
1742 http://cairographics.org/releases/pixman-0.10.0.tar.gz
1744 This library provides all software rendering for cairo, (the
1745 implementation of the image backend as well as any image fallbacks
1746 required for other backends). This is the same code that was
1747 previously included as part of cairo itself, but is now an external
1748 library so that it can be shared by both cairo and by the X server,
1749 (which is where the code originated).
1751 Improved PDF, PostScript, and SVG output (Thanks, Adrian!)
1752 ----------------------------------------------------------
1753 Users of the cairo-pdf, cairo-ps, and cairo-svg should see a dramatic
1754 improvement from cairo 1.2/1.4 to 1.6. With this release there are now
1755 almost no operations that will result in unnecessary rasterization in
1756 the PDF and PostScript. Rasterized "image fallbacks" are restricted
1757 only to minimal portions of the document where something is being
1758 drawn with cairo that is beyond the native capabilities of the
1759 document, (this is rare for PDF or SVG, but occurs when blending
1760 translucent objects for PostScript).
1762 This means that the final output will be of higher quality, and will
1763 also be much smaller, and therefore will print more quickly. The
1764 machinery for doing analysis and minimal fallbacks also benefits the
1765 win32-printing surface described below.
1767 In addition to doing less rasterization, the PostScript and PDF output
1768 also has several other improvements to make the output more efficient
1769 and more compatible with specifications.
1771 [*] Note: Just before this release, a bug has been reported that the
1772 PostScript output from cairo can crash some printers, (so far the
1773 following models have been reported as problematic Xerox Workcentre
1774 7228 or 7328 and Dell 5100cn). We will implement a workaround as soon
1775 as we can learn exactly what in cairo's output these printers object
1776 to, (and we could use help from users that have access to misbehaving
1777 printers). This bug is being tracked here:
1779 Printing some PDFs from evince is crashing our Xerox printer
1780 http://bugs.freedesktop.org/show_bug.cgi?id=15348
1782 New support for arbitrary X server visuals (Thanks, Keith and Behdad!)
1783 ----------------------------------------------------------------------
1784 As of cairo 1.6, cairo should now work with an arbitrary TrueColor or
1785 8-bit PseudoColor X server visual. Previous versions of cairo did not
1786 support these X servers and refused to draw anything. We're pleased to
1787 announce that this limitation has been lifted and people stuck with
1788 ancient display systems need no longer be stuck with ancient software
1789 just because of cairo.
1791 New, supported Quartz backend for Mac OS X (Thanks, Brian and Vladimir!)
1792 ------------------------------------------------------------------------
1793 As of cairo 1.6, the cairo-quartz backend is now marked as "supported"
1794 rather than "experimental" as in previous cairo releases. Its API now
1795 has guarantees of API stability into future cairo releases, and its
1796 output quality is comparable to other backends. There have been
1797 significant improvements to cairo-quartz since 1.4. It now uses many
1798 fewer image fallbacks, (meaning better performance), and has greatly
1799 improved text rendering.
1801 New, "win32 printing" backend (Thanks, Adrian and Vladimir!)
1802 ------------------------------------------------------------
1803 A new win32-printing surface has been added with an interface very
1804 similar to the original win32 surface, (both accept an HDC
1805 parameter). But this new surface should only be called with a printing
1806 DC, and will result in all drawing commands being stored into a
1807 meta-surface and emitted after each page is complete. This allows
1808 cairo to analyze the contents, (as it does with PDF, PostScript, and
1809 SVG backends), and to do minimal image-based fallbacks as
1810 necessary. The analysis keeps things as efficient as possible, while
1811 the presence of fallbacks, (when necessary), ensure the consistent,
1812 high-quality output expected from cairo.
1814 Robustness fixes (Thanks, Chris!)
1815 ---------------------------------
1816 There has been a tremendous number of improvements to cairo's
1817 robustness. Areas that have been improved include:
1819 * Proper reporting of errors
1821 * Responding correctly to invalid input
1823 * Avoiding integer overflows
1825 * Avoiding memory leaks on error-recovery paths
1827 * Making reference counting thread safe
1829 * Exhaustive testing of memory allocation points
1831 Other fixes (Thanks, everybody!)
1832 --------------------------------
1833 Cairo's internal fixed-point representation has been changed from
1834 16.16 to 24.8. This has a direct impact on applications as it allows
1835 much larger objects to be drawn before internal limits in cairo make
1836 the drawing not work.
1838 The CAIRO_EXTEND_PAD mode is now fully supported by surface
1839 patterns. This mode allows applications to use cairo_rectangle and
1840 cairo_fill to draw scaled images with high-quality bilinear filtering
1841 for the internal of the image, but without any objectionably blurry
1842 edges, (as would happen with the default EXTEND_NONE and cairo_paint).
1844 Rendering with CAIRO_ANTIALIAS_NONE has been fixed to be more
1845 predictable, (previously image rendering and geometry rendering would
1846 be slightly misaligned with respect to each other).
1848 The reference manual at http://cairographics.org/manual now documents
1849 100% of the functions and types in cairo's public API.
1853 Several small features have been added to cairo with new API functions:
1855 cairo_format_stride_for_width
1857 Must be called to compute a properly aligned stride value before
1858 calling cairo_image_surface_create_for_data.
1860 cairo_has_current_point
1862 Allows querying if there is a current point defined for the
1867 Allows querying for path extents, (independent of any fill or
1870 cairo_surface_copy_page
1871 cairo_surface_show_page
1873 Allow beginning a new document page without requiring a cairo_t
1876 cairo_ps_surface_restrict_to_level
1878 cairo_ps_level_to_string
1879 cairo_ps_surface_set_eps
1881 Allow controlling the Post PostScript level, (2 or 3), to
1882 target, as well as to generate Encapsulated PostScript (EPS).
1884 cairo_quartz_font_face_create_for_cgfont
1886 Create a quartz-specific cairo_font_face_t from a CGFontRef.
1888 cairo_win32_font_face_create_for_logfontw_hfont
1890 Create a win32-specific cairo_font_face from a LOGFONTW and an
1895 I've accounted for 32 distinct people with attributed code added to
1896 cairo between 1.4.14 and 1.6.0, (their names are below). That's an
1897 impressive number, but there are certainly dozens more that
1898 contributed with testing, suggestions, clarifying questions, and
1899 encouragement. I'm grateful for the friendships that have developed as
1900 we have worked on cairo together. Thanks to everyone for making this
1903 Adrian Johnson, Alp Toker, Antoine Azar, Behdad Esfahbod,
1904 Benjamin Otte, Bernardo Innocenti, Bertram Felgenhauer,
1905 Boying Lu, Brian Ewins, Carl Worth, Chris Heath, Chris Wilson,
1906 Claudio Ciccani, Emmanuel Pacaud, Jeff Muizelaar, Jeremy Huddleston,
1907 Jim Meyering, Jinghua Luo, Jody Goldberg, Jonathan Gramain,
1908 Keith Packard, Ken Herron, Kouhei Sutou, Kristian Høgsberg,
1909 Larry Ewing, Martin Ejdestig, Nis Martensen, Peter Weilbacher,
1910 Richard Hult, Shailendra Jain, Søren Sandmann Pedersen,
1913 Snapshot 1.5.20 (2008-04-04 Carl Worth <cworth@cworth.org>)
1914 ===========================================================
1915 This is the tenth snapshot in cairo's unstable 1.5 series. It comes
1916 just two days (and only one working day) after the 1.5.18
1917 snapshot. The quick snapshot is due to two embarrassing bugs (both
1918 affecting cairo-xlib) that had been introduced in the 1.5.18
1919 snapshot. The fixes for these are described below along with a few
1920 other fixes, (which hopefully aren't introducing new bugs this time).
1924 Revert fix from 1.5.18 to allow pattern expansion based on the filter
1925 mode. This fix seemed so boring, (the use case it addresses is almost
1926 never used in practice), that it didn't even get mentioned in the
1927 1.5.18 release notes. However, the "fix" happened to break rendering
1928 that is always used resulting in corrupt image rendering in mozilla,
1929 evolution, and probably everything else that uses cairo.
1931 Fix to avoid BadMatch errors in cairo_surface_create_similar. These
1932 were introduced, (inadvertently, of course), as part of the fix in
1933 1.5.18 for creating similar surfaces without the Render
1934 extension. Again, thanks to mozilla, (and Vladimir Vukicevic in
1935 particular), for noticing our mistake.
1939 Correctly handle an in-error surface in
1940 cairo_surface_write_to_png. Previously this function would cause an
1941 assertion failure if you gave it a finished surface. Now it cleanly
1942 returns a CAIRO_STATUS_SURFACE_FINISHED result instead.
1944 Avoid potentially infinite wandering through memory inside
1945 _cairo_hull_prev_valid. Thanks to Jonathan Watt for noticing this
1948 https://bugzilla.mozilla.org/show_bug.cgi?id=306649#c21
1952 Fix generation of "soft" masks made by drawing to a similar surface
1953 and then calling cairo_mask_surface() with it.
1957 Fix for code that uses cairo_mask() on an intermediate surface which
1958 is later passed to cairo_mask_surface().
1960 Snapshot 1.5.18 (2008-04-05 Carl Worth <cworth@cworth.org>)
1961 ===========================================================
1962 This is the ninth snapshot in cairo's unstable 1.5 series. It comes
1963 just 4 days after the 1.5.16 snapshot. We had hoped to not need
1964 another snapshot before the final 1.6.0 release, but several critical
1965 bugs were found and fixed in the last few days, so we thought it
1966 important to let people test the fixes with this snapshot. See below
1971 The README now lists necessary dependencies.
1973 Various graphics state defaults are now documented, (source pattern is
1974 opaque black, line width is 2.0, line join is miter, line cap is butt,
1975 miter limit is 10.0, etc.).
1979 Several cleanups have been made along many error-path returns,
1980 (carefully propagating up the original error status values, cleaning
1981 up memory leaks during error recovery, etc.). This is yet another in
1982 Chris "ickle" Wilson's long series of error-handling cleanups during
1985 Avoid undesired clipping when drawing scaled surface patterns with
1990 Fix emission of 1-bit alpha masks in PDF output.
1992 Fix a bug that would cause glyphs to be misplaced along the Y axis:
1994 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=%23474136
1996 Originally, an issue about a crash, but later leading to the
1997 misplaced glyphs issue being discovered.
2001 Fix misplaced glyphs in cairo's PostScript output.
2003 This issue occurs when consecutive glyphs are placed far
2004 apart. This case is exercised by the new ft-show-glyphs-table test
2005 case, which was originally inspired by the Debian bug #23474136
2008 Fix more misplaced glyphs in cairo's PostScript output:
2010 The issue here showed up under very particular circumstance, (when
2011 converting a PDF file with a CFF font with CID Identity-H encoding
2012 and using glyph 0, (defined by the CFF specification as .notdef)
2013 as a space instead). More concretely, this problem appeared when
2014 converting the UbuntuDesktop.pdf file mentioned in this bug
2017 https://bugs.freedesktop.org/show_bug.cgi?id=15348#c3
2019 As usual with arcane font-encoding-specific bugs like this, many
2020 thanks to Adrian Johnson for his magical ability to dive into
2021 specifications and emerge almost instantaneously with fixes. And
2022 thanks to Sebastien Bacher for bringing the bug to our attention.
2026 Fix serious failure on X servers without the Render extension.
2028 Since the 1.5.14 snapshot (with support for PseudoColor visuals),
2029 any application attempting to create a "similar" xlib surface would
2030 fail on an X server without the Render extension. Thanks to
2031 Frederic Crozat for pointing out that cairo's test suite was
2032 entirely failing when run against Xvfb.
2034 Avoid crashing cairo-xlib applications for too-large glyphs
2036 Naively sending glyphs of any size to the X server will eventually
2037 violate the X limit on maximum request sizes. We now properly
2038 detect when a glyph would be too large and use existing fallbacks
2039 to render the glyph rather than trying to send it to the X server.
2041 Enable the buggy_repeat workaround for Xorg servers < 1.4
2043 We have determined that Xorg 1.3.0 (as packaged in Fedora 8 at
2044 least) has a bug that can result in an X server crash when cairo
2045 uses certain X Render repeat operations, (as exercised by cairo's
2046 extend-reflect test). We avoid this crash by using fallbacks
2047 whenever a repeating surface is needed for any Xorg server with a
2048 version less than 1.4. This is slower, but should prevent the
2051 (Meanwhile, there appears to be a separate bug where some X
2052 servers or specific X-server drivers will use random pixmap data
2053 when asked to draw a repeating surface. The buggy_repeat
2054 workaround would also avoid those problems, but we have not yet
2055 characterized whether the new "version < 1.4" is a good
2056 characterization of those problems or not.)
2060 Implement cairo_font_extents for this backend.
2062 The cairo-quartz-font implementation added in the 1.5.14 snapshot was
2063 entirely missing support for the cairo_font_extents function. Thanks to
2064 Richard Hult for pointing out this obvious shortcoming, (and obvious
2065 lack of coverage in our test suite):
2067 CGFont backend returns 0 font extents
2068 https://bugs.freedesktop.org/show_bug.cgi?id=15319
2070 Snapshot 1.5.16 (2008-04-01 Carl Worth <cworth@cworth.org>)
2071 ===========================================================
2072 This is the eighth snapshot in cairo's unstable 1.5 series. It comes
2073 less than two weeks after the 1.5.14 snapshot and it really is a
2074 legitimate snapshot, (in spite of sharing this date with that of many
2075 bogus announcements). The major change in this snapshot is that the
2076 cairo-quartz backend is now officially "supported", including new API
2077 to construct a font face from a CGFontRef . Also several bug fixes
2078 have been fixed in many backends. See below for details.
2082 Cairo now depends on pixman 0.10.0 which was recently released. The
2083 latest pixman release can always be found alongside cairo releases at:
2085 http://cairographics.org/releases
2087 Increase the precision of color stops for gradients. This fixes a
2088 regression in gradient rendering that had been present since the
2091 paginated (all of ps, pdf, svg, and win32-printing)
2092 ---------------------------------------------------
2093 Fix assertion failure when some drawing elements are outside the page
2094 boundaries, (this bug was noticed when using Inkscape to print a
2095 drawing with landscape orientation to a portrait-oriented piece of
2100 Fix of bug causing incorrect glyph positioning.
2102 Fix handling of CAIRO_OPERATOR_SOURCE.
2106 More reduction of unnecessary digits of precision in PDF output.
2108 Fix handling of CAIRO_OPERATOR_SOURCE.
2112 Fix bug in usage of libpng that was preventing cairo_mask from working
2113 with the svg backend.
2115 Fix transformation of source pattern for cairo_stroke().
2117 cairo-win32-printing
2118 --------------------
2119 Fix fallback resolution, (thanks again to inkscape users/developers
2120 for helping us find this one).
2124 Mark the cairo-quartz backend as "supported" rather than
2125 "experimental". This means the following:
2127 * The backend will now be built by default (if possible).
2129 * We are committing that the backend-specific API (as published in
2130 cairo-quartz.h) are stable and will be supported in all future
2133 * We are committing that the output quality of this backend
2134 compares favorably with other cairo backends, (and that quality
2135 is ensured by good results from the cairo test suite).
2137 * We recommend that distributions build and distribute this
2138 backend when possible.
2140 Note that the cairo_quartz_image API (in cairo-quartz-image.h) is
2141 still experimental, will not build by default, (pass
2142 --enable-quartz-image to configure to build it), and may see API
2143 changes before it is marked as "supported" in a future release.
2145 Put the CAIRO_FONT_TYPE_ATSUI name back into
2146 cairo-deprecated.h. Without this, the cairo 1.5.14 snapshot broke all
2147 builds for applications using the C++ cairomm bindings (and perhaps
2148 others) which have the CAIRO_FONT_TYPE_ATSUI name in their header
2149 files. This breakage happened even for applications not using
2150 cairo-quartz at all.
2152 Note: Even though the CAIRO_FONT_TYPE_ATSUI name is provided to
2153 avoid this build breakage, we still recommend that bindings and
2154 applications move to the new, and more accurate,
2155 CAIRO_FONT_TYPE_QUARTZ name.
2157 Replace the implementation of cairo-quartz-font to use CFFont instead
2158 of ATSUI. The CGFont API is a better fit than ATSUI, and this new
2159 implementation is also more correct than the old one as well.
2161 This also adds the following new API call:
2163 cairo_public cairo_font_face_t *
2164 cairo_quartz_font_face_create_for_cgfont (CGFontRef font);
2166 The previous cairo_quartz_font_face_create_for_atsu_font_id function
2167 continues to exist and is part of the supported API going
2168 forward. (However, the old name of that same function, which was
2169 cairo_atsui_font_face_create_for_atsu_font_id is officially
2170 deprecated. Any source code using the old name should be updated to
2173 Fix transformation of source pattern for cairo_stroke().
2177 Avoid crash in create_similar is cairo_win32_surface_create fails.
2179 Snapshot 1.5.14 (2008-03-20 Carl Worth <cworth@cworth.org>)
2180 ===========================================================
2181 This is the seventh snapshot in cairo's unstable 1.5 series. It comes
2182 3 weeks after the 1.5.12 snapshot. This snapshot includes support for
2183 arbitrary X server visuals, (including PseudoColor), which was the
2184 final remaining cairo-specific item on the cairo 1.6 roadmap. It also
2185 includes a huge number of improvements to the cairo-quartz backend. So
2186 this is effectively a cairo 1.6 release candidate. We expect very few
2187 changes from now until 1.6 and only for specific bug fixes.
2191 Rename ATSUI font backend to Quartz font backend. This affects the
2194 --enable-atsui -> --enable-quartz-font
2195 CAIRO_HAS_ATSUI_FONT -> CAIRO_HAS_QUARTZ_FONT
2196 CAIRO_FONT_TYPE_ATSUI -> CAIRO_FONT_TYPE_QUARTZ
2198 cairo_atsui_font_face_create_for_atsu_font_id ->
2199 cairo_quartz_font_font_create_for_atsu_font_id
2201 This API change is justified by the cairo-quartz backend still be
2202 marked as "experimental" rather than "supported", (though this is one
2203 step toward making the change to "supported" before 1.6). Cairo will
2204 still provide ABI compatibility with the old symbol name, however.
2206 paginated (all of ps, pdf, svg, and win32-printing)
2207 ---------------------------------------------------
2208 Optimize by not analyzing an image surface for transparency more than
2209 once, (previously all images were analyzed twice).
2211 cairo-ps and cairo-pdf
2212 ----------------------
2213 Avoiding emitting a matrix into the stroke output when unnecessary,
2214 (making output size more efficient).
2216 Reduce rounding error of path shapes by factoring large scale factors
2217 out of the path matrix, (ensuring that a fixed-number of printed
2218 digits for path coordinates contains as much information as possible).
2220 Reduce excess digits for text position coordinates. This makes the
2221 output file size much smaller without making the result any less
2226 Eliminate bug causing extraneous text repetition on Linux PostScript
2227 output in some cases.
2229 See: Mozilla Bug 419917 – Printed page contents are reflected
2230 inside bordered tables (Linux-only)
2232 https://bugzilla.mozilla.org/show_bug.cgi?id=419917
2234 Optimize output when EXTEND_PAD is used.
2238 Fix to not use fill-stroke operator with transparent fill, (else PDF
2239 output doesn't match the cairo-defined correct result). See:
2241 https://bugs.launchpad.net/inkscape/+bug/202096
2245 Fix stroke of path with a non-solid-color source pattern:
2247 http://bugs.freedesktop.org/show_bug.cgi?id=14556
2251 Fix text rendering with gradient or image source pattern.
2253 Handling antialiasing correctly for cairo_stroke(), cairo_clip(), and
2254 cairo_show_text()/cairo_show_glyphs().
2256 Correctly handle gradients with non-identity transformations:
2258 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=14248
2260 Add native implementation of REPEAT and REFLECT extend modes for
2263 Fix implementation for the "unbounded" operators, (CAIRO_OPERATOR_OUT,
2264 _IN, _DEST_IN, and _DEST_ATOP).
2266 Correctly handle endiannees in multi-architecture compiles on Mac OS
2269 Avoid behavior which would cause Core Graphics to print warnings to
2270 the console in some cases.
2274 Fix handling of miter limit.
2276 cairo-win32-printing
2277 --------------------
2278 Fix to not use a 1bpp temporary surface in some cases while printing,
2279 (so grayscale data is preserved rather than just becoming black and
2284 Add support for rendering to arbitrary TrueColor X server
2285 visuals. This fixes at least the following bugs:
2287 cairo doesn't support 8-bit truecolor visuals
2288 https://bugs.freedesktop.org/show_bug.cgi?id=7735
2290 cairo doesn't support 655 xlib format
2291 https://bugs.freedesktop.org/show_bug.cgi?id=9719
2293 Add support for rendering to 8-bit PseudoColor X server visuals. This
2294 fixes the following bug:
2296 Cairo doesn't support 8-bit pseudocolor visuals
2297 https://bugs.freedesktop.org/show_bug.cgi?id=4945
2299 Snapshot 1.5.12 (2008-02-28 Carl Worth <cworth@cworth.org>)
2300 ===========================================================
2301 This is the sixth snapshot in cairo's unstable 1.5 series. It comes 1
2302 week after the 1.5.10 snapshot. This snapshot includes the
2303 long-awaited change from 16.16 to 24.8 fixed-point values, (see below
2304 for why you should care). It also includes several backend-specific
2307 24.8 fixed-point format
2308 -----------------------
2309 Cairo has always converted path coordinates to a fixed-point
2310 representation very early in its processing. Historically, this has
2311 been a 32-bit representation with 16 bits of integer for the
2312 device-pixel grid and 16 bits of sub-pixel positioning. The choice of
2313 16 bits for the integer coordinate space was based on the 16-bit limit
2314 for X Window drawables.
2316 This 16-bit limit has proven problematic for many applications. It's
2317 an especially vexing problem when targeting non-X backends that don't
2318 have any 16-bit restriction. But even when targeting cairo-xlib, it's
2319 often desirable to draw a large shape, (say a background rectangle),
2320 that extends beyond the surface bounds and expect it to fill the
2321 surface completely, (rather than overflowing and triggering random
2324 Meanwhile, nobody has ever really needed 16 bits of sub-pixel
2327 With this snapshot, the fixed-point system is still in place and is
2328 still using a 32-bit representation, (future versions of cairo might
2329 move entirely to floating-point when targeting PDF output for
2330 example). But the representation now provides 24 bits of pixel
2331 addressing and only 8 bits of sub-pixel positioning. This should give
2332 a much less stifling space to many applications.
2334 However, the underlying pixman library still has 16-bit limitations in
2335 many places, (it has its roots in the X server as well). Until those
2336 are also fixed, applications targeting cairo image surfaces, or
2337 hitting software fallbacks when targeting other surfaces will still
2338 encounter problems with device-space values needing more than 16
2343 Add a few tests to the test suite to increase coverage.
2345 Cleanup a few error-handling paths, (propagate error correctly).
2349 Fix handling of font sizes smaller than 1 device pixel.
2353 Fix to properly save/restore clip when analyzing meta-surface
2354 patterns, (fixing a couple of test-suite failures).
2356 Implement native support for CAIRO_OPERATOR_SOURCE when the source
2359 Emit rectangles as PDF rectangles ("re" operator) rather than as
2364 Fix to work properly with the 16.16->24.8 change.
2368 Fix CAIRO_EXTEND_REFLECT by using an image fallback, (there's no
2369 direct SVG support for reflected patterns).
2371 Fix the use of alpha-only masks, (such as CAIRO_FORMAT_A8).
2375 Add new API for efficiently using image data as a source:
2378 cairo_quartz_image_surface_create (cairo_surface_t *image_surface);
2381 cairo_quartz_image_surface_get_image (cairo_surface_t *surface);
2383 For full documentation, see:
2385 http://cairographics.org/manual/cairo-Quartz-Surfaces.html#cairo-quartz-image-surface-create
2387 Several fixes for cairo_mask().
2391 Change default from from Monaco to Helvetica to be more consistent
2392 with other font backends.
2394 Snapshot 1.5.10 (2008-02-20 Carl Worth <cworth@cworth.org>)
2395 ===========================================================
2396 This is the fifth snapshot in cairo's unstable 1.5 series. It comes 3
2397 weeks after the 1.5.8 snapshot. This snapshot adds one new API
2398 function, (cairo_has_current_point), and the usual mix of
2399 improvements, (more efficient PostScript/PDF output, optimized
2400 stroking), and fixes (more robust error-handling, etc.). See below for
2405 Add a new function to query if there is a current point:
2408 cairo_has_current_point (cairo_t *cr);
2410 There is no current point immediately after cairo_create(), nor after
2411 cairo_new_path() or cairo_new_sub_path(). There is a current point
2412 after any of the path-creation functions, (cairo_move_to,
2413 cairo_line_to, cairo_curve_to, etc.).
2415 With this new function, we also revert the change of the return type
2416 of cairo_get_current_point from cairo 1.5.8, (it's now a void function
2421 Optimize stroking code to avoid repeated calculation of redundant
2422 values, (particularly significant for very large, offscreen paths).
2426 Patch a few more potential buffer overruns, (due to integer
2429 Many fixes and improvements to cairo's error-handling, (ensure that
2430 correct error values are returned, clean up memory leaks on
2431 error-handling paths, etc.).
2433 Fix a potential infinite loop when stroking a spline with a pen that
2434 has been transformed to a line segment.
2436 Remove treating NULL as a synonym for a valid cairo_font_options_t*
2437 with default values, (a change that had been introduced as of cairo
2440 Remove the altered handling of tolerance and fallback-resolution that
2441 had been introduced as of cairo 1.5.4.
2445 Pass the original Drawable, (as opposed to the root window), to
2446 XCreatePixmap when creating a similar surface. This gives the X server
2447 more information so that it can be clever and efficient.
2451 Fix the rendering of repeating and reflecting patterns.
2453 Ensure miter limit is always >= 1, (smaller limits are not meaningful,
2454 but they can cause some PDF viewers to fail to display pages).
2456 Generate more efficient output when the same path is used for both
2461 Start sharing much of the cairo-pdf code rather than implementing very
2462 similar code in cairo-ps.
2464 Implement native support for repeating and reflecting linear
2467 Implement reflected surface patterns.
2469 Ensure miter limit is always >= 1, (smaller limits are not meaningful,
2470 but they can cause some PostScript viewers to crash).
2472 Generate PostScript that will perform more efficiently and use less
2473 memory on printers, (use currentfile instead of a giant string array
2474 for image data, and avoid using PostScript patterns for paint() and
2475 fill() when possible).
2479 Avoid unnecessary rasterization when copying a "similar" surface to
2480 another svg surface, (allow the SOURCE operator to be implemented with
2481 all-vector operations if there are no underlying objects).
2485 Eliminate infinite loop when attempting to render an empty string.
2487 Snapshot 1.5.8 (2008-01-30 Carl Worth <cworth@cworth.org>)
2488 ==========================================================
2489 This is the fourth snapshot in cairo's unstable 1.5 series. It comes 2
2490 weeks after the 1.5.6 snapshot. It adds a few new API functions. Most
2491 notably all callers of cairo_image_surface_create_for_data should now
2492 be calling cairo_format_stride_for_width to compute a legal stride
2493 value. See below for more details.
2495 New API in cairo 1.5.8
2496 ----------------------
2497 We've added a new function that should be called to compute a legal
2498 stride value before allocating data to be used with
2499 cairo_image_surface_create_for_data:
2502 cairo_format_stride_for_width (cairo_format_t format,
2505 We've also added a new cairo_path_extents function that can be used to
2506 compute a bounding box for geometry such as a single line segment,
2507 (contrast with cairo_path_extents and cairo_stroke_extents):
2510 cairo_path_extents (cairo_t *cr,
2511 double *x1, double *y1,
2512 double *x2, double *y2);
2514 And finally, we've added a function to allow for querying the
2515 XRenderPictFormat of a cairo-xlib surface:
2518 cairo_xlib_surface_get_xrender_format (cairo_surface_t *surface);
2522 Fix return types of cairo_surface_show_page and
2523 cairo_surface_copy_page. This is an API change to functions that are
2524 new in the 1.5 series, so not an API break compared to any stable
2525 cairo release, (1.0.x, 1.2.x, 1.4.x).
2527 Change the return type of cairo_get_current_point() from void to
2528 cairo_status_t. This allows the caller to receive a
2529 CAIRO_STATUS_NO_CURRENT_POINT value to distinguish the a current point
2530 at the origin from no current point existing.
2532 Performance improvement
2533 -----------------------
2534 Improve performance of clipping by using an optimized code path
2535 internally, (with the ADD operator instead of IN).
2539 Fix various cairo_*_extents functions to initialize the return-value
2540 variables even in the case of a cairo_t in error.
2542 Treat NULL as a legitimate value for cairo_font_options_t*. [NOTE:
2543 On discussion afterwards, we decided against this change so it has
2544 been removed as of cairo 1.5.10.]
2546 Fix rendering with CAIRO_ANTIALIAS_NONE to be more predictable, (that
2547 is, to avoid seams appearing when geometry and imagery share an
2548 identical edge). Portions of this fix are in the pixman library and
2549 will appear in a future release of that library.
2551 Avoid triggering an error for a font size of 0.
2553 Miscellaneous changes
2554 ---------------------
2555 Require pixman >= 0.9.6.
2557 There has been a tremendous amount improvement to cairo's
2558 documentation. We're delighted that 100% of the public API has at
2559 least some documentation in the API reference manual. Many thanks to
2560 Behdad Esfahbod and Nis Martensen for leading this effort.
2562 cairo-pdf and cairo-ps
2563 ----------------------
2564 Eliminate failure when a Type 1 font is embedded with an explicit
2569 Implement a more correct and more efficient approach for patterns with
2570 an extend mode of CAIRO_EXTEND_REFLECT.
2574 Fix image masks to properly pack and pad mask bits.
2578 Take care to only use DrawTiledImage for integer-aligned images, (and
2579 use slower paths to get the correct result in other cases).
2583 Fix for older versions of mingw.
2585 Improve the handling of the clipping with the win32 and win32-printing
2588 Fix rendering of non black/white text.
2590 Snapshot 1.5.6 (2008-01-15 Carl Worth <cworth@cworth.org>)
2591 ==========================================================
2592 This is the third snapshot in cairo's unstable 1.5 series. It comes
2593 about 6 weeks after the 1.5.4 snapshot. The only API addition compared
2594 to 1.5.4 is very minor, (a new value CAIRO_STATUS_TEMP_FILE_ERROR).
2595 The remainder of the changes are the usual accumulation of bug fixes
2596 and improvements. See below for details.
2600 Fix handling of fonts that contain a mixture of outline and bitmapped
2601 glyphs. There was a change in this handling in 1.5.4 that improved
2602 some cases and also regressed other cases. Now, all cases should be
2605 Fix alignment issues that were causing SIGBUS failures on SPARC.
2607 Fix a regression (which first appeared in 1.5.2) where stroking under
2608 a large scale would sometimes incorrectly replace a miter join with a
2609 bevel join. (Thanks to Keith Packard.)
2611 Fix reporting of zero-sized extents to be {0,0} rather than
2612 {INT_MAX,INT_MIN}. This avoids several integer overflow and
2613 allocations of massive regions in some cases.
2615 Fix failures of gradients with no stops, (quartz, ps, and pdf).
2617 Fix handling of Type 1 fonts on Windows platforms.
2619 Fix handling of Type 1 fonts with no specific family name in the font
2620 itself, (generate a CairoFont-x-y name).
2622 Handle NULL string values in cairo_show_text, cairo_show_glyphs, and
2625 Many robustness improvements along error-handling paths, (thanks as
2626 always, to Chris "ickle" Wilson).
2628 Various other minor fixes.
2630 Paginated backends (PDF/PostScript/win32-printing)
2631 --------------------------------------------------
2632 Avoid unnecessary rasterization when using a paginated surface as a
2633 source, (such as drawing from one pdf surface to another).
2635 Fix replaying of paginated surface with more than one level of push/pop
2640 Fix xlib backend to not consider recent X server release as having a
2641 buggy repeat implementation in the Render extension.
2645 Fix PDF output to avoid triggering very slow rendering in PDF viewers,
2646 (avoid starting and stopping the content stream for each pattern
2649 Support CAIRO_OPERATOR_SOURCE in cases where there is nothing below
2650 the object being drawn.
2652 Fix to avoid seams appearing between multiple fallback regions.
2654 cairo-ps (PostScript)
2655 ---------------------
2656 Use correct bounding box in Type 3 fonts.
2658 Fix several bugs in cairo's PostScript output. These include making
2659 the PostScript output more compatible with recent versions of
2660 ghostscript that are more strict about Type 3 fonts, for
2663 Fix for win32 to not attempt to create temporary files in the root
2664 directory, (where the user may not have write permission).
2666 Avoid generating Level 3 PostScript if Level 2 is sufficient. Also,
2667 add code in output documents to alert the user if Level 3 PostScript
2668 is handed to a device that cannot handle PostScript beyond Level
2673 Various performance optimizations.
2675 Fixed support for small surfaces (less than 8x8).
2677 Provide support for environment variables CAIRO_DIRECTFB_NO_ACCEL to
2678 disable acceleration and CAIRO_DIRECTFB_ARGB_FONT to enable ARGB fonts
2683 Allow OS/2 APIs instead of C library allocation functions.
2685 Snapshot 1.5.4 (2007-12-05 Carl Worth <cworth@cworth.org>)
2686 ==========================================================
2687 This is the second snapshot in cairo's unstable 1.5 series. It comes
2688 just over 1 month after the 1.5.2 snapshot. There are no API changes
2689 or additions in 1.5.4 compared to 1.5.2, but there are several bug
2690 fixes, and some optimizations. Most of these apply to particular
2691 backends. See below for details.
2693 General improvements
2694 --------------------
2695 Use less memory for spline approximation calculations.
2697 Change how the tolerance value is interpreted with regard to
2698 fallback-resolution. [Note: On further discussion, we decided against
2699 this change for now. It is removed as of cairo 1.5.10.]
2701 Fix precision of floating-point values in vector-output backends to
2702 avoid rounding errors with very small numbers.
2706 Fix bug in glyph rendering with xlib, (due to everything being clipped
2707 out). This was a regression in the 1.5.2 snapshot that was visible in
2708 the GIMP, for example. See:
2710 cairo 1.5.2 causes font problems in GIMP 2.4 status bar and evolution 2.12.1
2711 https://bugs.freedesktop.org/show_bug.cgi?id=13084
2713 PostScript improvements
2714 -----------------------
2715 Fix bug leading to invalid PostScript files when rendering
2716 text, (need "0 0 xyshow" instead of "0 xyshow").
2718 Fix many issues with Type 3 fonts, including making the resulting text
2723 Fix font metrics height value for ATSUI, (helps webkit on GTK+ OS X
2728 Fix EXTEND_NONE mode for patterns.
2730 Fix cairo_quartz_surface_create to properly clear the new surface
2731 in cairo_quartz_surface_create.
2733 Fix to correctly handle 0x0 sized surfaces.
2735 Optimize drawing of EXTEND_REPEAT patterns for OS X 10.5.
2737 Snapshot 1.5.2 (2007-10-30 Carl Worth <cworth@cworth.org>)
2738 ==========================================================
2739 This is the first snapshot in cairo's unstable 1.5 series. It comes 4
2740 months after the 1.4.10 release. This snapshot includes significant
2741 improvements to PDF and PostScript output, which is one of the things
2742 in which we're most interested in getting feedback. There are a couple
2743 of minor API additions, and several optimizations, (primarily in the
2744 "print/vector" backends). And there are dozens of bug fixes and
2745 robustness improvements.
2747 New dependency on external pixman library
2748 -----------------------------------------
2749 A significant change in this snapshot compared to all previous cairo
2750 releases is that cairo now depends on an external "pixman" library for
2751 its software rendering. Previously this same code was compiled
2752 internally as part of cairo, but now the code is separate so that both
2753 cairo and the X server can now share common code, (thanks very much to
2754 Søren Sandmann for his work on separating pixman and maintaining it).
2756 So users will need to acquire and build pixman before being able to
2757 build cairo. The current release is 0.9.6 and can be obtained from
2760 http://cairographics.org/releases/pixman-0.9.6.tar.gz
2762 which can be verified with:
2764 http://cairographics.org/releases/pixman-0.9.6.tar.gz.sha1
2765 66f01a682c64403a3d7a855ba5aa609ed93bcb9e pixman-0.9.6.tar.gz
2767 http://cairographics.org/releases/pixman-0.9.6.tar.gz.sha1.asc
2768 (signed by Carl Worth)
2770 Major PDF/PostScript improvements
2771 ---------------------------------
2772 Adrian Johnson has done some long-awaited work to make cairo's PDF and
2773 PostScript output more interesting than ever before. First, many
2774 operations that previously triggered image fallbacks will now be
2775 rendered as native vectors. These operations include:
2777 PDF: cairo_push_group, cairo_surface_create_similar,
2778 cairo_mask, A8/A1 surface sources, repeating/reflecting linear
2781 PostScript: cairo_push_group, cairo_surface_create_similar,
2782 gradients, bilevel alpha masks, (for example, all values either 0 or
2783 255 for an A8 mask).
2785 Not only that, but when an image fallback is required, it will now be
2786 limited to only the necessary region. For example, a tiny translucent
2787 image overlaying a small portion of text would previously caused an
2788 entire PostScript page to be rendered as a giant image. Now, the
2789 majority of that page will be nice text, and there will only be a tiny
2790 image in the output.
2792 Additionally, the PostScript output now carefully encodes text so that
2793 if it is subsequently converted to PDF, the text will be
2796 This is very exciting progress, and we're hoping to hear from users
2797 during the 1.5 series about how things have improved, (for example,
2798 inkscape users doing cairo-based PDF export: please let us know how
2799 things look). And feel free to pass your thanks along to Adrian for his excellent work.
2801 NOTE: This much improved PDF output makes more sophisticated use of
2802 functionality in the PDF specification. This means that cairo's output
2803 will sometimes expose bugs in some free software PDF viewers, (evince,
2804 poppler, and xpdf, for example), that are not yet ready for such PDF
2805 files. We're working with the poppler maintainers to get these bugs
2806 fixed as quickly as possible. In the meantime, please double-check
2807 with other PDF viewers if cairo-generated PDF files are not being
2808 rendered correctly. It may be due to a bug in the viewer rather than
2809 in the PDF file that cairo has created.
2811 Robustness improvements
2812 -----------------------
2813 Chris Wilson has made the largest contribution by far to cairo 1.5.2,
2814 (in number of commits). His more than 150 commits include a huge
2815 number of fixes to increase cairo's robustness. These fixes make cairo
2816 more robust against invalid and degenerate input, (NaN, empty path,
2817 etc.), against size-0 malloc calls, against memory leaks on
2818 error-recovery paths, and against other failures during error
2819 handling. He also implemented atomic operations to cairo, and used
2820 them to fix cairo's previously non-thread-safe reference counting,
2821 again improving robustness.
2823 Chris has put a tremendous amount of time and effort into writing
2824 analysis tools for this work, and in running those tools and fixing
2825 the problems they report. We're very grateful for this work, and hope
2826 that all cairo users appreciate the more robust implementation that
2829 This work is largely thankless, so it might make sense to notice
2830 sometime that cairo has been running quite smoothly for you, and when
2831 you do, send a quick "thank you" off to Chris Wilson, since it
2832 is all definitely running smoother thanks to his work.
2836 There are no major additions to cairo's core API. The only new,
2837 generic functions are:
2840 cairo_surface_copy_page (cairo_surface_t *surface);
2843 cairo_surface_show_page (cairo_surface_t *surface);
2845 which can now be used much more conveniently than the existing
2846 cairo_copy_page and cairo_show_page functions in some
2847 situations. These functions act identically, but require only a
2848 cairo_surface_t* and not a cairo_t*.
2850 All other API additions are specific to particular backends.
2852 New cairo-win32 API (new font face function and "win32 printing" surface)
2853 -------------------------------------------------------------------------
2854 There is a new function for creating a win32 font face for both a
2855 logfontw and an hfont together. This complements the existing
2856 functions for creating a font face from one or the other:
2859 cairo_win32_font_face_create_for_logfontw_hfont (LOGFONTW *logfont,
2862 There is also a new "win32 printing" surface:
2865 cairo_win32_printing_surface_create (HDC hdc);
2867 This interface looks identical to the original
2868 cairo_win32_surface_create, (both accept and HDC), but the behavior of
2869 this new surface is very different. It should only be called with a
2870 printing DC, and will result in all drawing commands being stored into
2871 a meta-surface and emitted after each page is complete, with analysis
2872 to do as minimal image-based fallbacks as necessary. The behavior and
2873 implementation shares much with the PDF and PostScript backends.
2875 New cairo-ps API (EPS and PostScript level control)
2876 ---------------------------------------------------
2877 An often requested feature has been the ability to generate
2878 Encapsulated PostScript (EPS) with cairo. We have that now with the
2879 following very simple API. Just do cairo_ps_surface_create as usual
2880 then call this function with a true value:
2883 cairo_ps_surface_set_eps (cairo_surface_t *surface,
2886 [NOTE: As always with snapshots, it's possible---though not very
2887 likely---that the API could still be modified before a final
2888 release. For example, this is the first public cairo function that
2889 accepts a Boolean parameter. I'm generally opposed to Boolean
2890 parameters, but this is probably the one case where I'm willing to
2891 accept one, (namely a "set" function that accepts a single Boolean).]
2893 Also, it is now possible to control what PostScript level to target,
2894 (either level 2 or level 3), with the following new API:
2896 typedef enum _cairo_ps_level {
2902 cairo_ps_surface_restrict_to_level (cairo_surface_t *surface,
2903 cairo_ps_level_t level);
2906 cairo_ps_get_levels (cairo_ps_level_t const **levels,
2910 cairo_ps_level_to_string (cairo_ps_level_t level);
2912 Improvement for cairo-quartz
2913 ----------------------------
2914 Brian Ewins had contributed several improvements to cairo-quartz. These
2915 include an implementation of EXTEND_NONE for linear and radial
2916 gradients, (so this extend mode will no longer trigger image fallbacks
2917 for these gradients), as well as native surface-mask clipping, (only
2918 on OS X 10.4+ where the CGContextClipToMask function is available).
2920 He also fixed a semantic mismatch between cairo and quartz for dashing
2921 with an odd number of entries in the dash array.
2923 We're grateful for Brian since not many quartz-specific improvements
2924 to cairo would be happening without him.
2928 Optimize SVG output for when the same path is both filled and stroked,
2929 and avoid unnecessary identity matrix in SVG output. (Emmanuel Pacaud).
2931 Optimize PS output to take less space (Ken Herron).
2933 Make PS output more compliant with DSC recommendations (avoid initclip
2934 and copy_page) (Adrian Johnson).
2936 Make PDF output more compact (Adrian Johnson).
2938 Release glyph surfaces after uploading them to the X server, (should
2939 save some memory for many xlib-using cairo application). (Behdad
2942 Optimize cairo-win32 to use fewer GDI objects (Vladimir Vukicevic).
2944 win32-printing: Avoid falling back to images when alpha == 255
2945 everywhere. (Adrian Johnson).
2947 win32-printing: Avoid falling back for cairo_push_group and
2948 cairo_surface_create_similar. (Adrian Johnson)
2952 Avoid potential integer overflows when allocating large buffers
2953 (Vladimir Vukicevic).
2955 Preparations to allow the 16.16 fixed-point format to change to
2956 24.8 (Vladimir Vukicevic).
2958 Fix bugs for unsupported X server visuals (rgb565, rgb555, bgr888, and
2959 abgr8888). (Carl Worth and Vladimir Vukicevic)
2961 Fix bugs in PDF gradients (Adrian Johnson).
2963 Fix cairo-xlib to build without requiring Xrender header
2964 files (Behdad Esfahbod).
2966 Make cairo more resilient in the case of glyphs not being available in
2967 the current font. (Behdad Esfahbod)
2969 Prevent crashes when both atsui and ft font backends are compiled in
2972 Make font subsetting code more robust against fonts that don't include
2973 optional tables (Adrian Johnson).
2975 Fix CFF subsetting bug, (which manifested by generating PDF files that
2976 Apple's Preview viewer could not read) (Adrian Johnson).
2978 Fixed error handling for quartz and ATSUI backends (Brian Ewins).
2980 Avoid rounding problems by pre-transforming to avoid integer-only
2981 restrictions on transformation in GDI (Adrian Johnson).
2983 Fixed an obscure bug (#7245) computing extents for some stroked
2986 Fix crashes due to extreme transformation of the pen, (seems to show
2987 up in many .swf files for some reason) (Carl Worth).
2989 Release 1.4.10 (2007-06-27 Carl Worth <cworth@cworth.org>)
2990 ==========================================================
2991 This is the fifth update in cairo's stable 1.4 series. It comes
2992 roughly three weeks after the 1.4.8 release. The most significant
2993 change in this release is a fix to avoid an X error in certain cases,
2994 (that were causing OpenOffice.org to crash in Fedora). There is also a
2995 semantic change to include child window contents when using an xlib
2996 surface as a source, an optimization when drawing many rectangles, and
2997 several minor fixes.
2999 Eliminate X errors that were killing OO.o (Chris Wilson)
3000 --------------------------------------------------------
3001 Cairo is fixed to avoid the X errors propagated when cleaning up
3002 Render Pictures after the application had already destroyed the
3003 Drawable they reference. (It would be nice if the X server wouldn't
3004 complain that some cleanup work is already done, but there you have
3005 it.) This fixes the bug causing OpenOffice.org to crash as described
3008 XError on right click menus in OOo.
3009 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=243811
3011 Use IncludeInferiors when using xlib surface as a source (Ryan Lortie)
3012 ----------------------------------------------------------------------
3013 When an xlib surface is used as the source of a draw operation the
3014 contents of child windows are now included in the source data. The
3015 semantics of drawing to xlib surfaces are unchanged (ie: draws are
3016 still clipped by child windows overlapping the destination window).
3018 Optimize drawing of many rectangles (Vladimir Vukicevic)
3019 --------------------------------------------------------
3020 Avoid O(N*N) loop when filling many axis-aligned rectangles, (either
3021 many rectangles as separate sub-paths or due to dashing).
3025 Fix cairo-perf on Solaris by linking to librt. (Behdad Esfahbod)
3027 Fix make check for systems that require executable files to have a
3028 particular extension. (Behdad Esfahbod)
3030 Eliminate some warnings in cairo-quartz. (Brian Ewins)
3032 Fix build-breaking typo for cairo-directfb. (Chris Wilson)
3034 Release 1.4.8 (2007-06-07 Carl Worth <cworth@cworth.org>)
3035 =========================================================
3036 This is the fourth update in cairo's stable 1.4 series. It comes just
3037 over five weeks after the 1.4.6 release. This release includes a
3038 thread-safe surface-cache for solid patterns which significantly
3039 improves text rendering with the xlib backend. Also, dozens of error
3040 paths in cairo have been fixed thanks to extensive fault-injection
3041 testing by Chris Wilson.
3043 Surface cache for solid patterns
3044 --------------------------------
3045 Originally written by Jorn Baayen, the introduction of a small cache
3046 for surfaces created for solid patterns improves performance
3047 dramatically. For example, this reduces the volume of X requests
3048 during text rendering to the same level as Xft.
3050 This cache first made its appearance in a 1.3.x snapshot, but was
3051 removed before appearing in any previous major release due to
3052 complications with multi-threaded programs. For example, programs like
3053 evince that would carefully restrict usage of cairo-xlib to a single
3054 thread were unpleasantly surprised to find that using cairo-image in a
3055 separate thread could trigger X requests.
3057 Behdad Esfahbod designed a fix which was implemented by Chris
3058 Wilson. Now, the necessary X requests are queued up until the next
3059 time the application directly operates on an xlib surface.
3061 Improved error handling paths
3062 ------------------------------
3063 Chris Wilson continued the excellent work he started in cairo 1.4.4 to
3064 make cairo much more robust against out-of-memory and other errors. He
3065 applied his memory allocation fault injection cairo's main test suite,
3066 (previously he had applied it to cairo's performance suite).
3068 Chris's testing found dozens of bugs which he fixed. Many of these
3069 bugs had perhaps never been hit by any users. But at least one was
3070 hit by the gnome-about program which resulted in dozens of duplicated
3071 bug reports against that program:
3073 http://bugzilla.gnome.org/show_bug.cgi?id=431990
3075 We were very pleasantly surprised to see this bug get fixed as a
3076 side-effect of Chris's work. Well done, Chris!
3080 Cleanup of mutex declarations (Behdad Esfahbod)
3082 Remove unnecessary clip region from SVG output (Emmanuel Pacaud)
3084 Remove Xsun from the buggy_repeat blacklist (Elaine Xiong)
3086 ATSUI: Fix glyph measurement: faster and more correct (Brian Ewins)
3088 Quartz: fixed 'extend' behaviour for patterns, improved pattern performance,
3089 and a few smaller correctness fixes. (Brian Ewins, Vladimir Vukicevic)
3091 Release 1.4.6 (2007-05-01 Carl Worth <cworth@cworth.org>)
3092 =========================================================
3093 This is the third update in cairo's stable 1.4 series. It comes a
3094 little less than three weeks since the 1.4.4 release. This release
3095 fixes the broken mutex initialization that made cairo 1.4.4 unusable
3096 on win32, OS/2, and BeOS systems. This release also adds significant
3097 improvements to cairo's PDF backend, (native gradients!), and a couple
3098 of performance optimizations, (one of which is very significant for
3099 users of the xlib backend). See below for more details.
3101 Repaired mutex initialization
3102 -----------------------------
3103 We apologize that cairo 1.4.4 did little more than crash on many
3104 platforms which are less-frequently used by the most regular cairo
3105 maintainers, (win32, OS/2, and BeOS). The mutex initialization
3106 problems that caused those crashes should be fixed now. And to avoid
3107 similar problems in the future, we've now started posting pre-release
3108 snapshots to get better testing, (subscribe to cairo@cairographics.org
3109 if you're interested in getting notified of those and testing them).
3113 Thanks to Adrian Johnson, (cairo PDF hacker extraordinaire), we have
3114 several improvements to cairo's PDF backend to announce:
3118 As of cairo 1.4.6, cairo will now generate native PDF gradients in
3119 many cases, (previously, the presence of a gradient on any page
3120 would force rasterized output for that page). Currently, only
3121 gradients with extend types of PAD (the default) or NONE will
3122 generate native PDF gradients---others will still trigger
3123 rasterization, (but look for support for other extend modes in a
3124 future release). Many thanks to Miklós Erdélyi as well, who did the
3125 initial work for this support.
3127 Better compatibility with PDF viewers:
3129 The PDF output from cairo should now be displayed correctly by a
3130 wider range of PDF viewers. Adrian tested cairo's PDF output against
3131 many PDF viewers, identified a common bug in many of those viewers
3132 (ignoring the CTM matrix in some cases), and modified cairo's output
3133 to avoid triggering that bugs (pre-transforming coordinates and
3134 using an identity matrix).
3136 Better OpenType/CFF subsetting:
3138 Cairo will now embed CFF and TrueType fonts as CID fonts.
3140 Performance optimizations
3141 -------------------------
3142 Faster cairo_paint_with_alpha:
3144 The cairo_paint_with_alpha call is used to apply a uniform alpha
3145 mask to a pattern. For example, it can be used to gradually fade an
3146 image out or in. Jeff Muizelaar fixed some missing/broken
3147 optimizations within the implementation of this function resulting
3148 in cairo_paint_with_alpha being up to 4 times faster when using
3149 cairo's image backend.
3151 Optimize rendering of "off-screen" geometry:
3153 Something that applications often do is to ask cairo to render
3154 things that are either partially or wholly outside the current clip
3155 region. Since 1.4.0 the image backend has been fixed to not waste
3156 too much time in this case. But other backends have still been
3159 In particular, the xlib backend has often performed quite badly in
3160 this situation. This is due to a bug in the implementation of
3161 trapezoid rasterization in many X servers.
3163 Now, in cairo 1.4.6 there is a higher-level fix for this
3164 situation. Cairo now eliminates or clips trapezoids that are wholly
3165 or partially outside the clip region before handing the trapezoids
3166 to the backend. This means that the X server's performance bug is
3167 avoided in almost all cases.
3169 The net result is that doing an extreme zoom-in of vector-based
3170 objects drawn with cairo might have previously brought the X server
3171 to its knees as it allocated buffers large enough to fit all of the
3172 geometry, (whether visible or not). But now the memory usage should
3173 be bounded and performance should be dramatically better.
3177 Behdad contributed an impressively long series of changes that
3178 organizes cairo's internals in several ways that will be very
3179 beneficial to cairo developers. Thanks, Behdad!
3181 Behdad has also provided a utility for generating malloc statistics,
3182 (which was used during the great malloc purges of 1.4.2 and
3183 1.4.4). This utility isn't specific to cairo so may be of benefit to
3184 others. It is found in cairo/util/malloc-stats.c and here are Behdad's
3189 make malloc-stats.so
3191 inside util/, and to use, run:
3193 LD_PRELOAD=malloc-stats.so some-program
3195 For binaries managed by libtool, eg, cairo-perf, do:
3197 ../libtool --mode=execute /bin/true ./cairo-perf
3198 LD_PRELOAD="../util/malloc-stats.so" .libs/lt-cairo-perf
3200 Finally, the cairo-perf-diff-files utility was enhanced to allow for
3201 generating performance reports from several runs of the same backend
3202 while some system variables were changed. For example, this is now
3203 being used to allow cairo-perf to measure the performance of various
3204 different acceleration architectures and configuration options of the
3207 Release 1.4.4 (2007-04-13 Carl Worth <cworth@cworth.org>)
3208 =========================================================
3209 This is the second update release in cairo's stable 1.4 series. It
3210 comes just less than a month after 1.4.2. The changes since 1.4.2
3211 consist primarily of bug fixes, but also include at least one
3212 optimization. See below for details.
3214 Of all the work that went into the 1.4.4 release
3216 There have been lots of individuals doing lots of great work on cairo,
3217 but two efforts during the 1.4.4 series deserve particular mention:
3219 Internal cleanup of error handling, (Chris Wilson)
3220 --------------------------------------------------
3221 Chris contributed a tremendous series of patches (74 patches!) to
3222 improve cairo's handling of out-of-memory and other errors. He began
3223 by adding gcc's warn_unused_attribute to as many functions as
3224 possible, and then launched into the ambitious efforts of adding
3225 correct code to quiet the dozens of resulting warnings.
3227 Chris also wrote a custom valgrind skin to systematically inject
3228 malloc failures into cairo, and did all the work necessary to verify
3229 that cairo's performance test suite runs to completion without
3232 The end result is a much more robust implementation. Previously, many
3233 error conditions would have gone unnoticed and would have led to
3234 assertion failures, segmentation faults, or other harder-to-diagnose
3235 problems. Now, more than ever, cairo should cleanly let the user know
3236 of problems through cairo_status and other similar status
3237 functions. Well done, Chris!
3239 More malloc reduction, (Mathias Hasselmann)
3240 -------------------------------------------
3241 After 1.4.0, Behdad launched an effort to chase down excessive calls
3242 to malloc within the implementation of cairo. He fixed a lot of
3243 malloc-happy objects for 1.4.2, but one of the worst offenders,
3244 (pixman regions), was left around. Mathias contributed an excellent
3245 series of 15 patches to finish off this effort.
3247 The end result is a cairo that calls malloc much less often than it
3248 did before. Compared to 1.4.2, 55% of the calls to malloc have been
3249 eliminate, (and 60% have been eliminated compared to 1.4.0). Well
3252 Other improvements since 1.4.2
3253 ------------------------------
3254 • Centralize mutex declarations (will reduce future build breaks),
3255 (Mathias Hasselmann)
3257 • Reduce malloc by caching recently freed pattern objects (Chris
3260 • Fix some broken composite operations (David Reveman)
3261 https://bugs.freedesktop.org/show_bug.cgi?id=5777
3263 Backend-specific fixes
3264 ----------------------
3266 • Use TJ operator for more compact representation of glyphs (Adrian
3269 • Fix glyph positioning bug when glyphs are not horizontal
3270 http://lists.freedesktop.org/archives/cairo/2007-April/010337.html
3273 • Fix crash when rendering with bitmap fonts (Carl Worth)
3274 https://bugzilla.mozilla.org/show_bug.cgi?id=376498
3277 • Turn metrics-hinting on by default (Behdad Esfahbod)
3279 • Fix edge-effect problem with transformed images drawn to xlib
3281 https://bugs.freedesktop.org/show_bug.cgi?id=10508
3283 • Avoid dereferencing a NULL screen. (Chris Wilson)
3284 https://bugs.freedesktop.org/show_bug.cgi?id=10517
3287 • Fix scaling of glyph surfaces
3289 https://bugs.freedesktop.org/show_bug.cgi?id=9568
3291 • Fix compilation failure when both xlib and quartz enabled
3294 • Fix rounding bug leading to incorrectly positioned glyphs
3296 https://bugs.freedesktop.org/show_bug.cgi?id=10531
3298 Release 1.4.2 (2007-03-19 Carl Worth <cworth@cworth.org>)
3299 =========================================================
3300 This is the first update release in cairo's stable 1.4 series. It
3301 comes just less than 2 weeks after 1.4.0. We hadn't anticipated an
3302 update this early, but we've managed to collect some important fixes
3303 that we wanted to get out to cairo users as soon as possible, (6 fixes
3304 for crashes, 1 case where graphical elements would not be drawn at
3305 all, a handful of backend-specific bugs, and several important build
3308 There's almost nothing but bug fixes in this release, (see below one
3309 optimization that Behdad did sneak in), so we recommend that everyone
3310 upgrade to this release when possible.
3312 Thanks to the many people that worked to fix these bugs, and those
3313 that did the work to report them and to test the fixes, (wherever
3314 possible both names are credited below).
3318 • Fix a crash due to a LOCK vs. UNLOCK typo (M. Drochner fixing Carl
3319 Worth's embarrassing typo).
3321 http://bugs.freedesktop.org/show_bug.cgi?id=10235
3323 • Fix potential buffer overflow, which on some systems with a checking
3324 variant of snprintf would lead to a crash (Adrian Johnson, Stanislav
3327 https://bugs.freedesktop.org/show_bug.cgi?id=10267
3328 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=232576
3330 • Fix a crash in cairo_stroke_extents or cairo_in_stroke when line
3331 width is 0.0. (Carl Worth and Sebastien Bacher)
3333 https://bugs.freedesktop.org/show_bug.cgi?id=10231
3335 • Fix a crash on certain combinations of X server/video drivers (Carl
3336 Worth and Tomas Carnecky).
3338 https://bugs.freedesktop.org/show_bug.cgi?id=10250
3340 • Fix a crash due to mishandling of invalid user input (Carl Worth and
3341 Alexander Darovsky).
3343 https://bugs.freedesktop.org/show_bug.cgi?id=9844
3345 • xlib: Cleanup server-side glyph caches on XCloseDisplay. This
3346 eliminated a crash detected by the perf suite, (and that
3347 applications could have run into as well). (Chris Wilson)
3351 • Fix for some geometry which simply disappeared under some
3352 transformations---a stroked line with an extreme skew in X, for
3353 example (Carl Worth and Jonathan Watt).
3355 https://bugzilla.mozilla.org/show_bug.cgi?id=373632
3357 • SVG: Fix radial gradients for CAIRO_EXTEND_REFLECT and when r0 > r1
3360 • PDF: Set page group color space to DeviceRGB.
3362 This fixes incorrect (muddy) transparent colors when rendering cairo
3363 PDF output in some viewers. (Adrian Johnson, Adam Goode, and
3366 http://lists.freedesktop.org/archives/cairo/2006-November/008551.html
3368 • win32: Return correct metrics when hinting is off, and fix font
3369 descent computation (Behdad Esfahbod).
3371 • quartz: Fix glyph interfaces to correctly return user-space rather
3372 than device-space coordinates (Brian Ewins).
3374 https://bugs.freedesktop.org/show_bug.cgi?id=9568
3376 • xcb: Fix parameter-order confusion with xcb_create_pixmap, which now
3377 makes all tests that pass with xlib now pass with xcb (Carl Worth,
3380 • Fix some memory leaks in the perf suite (Chris Wilson).
3382 • Fix perf suite to consider changes in pixman/src (Mathias
3387 • Don't include pre-generated cairo-features.h file. This was causing
3388 build failures when building with the directfb backend enabled
3391 https://bugs.freedesktop.org/show_bug.cgi?id=10189
3393 • Eliminate use of maintainer mode from cairo's automake/configure
3394 script. This means that updates to files such as Makefile.am will
3395 take effect, (by rerunning automake and friends as necessary) when
3396 invoking make rather than being silently ignored. (Behdad Esfahbod)
3398 • Don't compile cairo-deflate-stream.c, which depends on zlib, unless
3399 building the pdf backend which requires it. (Carl Worth, Tor
3402 https://bugs.freedesktop.org/show_bug.cgi?id=10202
3404 • Don't make the ps backend link against zlib anymore, since it
3405 doesn't require it (Carl Worth).
3407 • Use "find !" rather than "find -not" for better portability (Thomas
3410 https://bugs.freedesktop.org/show_bug.cgi?id=10226
3412 • Don't use unsupported visibility attribute "hidden" on Solaris
3413 (Gilles Dauphin, Thomas Klausner).
3415 https://bugs.freedesktop.org/show_bug.cgi?id=10227
3419 • It was Behdad that suggested we focus strictly on bug fixes now that
3420 we shipped so many performance improvements in 1.4.0, but it was
3421 also Behdad that got distracted by the chance to remove a lot of
3422 mallocs from cairo. Paths, gstates, trapezoids, splines, polygons,
3423 and gradient color stops will now use small, stack-allocated buffers
3424 in the most common cases rather than calling malloc as
3425 often. (Behdad Esfahbod). And look for more from Mathias Hasselmann
3428 Release 1.4.0 (2007-03-06 Carl Worth <cworth@cworth.org>)
3429 =========================================================
3430 The many people[*] who have been working hard on cairo are very
3431 pleased to announce the long-awaited release of cairo 1.4. This
3432 release comes 4 months after the last stable update release (1.2.6)
3433 and 9 months since the initial release of 1.2.0.
3435 The release notes below are intended to capture the highlights of the
3436 changes that have occurred from the 1.2 series to the new 1.4.0
3439 Performance improvements
3440 ------------------------
3441 Within the cairo project, the last 6 months or so has seen an intense
3442 effort focusing on the performance of cairo itself. That effort has
3443 paid off considerably, as can be seen in the following highlights of
3444 some of the performance differences from cairo 1.2.6 to cairo 1.4.0.
3446 (Note: The performance results reported here were measured on an x86
3447 laptop. Many of the improvements in 1.4---particular those involving
3448 text rendering---are even more dramatic on embedded platforms without
3449 hardware floating-point units. Such devices played an important part
3450 of many of the optimizations that found their way into cairo over the
3453 • Dramatic improvement when drawing objects that are mostly off-screen
3454 with the image backend (with the xlib backend this case is still
3455 slow due to an X server bug):
3457 image-rgba long-lines-uncropped-100 479.64 -> 4.98: 96.24x speedup
3458 ███████████████████████████████████████████████▋
3460 • Dramatic improvement when copying a small fraction of an image
3461 surface to an xlib surface:
3463 xlib-rgba subimage_copy-512 3.93 -> 0.07: 54.52x speedup
3464 ██████████████████████████▊
3466 • Dramatic improvement to tessellation speed for complex objects:
3468 image-rgb tessellate-256-100 874.16 -> 34.79: 25.13x speedup
3470 xlib-rgba zrusin_another_fill-415 148.40 -> 13.85: 10.72x speedup
3472 xlib-rgb world_map-800 680.20 -> 345.54: 1.97x speedup
3475 • Dramatic improvement to the speed of stroking rectilinear shapes,
3476 (such as the outline of a rectangle or "box"):
3478 image-rgb box-outline-stroke-100 0.18 -> 0.01: 24.22x speedup
3480 xlib-rgb box-outline-stroke-100 0.46 -> 0.06: 8.05x speedup
3484 • Dramatic improvements to text rendering speeds:
3486 xlib-rgba text_image_rgba_over-256 63.12 -> 9.61: 6.57x speedup
3489 • 3x improvements to floating-point to fixed-point conversion speeds:
3491 image-rgba pattern_create_radial-16 9.29 -> 3.44: 2.70x speedup
3494 • 2x improvements to linear gradient computation:
3496 image-rgb paint_linear_rgb_source-512 26.22 -> 11.61: 2.26x speedup
3499 • 2x improvement to a case common in PDF rendering:
3501 image-rgb unaligned_clip-100 0.10 -> 0.06: 1.81x speedup
3504 • 1.3x improvement to rectangle filling speed (note: this improvement
3505 is new since 1.3.16---previously this test case was a 1.3x slowdown
3508 image-rgba rectangles-512 6.19 -> 4.37: 1.42x speedup
3510 xlib-rgba rectangles-512 7.48 -> 5.58: 1.34x speedup
3513 NOTE: In spite of our best efforts, there are some measurable
3514 performance regressions in 1.4 compared to 1.2. It appears that the
3515 primary problem is the increased overhead of the new tessellator when
3516 drawing many, very simple shapes. The following test cases capture
3517 some of that slowdown:
3519 image-rgba mosaic_tessellate_lines-800 11.03 -> 14.29: 1.30x slowdown
3521 image-rgba box-outline-fill-100 0.01 -> 0.01: 1.26x slowdown
3523 image-rgba fill_solid_rgb_over-64 0.20 -> 0.22: 1.12x slowdown
3525 image-rgba fill_image_rgba_over-64 0.23 -> 0.25: 1.10x slowdown
3527 xlib-rgb paint_image_rgba_source-256 3.24 -> 3.47: 1.07x slowdown
3529 We did put some special effort into eliminating this slowdown for the
3530 very common case of drawing axis-aligned rectangles with an identity
3531 matrix (see the box-outline-stroke and rectangles speedup numbers
3532 above). Eliminating the rest of this slowdown will be a worthwhile
3533 project going forward.
3535 Also note that the "box-outline-fill" case is a slowdown while
3536 "box-outline-stroke" is a (huge) speedup. These two test cases
3537 resulted from the fact that some GTK+ theme authors were filling
3538 between two rectangles to avoid slow performance from the more natural
3539 means of achieving the same shape by stroking a single rectangle. With
3540 1.4 that workaround should definitely be eliminated as it will now
3541 cause things to perform more slowly.
3543 Greatly improved PDF output
3544 ---------------------------
3545 We are very happy to be able to announce that cairo-generated PDF
3546 output will now have text that can be selected, cut-and-pasted, and
3547 searched with most capable PDF viewer applications. This is something
3548 that was not ever possible with cairo 1.2.
3550 Also, the PDF output now has much more compact encoding of text than
3551 before. Cairo is now much more careful to not embed multiple copies of
3552 the same font at different sizes. It also compresses text and font
3553 streams within the PDF output.
3557 There are several new functions available in 1.4 that were not
3558 available in 1.2. Curiously, almost all of the new functions simply
3559 allow the user to query state that has been set in cairo (many new
3560 "get" functions) rather than providing any fundamentally new
3561 operations. The new functionality is:
3563 • Getting information about the current clip region
3566 cairo_copy_clip_rectangle_list
3567 cairo_rectangle_list_destroy
3569 • Getting information about the current dash setting
3571 cairo_get_dash_count
3574 • Getting information from a pattern
3576 cairo_pattern_get_rgba
3577 cairo_pattern_get_surface
3578 cairo_pattern_get_color_stop_rgba
3579 cairo_pattern_get_color_stop_count
3580 cairo_pattern_get_linear_points
3581 cairo_pattern_get_radial_circles
3583 • Getting the current scaled font
3585 cairo_get_scaled_font
3587 • Getting reference counts
3589 cairo_get_reference_count
3590 cairo_surface_get_reference_count
3591 cairo_pattern_get_reference_count
3592 cairo_font_face_get_reference_count
3593 cairo_scaled_font_get_reference_count
3595 • Setting/getting user data on objects
3599 cairo_pattern_set_user_data
3600 cairo_pattern_get_user_data
3601 cairo_scaled_font_set_user_data
3602 cairo_scaled_font_get_user_data
3604 • New cairo-win32 functions:
3606 cairo_win32_surface_create_with_ddb
3607 cairo_win32_surface_get_image
3608 cairo_win32_scaled_font_get_logical_to_device
3609 cairo_win32_scaled_font_get_device_to_logical
3613 The CAIRO_FORMAT_RGB16_565 enum value has been deprecated. It never
3614 worked as a format value for cairo_image_surface_create, and it wasn't
3615 necessary for supporting 16-bit 565 X server visuals.
3617 A sampling of bug fixes in cairo 1.4
3618 ------------------------------------
3619 • Fixed radial gradients
3620 • Fixed dashing (degenerate and "leaky" cases)
3621 • Fixed transformed images in PDF/PS output (eliminate bogus repeating)
3622 • Eliminate errors from CAIRO_EXTEND_REFLECT and CAIRO_EXTEND_PAD
3623 • cairo_show_page no longer needed for single-page output
3624 • SVG: Fix bug preventing text from appearing in many viewers
3625 • cairo-ft: Return correct metrics when hinting is off
3626 • Eliminate crash in cairo_create_similar if nil surface is returned
3627 • Eliminate crash after INVALID_RESTORE error
3628 • Fix many bugs related to multi-threaded use and locking
3629 • Fix for glyph spacing 32 times larger than desired (cairo-win32)
3630 • Fixed several problems in cairo-atsui (assertion failures)
3631 • Fix PDF output to avoid problems when printing from Acrobat Reader
3632 • Fix segfault on Mac OS X (measuring a zero-length string)
3633 • Fix text extents to not include the size of non-inked characters
3634 • Fix for glyph cache race condition in glitz backend (Jinghua Luo)
3635 • Fix make check to work on OPD platforms (IA64 or PPC64)
3636 • Fix compilation problems of cairo "wideint" code on some platforms
3637 • Many, many others...
3639 Experimental backends (quartz, XCB, OS/2, BeOS, directfb)
3640 ---------------------------------------------------------
3641 None of cairo's experimental backends are graduating to "supported"
3642 status with 1.4.0, but two of them in particular (quartz and xcb), are
3645 The quartz baceknd has been entirely rewritten and is now much more
3646 efficient. The XCB backend has been updated to track the latest XCB
3647 API (which recently had a 1.0 release).
3649 We hope to see these backends become supported in a future release,
3650 (once they are passing all the tests in cairo's test suite).
3652 The experimental OS/2 backend is new in cairo 1.4 compared to cairo
3655 Documentation improvements
3656 --------------------------
3657 We have added documentation for several functions and types that
3658 were previously undocumented, and improved documentation on other
3659 ones. As of this release, there remain only two undocumented
3660 symbols: cairo_filter_t and cairo_operator_t.
3662 [*]Thanks to everyone
3663 ---------------------
3664 I've accounted for 41 distinct people with attributed code added to
3665 cairo between 1.2.6 and 1.4.0, (their names are below). That's an
3666 impressive number, but there are certainly dozens more that
3667 contributed with testing, suggestions, clarifying questions, and
3668 encouragement. I'm grateful for the friendships that have developed as
3669 we have worked on cairo together. Thanks to everyone for making this
3672 Adrian Johnson, Alfred Peng, Alp Toker, Behdad Esfahbod,
3673 Benjamin Otte, Brian Ewins, Carl Worth, Christian Biesinger,
3674 Christopher (Monty) Montgomery, Daniel Amelang, Dan Williams,
3675 Dave Yeo, David Turner, Emmanuel Pacaud, Eugeniy Meshcheryakov,
3676 Frederic Crozat, Hans Breuer, Ian Osgood, Jamey Sharp, Jeff Muizelaar,
3677 Jeff Smith, Jinghua Luo, Jonathan Watt, Joonas Pihlaja, Jorn Baayen,
3678 Kalle Vahlman, Kjartan Maraas, Kristian Høgsberg, M Joonas Pihlaja,
3679 Mathias Hasselmann, Mathieu Lacage, Michael Emmel, Nicholas Miell,
3680 Pavel Roskin, Peter Weilbacher, Robert O'Callahan,
3681 Soren Sandmann Pedersen, Stuart Parmenter, T Rowley,
3684 Snapshot 1.3.16 (2007-03-02 Carl Worth <cworth@cworth.org>)
3685 ===========================================================
3688 A few new public functions have been added to the cairo API since the
3689 1.3.14 snapshot. These include a function to query the current scaled
3692 cairo_get_scaled_font
3694 New functions to query the reference count of all cairo objects:
3696 cairo_get_reference_count
3698 cairo_surface_get_reference_count
3699 cairo_pattern_get_reference_count
3701 cairo_font_face_get_reference_count
3702 cairo_scaled_font_get_reference_count
3704 And new functions to allow the use of user_data with any cairo object,
3705 (previously these were only available on cairo_surface_t and
3706 cairo_font_face_t objects):
3711 cairo_pattern_set_user_data
3712 cairo_pattern_get_user_data
3714 cairo_scaled_font_set_user_data
3715 cairo_scaled_font_get_user_data
3717 Usability improvement for PDF/PS/SVG generation
3718 -----------------------------------------------
3719 In previous versions of cairo, generating single-page output with the
3720 cairo-pdf, cairo-ps, or cairo-svg backends required a final call to
3721 cairo_show_page. This was often quite confusing as people would port
3722 functional code from a non-paginated backend and be totally mystified
3723 as to why the output was blank until they learned to add this call.
3725 Now that call to cairo_show_page is optional, (it will be generated
3726 implicitly if the user does not call it). So cairo_show_page is only
3727 needed to explicitly separate multiple pages.
3729 Greatly improved PDF output
3730 ---------------------------
3731 We are very happy to be able to announce that cairo-generated PDF
3732 output will now have text that can be selected, cut-and-paste, and
3733 searched with most capable PDF viewer applications. This is something
3734 that was not ever possible with cairo 1.2.
3736 Also, the PDF output now has much more compact encoding of text than
3737 before. Cairo is now much more careful to not embed multiple copies of
3738 the same font at different sizes. It also compresses text and font
3739 streams within the PDF output.
3743 • Fixed radial gradients
3745 The rendering of radial gradients has been greatly improved. In
3746 the cairo 1.2 series, there was a serious regression affecting
3747 radial gradients---results would be very incorrect unless one of
3748 the gradient circles had a radius of 0.0 and a center point within
3749 the other circle. These bugs have now been fixed.
3753 Several fixes have been made to the implementation of dashed
3754 stroking. Previously, some dashed, stroked rectangles would
3755 mis-render and fill half of the rectangle with a large triangular
3756 shape. This bug has now been fixed.
3758 • Fixed transformed images in PDF/PS output
3760 In previous versions of cairo, painting with an image-based source
3761 surface pattern to the PDF or PS backends would cause many kinds
3762 of incorrect results. One of the most common problems was that an
3763 image would be repeated many times even when the user had
3764 explicitly requested no repetition with CAIRO_EXTEND_NONE. These
3765 bugs have now been fixed.
3767 • Eliminate errors from CAIRO_EXTEND_REFLECT and CAIRO_EXTEND_PAD
3769 In the 1.2 version of cairo any use of CAIRO_EXTEND_REFLECT or
3770 CAIRO_EXTEND_PAD with a surface-based pattern resulted in an
3771 error, (cairo would stop rendering). This bug has now been
3774 Now, CAIRO_EXTEND_REFLECT should work properly with surface
3777 CAIRO_EXTEND_PAD is still not working correctly, but it will now
3778 simply behave as CAIRO_EXTEND_NONE rather than triggering the
3781 New rewrite of quartz backend (still experimental)
3782 --------------------------------------------------
3783 Cairo's quartz backend has been entirely rewritten and is now much
3784 more efficient. This backend is still marked as experimental, not
3785 supported, but it is now much closer to becoming an officially
3786 supported backend. (For people that used the experimental nquartz
3787 backend in previous snapshots, that implementation has now been
3788 renamed from "nquartz" to "quartz" and has replaced the old quartz
3791 Documentation improvements
3792 --------------------------
3793 We have added documentation for several functions and types that
3794 were previously undocumented, and improved documentation on other
3795 ones. As of this release, there remain only two undocumented
3796 symbols: cairo_filter_t and cairo_operator_t.
3800 • cairo-svg: Fix bug that was preventing text from appearing in many
3803 • cairo-ft: Return correct metrics when hinting is off
3805 • Cairo 1.3.14 deadlocks in cairo_scaled_font_glyph_extents or
3806 _cairo_ft_unscaled_font_lock_face
3808 https://bugs.freedesktop.org/show_bug.cgi?id=10035
3810 • cairo crashes in cairo_create_similar if nil surface returned by
3811 other->backend->create_similar
3813 https://bugs.freedesktop.org/show_bug.cgi?id=9844
3815 • evolution crash in _cairo_gstate_backend_to_user()
3816 https://bugs.freedesktop.org/show_bug.cgi?id=9906
3818 • Fix memory leak in rectilinear stroking code
3820 Things not in this release
3821 --------------------------
3822 • Solid-surface-pattern cache: This patch had been applied during
3823 the 1.3.x series, but it was reverted due to some inter-thread
3824 problems it caused. The patch is interesting since it made a big
3825 benefit for text rendering performance---so we'll work to bring a
3826 corrected version of this patch back as soon as possible.
3828 Snapshot 1.3.14 (2006-02-13 Carl Worth <cworth@cworth.org>)
3829 ===========================================================
3830 This is the seventh development snapshot in the 1.3 series, (and there
3831 likely won't be many more before the 1.4.0 release). It comes just
3832 over 3 weeks after the 1.3.12 snapshot.
3834 Since we're so close to the 1.4.0 release, there are not a lot of new
3835 features nor even a lot of new performance improvements in this
3836 snapshot. Instead, there are a great number of bug fixes. Some are
3837 long-standing bugs that we're glad to say goodbye to, and several are
3838 fixes for regressions that were introduced as part of the optimization
3839 efforts during the 1.3.x series.
3841 PDF text selection fixed
3842 ------------------------
3843 The inability to correctly select text in cairo-generated PDF has been
3844 a defect ever since the initial support for the PDF backend in the
3845 cairo 1.2.0 release. With the 1.3.14 snapshot, in most situations, and
3846 with most PDF viewer applications, the PDF generated by cairo will
3847 allow text to be correctly selected for copy-and-paste, (as well as
3850 We're very excited about this new functionality, (and very grateful to
3851 Adrian Johnson, Behdad Esfahbod, and others that have put a lot of
3852 work into this lately). Please test this new ability and give feedback
3853 on the cairo@cairographics.org list.
3855 Many thread-safety issues fixed
3856 -------------------------------
3857 We've discovered that no release of cairo has ever provided safe text
3858 rendering from a multi-threaded application. With the 1.3.14 snapshot
3859 a huge number of the bugs in this area have been fixed, and multiple
3860 application dvelopers have now reported success at writing
3861 multi-threaded applications with cairo.
3865 Fixed a bug that was causing glyph spacing to be 32 times larger than
3866 desired when using cairo-win32.
3868 Fixed a regression in the rendering of linear gradients that had been
3869 present since the 1.3.8 snapshot.
3871 Fixed several problems in cairo-atsui that were leading to assertion
3872 failures when rendering text.
3874 Fix corrupted results when rendering a transformed source image
3875 surface to an xlib surface. This was a regression that had been
3876 present since the 1.3.2 snapshot.
3878 Fixed PDF output to prevent problems printing from some versions of
3879 Acrobat Reader, (a single glyph was being substituted for every
3882 And many other fixes as well, (see the logs for details).
3884 Snapshot 1.3.12 (2007-01-20 Carl Worth <cworth@cworth.org>)
3885 ===========================================================
3886 The relentless march toward the cairo 1.4 release continues, (even if
3887 slightly late out of the starting blocks in 2007). This is the sixth
3888 development snapshot in the 1.3 series. It comes 4 weeks after the
3893 As usual, this snapshot has some fun performance improvements to show
3896 image-rgba long-lines-uncropped-100 470.08 -> 4.95: 94.91x speedup
3897 ███████████████████████████████████████████████
3898 image-rgb long-lines-uncropped-100 461.60 -> 4.96: 93.02x speedup
3899 ██████████████████████████████████████████████
3901 This 100x improvement, (and yes, that's 100x, not 100%), in the image
3902 backend occurs when drawing large shapes where only a fraction of the
3903 shape actually appears in the final result, (the rest being outside
3904 the bounds of the destination surface). Many applications should see
3905 speedups here, and the actual amount of speedup depends on the ratio
3906 of non-visible to visible portions of geometry.
3908 [Note: There remains a similar performance bug when drawing mostly
3909 non-visible objects with the xlib backend. This is due to a similar
3910 bug in the X server itself, but we hope a future cairo snapshot will
3911 workaround that bug to get a similar speedup with the xlib backend.]
3913 image-rgba unaligned_clip-100 0.09 -> 0.06: 1.67x speedup
3915 image-rgb unaligned_clip-100 0.09 -> 0.06: 1.66x speedup
3918 This speedup is due to further MMX optimization by Soeren Sandmann for
3919 a case commonly hit when rendering PDF files, (and thanks to Jeff
3920 Muizelaar for writing code to extract the test case for us).
3922 There's another MMX optimization in this snapshot (without a fancy
3923 speedup chart) by Dan Williams which improves compositing performance
3924 specifically for the OLPC machine.
3926 Thanks to Adrian Johnson, cairo's PDF output is now much more
3927 efficient in the way it encodes text output. By reducing redundant
3928 information and adding compression to text output streams, Adrian
3929 achieved a ~25x improvement in the efficiency of encoding text in PDF
3930 files, (was ~45 bytes per glyph and is now ~1.6 bytes per glyph).
3934 In addition to those performance improvements, this snapshot includes
3937 * A huge number of bug fixes for cairo-atsui text rendering, (for mac
3938 OS X). These bugs affect font selection, glyph positioning, glyph
3939 rendering, etc. One noteworthy bug fixes is that
3940 cairo_select_font_face will no longer arbitrarily select bold nor
3941 italic when not requested, (at least not when using a standard CSS2
3942 font family name such as "serif", "sans-serif", "monospace", etc.).
3943 All these fixes are thanks to Brian Ewins who continues to do a
3944 great job as the new cairo-atsui maintainer.
3946 * Fix PDF output so that images that are scaled down no longer
3947 mysteriously repeat (Carl Worth).
3949 * Fix segfault on Mac OS X dues to attempt to measure extents of a
3950 zero-length string (Behdad Esfahbod).
3952 * Fix text extents to not include the size of initial/trailing
3953 non-inked characters (Behdad Esfahbod).
3957 Three functions have had API changes to improve consistency. Note that
3958 the API functions being changed here are all functions that were
3959 introduced as new functions during these 1.3.x snapshots. As always,
3960 there will not be any API changes to functions included in a major
3961 release (1.2.x, 1.4.x, etc.) of cairo.
3963 The changes are as follows:
3965 * Rename of cairo_copy_clip_rectangles to cairo_copy_clip_rectangle_list.
3967 * Change cairo_get_dash_count to return an int rather than accepting a
3968 pointer to an int for the return value.
3970 * Change cairo_get_dash to have a void return type rather than
3971 returning cairo_status_t.
3973 It's possible there will be one more round of changes to these
3974 functions, (and perhaps cairo_get_color_stop as well), as we seek to
3975 establish a unifying convention for returning lists of values.
3977 Snapshot 1.3.10 (2006-12-23 Carl Worth <cworth@cworth.org>)
3978 ===========================================================
3979 Santa Claus is coming just a little bit early this year, and he's
3980 bringing a shiny new cairo snapshot for all the good little boys and
3983 This is the fifth development snapshot in the 1.3 series. It comes 9
3984 days after the 1.3.8 snapshot, and still well within our goal of
3985 having a new snapshot every week, (though don't expect one next
3986 week---we'll all be too stuffed with sugar plums).
3988 Speaking of sugar plums, there's a sweet treat waiting in this cairo
3989 snapshot---greatly improved performance for stroking rectilinear
3990 shapes, like the ever common rectangle:
3992 image-rgb box-outline-stroke-100 0.18 -> 0.01: 25.58x speedup
3993 ████████████████████████▋
3994 image-rgba box-outline-stroke-100 0.18 -> 0.01: 25.57x speedup
3995 ████████████████████████▋
3996 xlib-rgb box-outline-stroke-100 0.49 -> 0.06: 8.67x speedup
3998 xlib-rgba box-outline-stroke-100 0.22 -> 0.04: 5.39x speedup
4001 In past releases of cairo, some people had noticed that using
4002 cairo_stroke to draw rectilinear shapes could be awfully slow. Many
4003 people had worked around this by using cairo_fill with a more complex
4004 path and gotten a 5-15x performance benefit from that.
4006 If you're one of those people, please rip that workaround out, as now
4007 the more natural use of cairo_stroke should be 1.2-2x faster than the
4008 unnatural use of cairo_fill.
4010 And if you hadn't ever implemented that workaround, then you just
4011 might get to see your stroked rectangles now get drawn 5-25x faster.
4013 Beyond that performance fix, there are a handful of bug fixes in this
4016 * Fix for glyph cache race condition in glitz backend (Jinghua Luo)
4018 * Many fixes for ATSUI text rendering (Brian Ewins)
4020 * Un-break recent optimization-triggered regression in rendering text
4021 with a translation in the font matrix (Behdad Esfahbod)
4023 * Fix make check to work on OPD platforms (IA64 or PPC64)
4026 * Fix a couple of character spacing issues on Windows
4029 Have fun with that, everybody, and we'll be back for more in the new
4030 year, (with a plan to add the last of our performance improvements in
4031 this round, fix a few bad, lingering bugs, and then finish off a nice,
4032 stable 1.4 release before the end of January).
4036 Snapshot 1.3.8 (2006-12-14 Carl Worth <cworth@cworth.org>)
4037 ==========================================================
4038 This is the fourth development snapshot in the 1.3 series. It comes
4039 just slightly more than one week after the 1.3.6 snapshot.
4041 After the bug fixes in 1.3.6, we're back to our original program of
4042 weekly snapshots, each one faster than the one from the week
4043 before. Cairo 1.3.8 brings a 2x improvement in the speed of rendering
4044 linear gradients (thanks to David Turner), and a significant reduction
4045 in X traffic when rendering text (thanks to Xan Lopez and Behdad
4046 Esfahbod), making cairo behave very much like Xft does.
4048 A few other things in the 1.3.8 snapshot worth noting include a more
4049 forgiving image comparator in the test suite, (using the "perceptual
4050 diff" metric and GPL implementation by Hector Yee[*]), a bug fix for
4051 broken linking on x86_64 (thanks to M Joonas Pihlaja) and an even
4052 better implementation of _cairo_lround, (not faster, but supporting a
4053 more complete input range), from Daniel Amelang.
4055 [*] http://pdiff.sourceforge.net/
4057 Snapshot 1.3.6 (2006-12-06 Carl Worth <cworth@cworth.org>)
4058 ==========================================================
4059 This is the third development snapshot in the 1.3 series. It comes two
4060 weeks after the 1.3.4 snapshot.
4062 We don't have fancy performance charts this week as the primary
4063 changes in this snapshot are bug fixes. The performance work continues
4064 and the next snapshot (planned for one week from today) should include
4065 several improvements. The bug fixes in this snapshot include:
4067 * Fix undesirable rounding in glyph positioning (Dan Amelang)
4069 This bug was noticed by several users, most commonly by seeing
4070 improper text spacing or scrambled glyphs as drawn by nautilus. For
4073 Update to cairo-1.3.4 worsen font rendering
4074 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=217819
4076 * Fix reduced range of valid input coordinates to tessellator
4079 This bug was causing lots of assertion failures in mozilla as
4082 CAIRO_BO_GUARD_BITS and coordinate space?
4083 http://lists.freedesktop.org/archives/cairo/2006-December/008743.html
4085 * Fix several regressions in new tessellator (M Joonas Pihlaja)
4087 Joonas just had a good eye for detail here. I don't think any
4088 external cairo users had noticed any of these bugs yet.
4090 * Fix compilation problems of cairo "wideint" code on some platforms
4093 * Fix failed configure due to broken grep (Dan Amelang)
4095 This bug was reported here:
4097 AX_C_FLOAT_WORDS_BIGENDIAN doesn't work because grep doesn't
4098 work with binary file
4099 https://bugs.freedesktop.org/show_bug.cgi?id=9124
4101 * Remove the pkg-config minimum version requirement (Behdad Esfahbod)
4103 Some systems ship with pkg-config 0.15 and there was really no good
4104 reason for cairo to insist on having version 0.19 before it would
4107 There is also one new (but inert) feature in this snapshot. There's a
4108 new option that can be passed to cairo's configure script:
4110 --disable-some-floating-point
4112 Disable certain code paths that rely heavily on double precision
4113 floating-point calculation. This option can improve
4114 performance on systems without a double precision floating-point
4115 unit, but might degrade performance on those that do.
4117 As of this snapshot, this option does not make any change to cairo,
4118 but it is possible that future versions of cairo will respect this
4119 option and change the implementation of various functions as
4122 Snapshot 1.3.4 (2006-11-22 Carl Worth <cworth@cworth.org>)
4123 ==========================================================
4124 This is the second development snapshot in the 1.3 series. It comes
4125 one week after the 1.3.2 snapshot.
4127 This snapshot has a couple of significant performance improvements,
4128 and also adds new support for producing multi-page SVG output, (when
4129 targeting SVG 1.2)---thanks to Emmanuel Pacaud. The details of the
4130 performance improvements are as follows:
4132 1. The long-awaited "new tessellator".
4134 The credit for this being an improvement goes to Joonas Pihlaja. He
4135 took my really slow code and really put it through its paces to get
4136 the dramatic performance improvement seen below (up to 38x faster
4137 on realistic cases, and more than 10x faster for the zrusin_another
4140 His own writeup of the work he did is quite thorough, but more than
4141 can be quoted here. Please see his post for the interesting details:
4143 http://lists.freedesktop.org/archives/cairo/2006-November/008483.html
4145 (Though note that this snapshot also includes some additional,
4146 significant improvements that were only sketched out in that
4147 email---see "Generating fewer trapezoids").
4149 2. More floating-point improvements
4151 Daniel Amelang continues to work the magic he began in the 1.3.2
4152 snapshot. This time he short-circuits floating-point
4153 transformations by identity matrices and applies the earlier
4154 floating-to-fixed-point technique to the problem of rounding.
4156 The improvements here will primarily benefit text performance, and
4157 will benefit platforms without hardware floating-point more than
4158 those that have it, (some text tests show 20% improvement on an x86
4159 machine and closer to 80% improvement on arm).
4161 The performance chart comparing 1.3.2 to 1.3.4 really speaks for
4162 itself, (this is on an x86 laptop). This is quite a lot of progress
4165 xlib-rgb stroke_similar_rgba_over-256 74.99 1.45% -> 2.03 68.38%: 36.86x speedup
4166 ███████████████████████████████████▉
4167 xlib-rgb stroke_similar_rgba_source-256 78.23 1.43% -> 3.30 67.05%: 23.71x speedup
4168 ██████████████████████▊
4169 xlib-rgba tessellate-256-100 820.42 0.15% -> 35.06 2.84%: 23.40x speedup
4170 ██████████████████████▍
4171 image-rgba tessellate-256-100 819.55 0.32% -> 35.04 3.56%: 23.39x speedup
4172 ██████████████████████▍
4173 xlib-rgb stroke_image_rgba_over-256 78.10 1.43% -> 4.33 65.56%: 18.04x speedup
4175 xlib-rgb stroke_image_rgba_source-256 80.11 1.63% -> 5.75 63.99%: 13.94x speedup
4177 xlib-rgba zrusin_another_tessellate-415 89.22 0.35% -> 8.38 5.23%: 10.65x speedup
4179 image-rgba zrusin_another_tessellate-415 87.38 0.89% -> 8.37 5.22%: 10.44x speedup
4181 image-rgba zrusin_another_fill-415 117.67 1.34% -> 12.88 2.77%: 9.14x speedup
4183 xlib-rgba zrusin_another_fill-415 140.52 1.57% -> 15.79 2.88%: 8.90x speedup
4185 image-rgba tessellate-64-100 9.68 3.42% -> 1.42 0.60%: 6.82x speedup
4187 xlib-rgba tessellate-64-100 9.78 4.35% -> 1.45 0.83%: 6.72x speedup
4189 xlib-rgb stroke_linear_rgba_over-256 46.01 2.44% -> 7.74 54.51%: 5.94x speedup
4191 xlib-rgb stroke_linear_rgba_source-256 48.09 2.15% -> 9.14 53.00%: 5.26x speedup
4193 xlib-rgb stroke_radial_rgba_over-256 50.96 2.34% -> 12.46 47.99%: 4.09x speedup
4195 xlib-rgb stroke_radial_rgba_source-256 53.06 1.57% -> 13.96 46.57%: 3.80x speedup
4197 image-rgba paint_similar_rgba_source-256 0.12 1.57% -> 0.08 9.92%: 1.42x speedup
4199 image-rgba paint_image_rgba_source-256 0.12 2.49% -> 0.08 10.70%: 1.41x speedup
4201 image-rgba world_map-800 356.28 0.46% -> 275.72 1.15%: 1.29x speedup
4203 xlib-rgba world_map-800 456.81 0.39% -> 357.95 1.39%: 1.28x speedup
4205 image-rgb tessellate-16-100 0.09 0.57% -> 0.07 3.43%: 1.23x speedup
4207 image-rgba tessellate-16-100 0.09 0.06% -> 0.07 2.46%: 1.23x speedup
4209 image-rgba text_solid_rgb_over-256 5.39 4.01% -> 4.47 0.70%: 1.21x speedup
4211 image-rgba text_solid_rgba_over-256 5.37 0.82% -> 4.45 0.75%: 1.21x speedup
4213 image-rgba text_image_rgb_over-64 0.78 0.10% -> 0.65 0.74%: 1.20x speedup
4215 image-rgba text_image_rgba_over-64 0.78 0.29% -> 0.65 0.68%: 1.19x speedup
4217 image-rgb text_solid_rgb_over-64 0.76 2.45% -> 0.63 0.81%: 1.19x speedup
4219 image-rgba text_solid_rgba_over-64 0.76 0.33% -> 0.64 0.66%: 1.19x speedup
4221 image-rgba text_similar_rgba_over-256 5.99 4.72% -> 5.04 1.09%: 1.19x speedup
4224 We should point out that there is some potential for slowdown in this
4225 snapshot. The following are the worst slowdowns reported by the cairo
4226 performance suite when comparing 1.3.2 to 1.3.4:
4228 image-rgba subimage_copy-256 0.01 0.87% -> 0.01 3.61%: 1.45x slowdown
4230 xlib-rgb paint_solid_rgb_over-256 0.31 10.23% -> 0.38 0.33%: 1.26x slowdown
4232 image-rgba box-outline-fill-100 0.01 0.30% -> 0.01 2.52%: 1.21x slowdown
4234 image-rgba fill_solid_rgb_over-64 0.20 1.22% -> 0.22 1.59%: 1.12x slowdown
4236 image-rgb fill_similar_rgb_over-64 0.21 1.04% -> 0.24 1.06%: 1.11x slowdown
4238 image-rgba fill_image_rgb_over-64 0.21 1.19% -> 0.24 0.72%: 1.11x slowdown
4240 image-rgba fill_similar_rgb_over-64 0.21 0.18% -> 0.24 0.30%: 1.11x slowdown
4242 image-rgb fill_solid_rgba_over-64 0.22 1.66% -> 0.24 1.15%: 1.11x slowdown
4244 image-rgb fill_image_rgb_over-64 0.21 0.14% -> 0.24 0.80%: 1.11x slowdown
4246 image-rgba fill_image_rgba_over-64 0.22 1.34% -> 0.25 0.20%: 1.11x slowdown
4248 image-rgba fill_solid_rgba_over-64 0.22 1.48% -> 0.24 0.95%: 1.11x slowdown
4250 image-rgb fill_similar_rgba_over-64 0.22 1.13% -> 0.25 1.25%: 1.10x slowdown
4253 The 45% slowdown for subimage_copy is an extreme case. It's unlikely
4254 to hit many applications unless they often use cairo_rectangle;
4255 cairo_fill to copy a single pixel at a time. In any case, it shows a
4256 worst-case impact of the overhead of the new tessellator. The other
4257 slowdowns (~ 10%) are probably more realistic, and still very
4260 We will work to ensure that performance regressions like these are not
4261 present from one major release of cairo to the next, (for example,
4264 But we're putting this 1.3.4 snapshot out there now, even with this
4265 potential slowdown so that people can experiment with it. If you've
4266 got complex geometry, we hope you will see some benefit from the new
4267 tessellator. If you've got primarily simple geometry, we hope things
4268 won't slowdown too much, but please let us know what slowdown you see,
4269 if any, so we can calibrate our performance suite against real-world
4272 Thanks, and have fun with cairo!
4274 Snapshot 1.3.2 (2006-11-14 Carl Worth <cworth@cworth.org>)
4275 ==========================================================
4276 This is the first development snapshot since the 1.2 stable series
4277 branched off shortly after the 1.2.4 release in August 2006.
4279 This snapshot includes all the bug fixes from the 1.2.6 release,
4280 (since they originated here on the 1.3 branch first and were
4281 cherry-picked over to 1.2). But more importantly, it contains some new
4282 API in preparation for a future 1.4 release, and most importantly, it
4283 contains several performance improvements.
4285 The bug fixes will not be reviewed here, as most of them are already
4286 described in the 1.2.6 release notes. But details for the new API and
4287 some performance improvements are included here.
4289 As with all snapshots, this is experimental code, and the new API
4290 added here is still experimental and is not guaranteed to appear
4291 unchanged in any future release of cairo.
4295 Several new API additions are available in this release. There is a
4296 common theme among all the additions in that they allow cairo to
4297 advertise information about its state that it was refusing to
4298 volunteer earlier. So this isn't groundbreaking new functionality, but
4299 it is essential for easily achieving several tasks.
4301 The new functions can be divided into three categories:
4303 Getting information about the current clip region
4304 -------------------------------------------------
4306 cairo_copy_clip_rectangles
4307 cairo_rectangle_list_destroy
4309 Getting information about the current dash setting
4310 --------------------------------------------------
4311 cairo_get_dash_count
4314 Getting information from a pattern
4315 ----------------------------------
4316 cairo_pattern_get_rgba
4317 cairo_pattern_get_surface
4318 cairo_pattern_get_color_stop_rgba
4319 cairo_pattern_get_color_stop_count
4320 cairo_pattern_get_linear_points
4321 cairo_pattern_get_radial_circles
4323 In each of these areas, we have new API for providing a list of
4324 uniform values from cairo. The closest thing we had to this before was
4325 cairo_copy_path, (which is rather unique in providing a list of
4328 The copy_clip_rectangles/rectangle_list_destroy functions follow a
4329 style similar to that of cairo_copy_path. Meanwhile, the dash and
4330 pattern color stop functions introduce a new style in which there is a
4331 single call to return the number of elements available (get_dash_count
4332 and get_color_stop_count) and then a function to be called once to get
4333 each element (get_dash and get_color_stop_rgba).
4335 I'm interested in hearing feedback from users of these new API
4336 functions, particularly from people writing language bindings. One
4337 open question is whether the clip "getter" functionality should adopt
4338 a style similar to that of the new dash and color_stop interfaces.
4342 The CAIRO_FORMAT_RGB16_565 enum value has been deprecated. It never
4343 worked as a format value for cairo_image_surface_create, and it wasn't
4344 necessary for supporting 16-bit 565 X server visuals.
4348 The XCB backend has been updated to track the latest XCB API (which
4349 recently had a 1.0 release).
4353 Vladimir Vukicevic has written a new "native quartz" backend which
4354 will eventually replace the current "image-surface wrapping" quartz
4355 backend. For now, both backends are available, (the old one is
4356 "quartz" and the new one is "nquartz"). But it is anticipated that the
4357 new backend will replace the old one and take on the "quartz" name
4358 before this backend is marked as supported in a release of cairo.
4362 Doodle and Peter Weilbacher have contributed a new, experimental
4363 backend for using cairo on OS/2 systems.
4365 Performance improvements
4366 ------------------------
4367 Here are some highlights from cairo's performance suite showing
4368 improvements from cairo 1.2.6 to cairo 1.3.2. The command used to
4369 generate this data is:
4371 ./cairo-perf-diff 1.2.6 HEAD
4373 available in the perf/ directory of a recent checkout of cairo's
4374 source, (the cairo-perf-diff script does require a git checkout and
4375 will not work from a tar file---though ./cairo-perf can still be used
4376 to generate a single report there and ./cairo-perf-diff-files can be
4377 used to compare two reports).
4379 Results are described below both for an x86 laptop (with an old Radeon
4380 video card, recent X.org build, XAA, free software drivers), as well
4381 as for a Nokia 770. First the x86 results with comments on each, (all
4382 times are reported in milliseconds).
4384 Copying subsets of an image surface to an xlib surface (much faster)
4385 --------------------------------------------------------------------
4386 xlib-rgba subimage_copy-512 10.50 -> : 53.97x speedup
4387 █████████████████████████████████████████████████████
4389 Thanks to Christopher (Monty) Montgomery for this big performance
4390 improvement. Any application which has a large image surface and is
4391 copying small pieces of it at a time to an xlib surface, (imagine an
4392 application that loads a single image containing all the "sprites" for
4393 that application), will benefit from this fix. The larger the ratio of
4394 the image surface to the portion being copied, the larger the benefit.
4396 Floating-point conversion (3x faster)
4397 -------------------------------------
4398 xlib-rgba pattern_create_radial-16 27.75 -> 3.93 : 2.94x speedup
4400 image-rgb pattern_create_radial-16 26.06 -> 3.74 : 2.90x speedup
4403 Thanks to Daniel Amelang, (and others who had contributed the idea
4404 earlier), for this nice improvement in the speed of converting
4405 floating-point values to fixed-point.
4407 Text rendering (1.3 - 2x faster)
4408 ------------------------------
4409 xlib-rgba text_image_rgba_source-256 319.73 -> 62.40 : 2.13x speedup
4411 image-rgb text_solid_rgba_over-64 2.85 -> 0.88 : 1.35x speedup
4414 I don't think we've ever set out to improve text performance
4415 specifically, but we did it a bit anyway. I believe the extra
4416 improvement in the xlib backend is due to Monty's image copying fix
4417 above, and the rest is due to the floating-point conversion speedup.
4419 Thin stroke improvements (1.5x faster)
4420 ---------------------------------------------
4421 image-rgb world_map-800 1641.09 -> 414.77 : 1.65x speedup
4423 xlib-rgba world_map-800 1939.66 -> 529.94 : 1.52x speedup
4426 The most modest stuff to announce in this release is the 50%
4427 improvement I made in the world_map case. This is in improvement that
4428 should help basically anything that is doing strokes with many
4429 straight line segments, (and the thinner the better, since that makes
4430 tessellation dominate rasterization). The fixes here are to use a
4431 custom quadrilateral tessellator rather than the generic tessellator
4432 for straight line segments and the miter joins.
4434 Performance results from the Nokia 770
4435 --------------------------------------
4436 xlib-rgba subimage_copy-512 55.88 -> 2.04 : 27.34x speedup
4437 ██████████████████████████▍
4438 xlib-rgb text_image_rgb_over-256 1487.58 -> 294.43 : 5.05x speedup
4440 image-rgb pattern_create_radial-16 187.13 -> 91.86 : 2.04x speedup
4442 xlib-rgba world_map-800 21261.41 -> 15628.02 : 1.36x speedup
4445 Here we see that the subimage_copy improvement was only about half as
4446 large as the corresponding improvement on my laptop, (27x faster
4447 compared to 54x) and the floating-point conversion fix also was quite
4448 as significant, (2x compared to 3x). Oddly the improvement to text
4449 rendering performance was more than twice as good (5x compared to
4450 2x). I don't know what the reason for that is, but I don't think it's
4451 anything anybody should complain about.
4453 Release 1.2.6 (2006-11-02 Behdad Esfahbod <behdad@behdad.org>)
4454 ==============================================================
4455 This is the third bug fix release in the 1.2 series, coming less than
4456 two months after the 1.2.4 release made on August 18.
4458 The 1.2.4 release turned out to be a pretty solid one, except for a crasher
4459 bug when forwarding an X connection where the client and the server have
4460 varying byte orders, eg. from a PPC to an i686. Other than that, various
4461 other small bugs have been fixed.
4463 Various improvements have been made in the testing infrastructure to prevent
4464 false positives, and to make sure the generated cairo shared object behaves as
4465 expected in terms of exported symbols and relocations.
4467 There were a total of 89 changes since 1.2.4. The following list the most
4472 - Avoid unsigned loop control variable to eliminate infinite,
4473 memory-scribbling loop. (#7593)
4474 - Fix cairo_image_surface_create to report INVALID_FORMAT errors.
4475 Previously the detected error was being lost and a nil surface was
4476 returned that erroneously reported CAIRO_STATUS_NO_MEMORY.
4477 - Change _cairo_color_compute_shorts to not rely on any particular
4478 floating-point epsilon value. (#7497)
4479 - Fix infinite-join test case (bug #8379)
4480 - Pass correct surface to create_similar in _cairo_clip_init_deep_copy().
4484 - Fix Type 1 embedding in PDF.
4485 - Correct the value of /LastChar in the PDF Type 1 font dictionary.
4486 - Improve error checking in TrueType subsetting.
4487 - Compute right index when looking up left side bearing. (bug #8180)
4488 - Correct an unsigned to signed conversion problem in truetype subsetting
4490 - Type1 subsetting: Don't put .notdef in Encoding when there are 256 glyphs.
4491 - Add cairo version to PS header / PDF document info dictionary.
4492 - Set CTM before path construction.
4496 - Get correct unhinted outlines on win32. (bug 7603)
4497 - Make cairo as a win32 static library possible.
4498 - Use CAIRO_FORMAT_RGB24 for BITSPIXEL==32 surfaces too.
4502 - Define WINVER if it's not defined. (bug 6456)
4503 - Fix the AMD64 final link by removing SLIM from pixman.
4504 - Misc win32 compilation fixes.
4505 - Add Sun Pro C definition of pixman_private.
4506 - Use pixman_private consistently as prefix not suffix.
4507 - Added three tests check-plt.sh, check-def.sh, and check-header.sh that check
4508 that the shared object, the .def file, and the public headers agree about
4509 the exported symbols.
4510 - Require pkg-config 0.19. (#8686)
4513 Release 1.2.4 (2006-08-18 Carl Worth <cworth@cworth.org>)
4514 =========================================================
4515 This is the second bug fix release in the 1.2 series, coming less than
4516 two weeks after the 1.2.2 release made on August 8.
4518 The big motivation for a quick release was that there were a log of
4519 build system snags that people ran into with the 1.2.2 release. But,
4520 by the time we got those all done, we found that we had a bunch of
4521 fixes for cairo's rendering as well. So there's a lot of goodness in
4522 here for such a short time period.
4526 Fix image surfaces to not be clipped when used as a source (Vladimir Vukicevic)
4527 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=72e25648c4c4bc82ddd938aa4e05887a293f0d8b
4529 Fix a couple of corner cases in dashing degenerate paths (Jeff Muizelaar)
4530 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=fbb1758ba8384650157b2bbbc93d161b0c2a05f0
4532 Fix support for type1 fonts on win32 (Adrian Johnson)
4533 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=da1019c9138695cb838a54f8b871bbfd0e8996d7
4535 Fix assertion failure when rotating bitmap fonts (Carl Worth)
4536 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=0bfa6d4f33b8ddb5dc55bbe419c15df4af856ff9
4538 Fix assertion failure when calling cairo_text_path with bitmap fonts (Carl Worth)
4539 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=9878a033531e6b96b5f27e69e10e90dee7440cd9
4541 Fix mis-handling of cairo_close_path in some situations (Tim Rowley, Carl Worth)
4542 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=53f74e59faf1af78f2f0741ccf1f23aa5dad4efc
4544 Respect font_matrix translation in _cairo_gstate_glyph_path (Behdad Esfahbod)
4545 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=f183b835b111d23e838889178aa8106ec84663b3
4547 Fix vertical metrics adjustment to work with non-identity shapes (Behdad Esfahbod)
4548 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=b7bc263842a798d657a95e539e1693372448837f
4550 [PS] Set correct ImageMatrix in _cairo_ps_surface_emit_bitmap_glyph_data (Behdad Esfahbod)
4551 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=d47388ad759b0a1a0869655a87d9b5eb6ae2445d
4555 Fix xlib detection to prefer pkg-config to avoid false libXt dependency (Behdad Esfahbod)
4556 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=0e78e7144353703cbd28aae6a67cd9ca261f1d68
4558 Fix typos causing win32 build problem with PS,PDF, and SVG backends (Behdad Esfahbod)
4559 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=aea83b908d020e26732753830bb3056e6702a774
4561 Fix configure cache to not use stale results (Behdad Esfahbod)
4562 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=6d0e3260444a2d5b6fb0cb223ac79f1c0e7b3a6e
4564 Fix to not pass unsupported warning options to the compiler (Jens Granseuer)
4565 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=97524a8fdb899de1ae4a3e920fb7bda6d76c5571
4567 Fix to allow env. variables such as png_REQUIRES to override configure detection (Jens Granseuer)
4568 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=abd16e47d6331bd3811c908e524b4dcb6bd23bf0
4570 Fix test suite to not use an old system cairo when converting svg2png (Behdad Esfahbod)
4571 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=6122cc85c8f71b1ba2df3ab86907768edebe1781
4573 Fix test suite to not require signal.h to be present (Behdad Esfahbod)
4574 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=6f8cf53b1e1ccdbe1ab6a275656b19c6e5120e40
4578 Many useful warnings cleanups from sparse, valgrind, and careful eyes
4579 (Kjartan Maraas, Pavel Roskin)
4581 Release 1.2.2 (2006-08-08 Carl Worth <cworth@cworth.org>)
4582 =========================================================
4583 This is the first bug fix release in the 1.2 series since the original
4584 1.2.0 release made six weeks ago.
4586 There were some very serious bugs in the 1.2.0 release, (see below),
4587 so everybody is encouraged to upgrade from 1.2.0 to 1.2.2. The 1.2.2
4588 release maintains source and binary compatibility with 1.2.0 and does
4589 not make any API additions.
4591 Fix crashes with BGR X servers
4592 ------------------------------
4593 With cairo 1.2.0 many people reported problems with all cairo-using
4594 programs, (including all GTK+ programs with GTK+ >= 2.8) immediately
4595 crashing with a complaint about an unsupported image format. This bug
4596 affected X servers that do not provide the Render extension and that
4597 provide a visual with BGR rather than RGB channel order.
4599 report: https://bugs.freedesktop.org/show_bug.cgi?id=7294
4600 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=9ae66174e774b57f16ad791452ed44efc2770a59
4602 Fix the "disappearing text" bug
4603 -------------------------------
4604 With cairo 1.2.0 many people reported that text would disappear from
4605 applications, sometimes reappearing with mouse motion or
4606 selection. The text would disappear after the first space in a string
4607 of text. This bug was caused by an underlying bug in (very common) X
4608 servers, and only affected text rendered without antialiasing, (either
4609 a bitmap font or a vector font with antialiasing disabled). The bug
4610 was also exacerbated by a KDE migration bug that caused antialiasing
4611 to be disabled more than desired.
4613 report: https://bugs.freedesktop.org/show_bug.cgi?id=7494
4614 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=456cdb3058f3b416109a9600167cd8842300ae14
4616 Xorg: https://bugs.freedesktop.org/show_bug.cgi?id=7681
4617 KDE: http://qa.mandriva.com/show_bug.cgi?id=23990
4619 Fix broken image fallback scaling (aka. "broken printing")
4620 ----------------------------------------------------------
4621 The various "print" backends, (pdf, ps, and svg), sometimes fallback
4622 to using image-based rendering for some operations. In cairo 1.2.0
4623 these image fallbacks were scaled improperly. Applications using cairo
4624 can influence the resolution of the image fallbacks with
4625 cairo_surface_set_fallback_resolution. With the bug, any value other
4626 than 72.0 would lead to incorrect results, (larger values would lead
4627 to increasingly shrunken output).
4629 report: https://bugs.freedesktop.org/show_bug.cgi?id=7533
4630 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=1feb4291cf7813494355459bb547eec604c54ffb
4632 Fix inadvertent semantic change of font matrix translation (Behdad Esfahbod)
4633 ----------------------------------------------------------------------------
4634 The 1.2.0 release introduced an inadvertent change to how the
4635 translation components of a font matrix are interpreted. In the 1.0
4636 series, font matrix translation could be used to offset the glyph
4637 origin, (though glyph metrics were reported incorrectly in
4638 1.0). However in 1.2.0, the translation was applied to the advance
4639 values between each glyph. The 1.2.0 behavior is fairly useless in
4640 practice, and it was not intentional to introduce a semantic
4641 change. With 1.2.2 we return to the 1.0 semantics, with a much better
4642 implementation that provides correct glyph metrics.
4644 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=84840e6bba6e72aa88fad7a0ee929e8955ba9051
4646 Fix create_similar to preserve fallback resolution and font options (Behdad Esfahbod)
4647 -------------------------------------------------------------------------------------
4648 There has been a long-standing issue with cairo_surface_create_similar
4649 such that font options and other settings from the original
4650 destination surface would not be preserved to the intermediate
4651 "similar" surface. This could result in incorrect rendering
4652 (particularly with respect to text hinting/antialiasing) with
4653 fallbacks, for example.
4655 report: https://bugs.freedesktop.org/show_bug.cgi?id=4106
4656 fixes: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=9fcb3c32c1f16fe6ab913e27eb54d18b7d9a06b0
4657 http://gitweb.freedesktop.org/?p=cairo;a=commit;h=bdb4e1edadb78a2118ff70b28163f8bd4317f1ec
4659 xlib: Fix text performance regression from 1.0 to 1.2.0 (Vladimir Vukicevic)
4660 ----------------------------------------------------------------------------
4661 Several people noticed that upgrading from cairo 1.0 to cairo 1.2.0
4662 caused a significant performance regression when using the xlib
4663 backend. This performance regression was particularly noticeable when
4664 doing lots of text rendering and when using a high-latency connection
4665 to the X server, (such as a remote X server over an ssh
4666 connection). The slowdown was identified and fixed in 1.2.2.
4668 report: https://bugs.freedesktop.org/show_bug.cgi?id=7514
4669 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=b7191885c88068dad57d68ced69a752d1162b12c
4671 PDF: Eliminate dependency on FreeType library dependency (Adrian Johnson)
4672 -------------------------------------------------------------------------
4673 The cairo 1.2 series adds a supported pdf backend to cairo. In cairo
4674 1.2.0 this backend required the freetype library, which was an
4675 undesirable dependency on systems such as win32, (cairo is designed to
4676 always prefer the "native" font system). As of cairo 1.2.2 the
4677 freetype library is not required to use the pdf backend on the win32
4680 report: https://bugs.freedesktop.org/show_bug.cgi?id=7538
4681 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=a0989f427be87c60415963dd6822b3c5c3781691
4683 PDF: Fix broken output on amd64 (Adrian Johnson)
4684 ------------------------------------------------
4685 report: http://bugzilla.gnome.org/show_bug.cgi?id=349826
4686 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=f4b12e497b7ac282b2f6831b8fb68deebc412e60
4688 PS: Fix broken output for truetype fonts > 64k (Adrian Johnson)
4689 ---------------------------------------------------------------
4690 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=067d97eb1793a6b0d0dddfbd0b54117844511a94
4692 PDF: Fix so that dashing doesn't get stuck on (Kent Worsnop)
4693 ------------------------------------------------------------
4694 Kent notices that with the PDF backend in cairo 1.2.0 as soon as a
4695 stroke was performed with dashing, all subsequent strokes would also
4696 be dashed. There was no way to turn dashing off again.
4698 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=778c4730a86296bf0a71080cf7008d7291792256
4700 Fix memory leaks in failure paths in gradient creation (Alfred Peng)
4701 --------------------------------------------------------------------
4702 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=db06681b487873788b51a6766894fc619eb8d8f2
4704 Fix memory leak in _cairo_surface_show_glyphs (Chris Wilson)
4705 ------------------------------------------------------------
4706 report: https://bugs.freedesktop.org/show_bug.cgi?id=7766
4707 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=e2fddcccb43d06486d3680a19cfdd5a54963fcbd
4709 Solaris: Add definition of cairo_private for some Sun compilers (Alfred Peng)
4710 -----------------------------------------------------------------------------
4711 report: https://bugzilla.mozilla.org/show_bug.cgi?id=341874
4712 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=04757a3aa8deeff3265719ebe01b021638990ec6
4714 Solaris: Change version number of Sun's Xorg server with buggy repeat (Brian Cameron)
4715 -------------------------------------------------------------------------------------
4716 report: https://bugs.freedesktop.org/show_bug.cgi?id=7483
4717 fix: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=e0ad1aa995bcec4246c0b8ab0d5a5a79871ce235
4719 Various memory leak fixes
4720 -------------------------
4721 Fix memory leak in _cairo_surface_show_glyphs (bug 7766)
4722 Fix file handle leak in failure path (bug 7616)
4723 Fix some memory leaks in the test cases.
4724 Fix some memory leaks in font subsetting code used in print backends.
4726 Documentation improvements (Behdad Esfahbod)
4727 --------------------------------------------
4728 Added new documentation for several functions (cairo_show_page,
4729 cairo_copy_page, cairo_in_stroke, cairo_in_fill).
4731 Fixed some syntax errors that were preventing some existing
4732 documentation from being published.
4734 Fixed several minor typographical errors.
4736 Added an index for new symbols in 1.2.
4738 Release 1.2.0 (2006-06-27 Carl Worth <cworth@cworth.org>)
4739 =========================================================
4740 This is the culmination of the work that has gone on within the 1.1
4743 There has been one API addition since the cairo 1.1.10 snapshot:
4745 cairo_xlib_surface_get_width
4746 cairo_xlib_surface_get_height
4748 There's also a new feature without any API change:
4750 Dots can now be drawn by using CAIRO_LINE_CAP_ROUND with
4751 degenerate sub-paths, (cairo_move_to() followed by either
4752 cairo_close_path() or a cairo_line_to() to the same location).
4754 And at least the following bugs have been fixed:
4756 6759 fontconfig option AntiAlias doesn't work in cairo 1.1.2
4757 6955 Some characters aren't displayed when using xlib (cache u...
4758 7268 positive device_offset values don't work as source
4759 * PDF emit_glyph function needs to support bitmapped glyphs
4760 * PS emit_glyph function needs to support bitmapped glyphs
4761 * SVG emit_glyph function needs to support bitmapped glyphs
4762 * PDF: minefield page one is falling back unnecessarily
4763 * PS/PDF: Fix broken placement for vertical glyphs
4764 * PS: Fix to not draw BUTT-capped zero-length dash segments
4765 * Do device offset before float->fixed conversion
4766 http://bugzilla.gnome.org/show_bug.cgi?id=332266
4767 * PS: Fix source surfaces with transformations
4768 * PS: Fix to not draw BUTT-capped degnerate sub-paths
4769 * PS: Don't walk off end of array when printing "~>"
4770 * Fix some memory leaks in the test suite rig
4771 * SVG: Fix memory leak when using cairo_mask
4772 * Fix EXTEND_REFLECT and EXTEND_PAD to not crash (though these are
4773 still not yet fully implemented for surface patterns).
4775 This has been a tremendous effort by everyone, and I'm proud to have
4776 been a part of it. Congratulations to all contributors to cairo!
4778 Snapshot 1.1.10 (2006-06-16 Carl Worth <cworth@cworth.org>)
4779 ===========================================================
4780 This is the fifth in a series of snapshots working toward the 1.2
4783 The primary motivation for this snapshot is to fix a long-standing bug
4784 that had long been silent, but as of the 1.1.8 snapshot started
4785 causing crashes when run against 16-bit depth X servers, (often Xvnc
4786 or Xnest). The fix for this adds a new CAIRO_FORMAT_RGB16_565 to the
4789 This snapshot also includes a rewrite of cairo's SVG backend to
4790 eliminate the dependency on libxml2. With this in place, cairo 1.2
4791 will not depend on any libraries that cairo 1.0 did not.
4793 As usual, there are also a few fixes for minor bugs.
4795 Snapshot 1.1.8 (2006-06-14 Carl Worth <cworth@cworth.org>)
4796 ==========================================================
4797 This is the fourth in a series of snapshots working toward the 1.2
4798 release of cairo. At this point, all major features of the 1.2 release
4799 are in place, leaving just a few bug fixes left.
4801 In particular, there well be no additional API changes between this
4802 1.1.8 snapshot and the 1.2 release.
4804 The announcement for 1.1.6 mentioned several API changes being
4805 considered. Only one of these changes was actually implemented
4806 (set_dpi -> fallback_resolution). This change does introduce one
4807 source-level incompatibility with respect to previous 1.1.x snapshots,
4808 so see below for details.
4810 Here is an abbreviated summary of changes since the 1.1.6 snapshot:
4814 According to the plan mentioned in the 1.1.6 notes, one source-level
4815 incompatible change has been implemented. The following three
4816 functions have been removed from cairo's API:
4818 cairo_pdf_surface_set_dpi
4819 cairo_ps_surface_set_dpi
4820 cairo_svg_surface_set_dpi
4822 and in their place the following function has been added:
4824 cairo_surface_set_fallback_resolution
4826 The signature and semantics of the function remains the same, so it is
4827 a simple matter of changing the name of the function when calling
4828 it. As a transition mechanism, this snapshot will (on many systems)
4829 build to include the old symbols so that code previously compiled will
4830 still run. However, all source code using the old names must be
4831 updated before it will compile. And the upcoming 1.2 release is not
4832 anticipated to include the old symbols.
4834 Finally, it should be pointed out that the old symbols never existed
4835 in the supported API of any stable release of cairo. (In the stable
4836 1.0 releases the PDF, PS, and SVG backends were advertised as
4837 experimental and unstable.)
4839 And, as always, cairo continues to maintain source and binary
4840 compatibility between major releases. So applications compiled against
4841 supported backends in a stable release of cairo (1.0.4 say) will
4842 continue to compile and run without modification against new major
4843 releases (1.2.0 say) without modification.
4847 The following new functions have been added to cairo's API:
4849 cairo_surface_get_content
4850 cairo_debug_reset_static_data
4851 cairo_image_surface_get_data
4852 cairo_image_surface_get_format
4853 cairo_image_surface_get_stride
4854 cairo_win32_font_face_create_for_hfont
4856 New, backend-specific pkg-config files
4857 --------------------------------------
4858 In addition to the original cairo.pc file, cairo will also now install
4859 a pkg-config files for each configured backend, (for example
4860 cairo-pdf.pc, cairo-svg.pc, cairo-xlib.pc, cairo-win32.pc, etc.) this
4861 also includes optional font backends (such as cairo-ft.pc) and the
4862 optional png functionality (cairo-png.pc).
4864 These new pkg-config files should be very convenient for allowing
4865 cairo-using code to easily check for the existing of optional
4866 functionality in cairo without having to write complex rules to grub
4867 through cairo header files or the compiled library looking for
4870 Printing backend (PS, PDF, and SVG)
4871 -----------------------------------
4872 Improving the quality of the "printing" backends has been a priority
4873 of the development between cairo 1.1.6 and cairo 1.1.8.
4875 The big improvement here is in the area of text output. Previously, at
4876 best, text was output as paths without taking advantage of any font
4877 support available in the output file format.
4879 Now, at the minimum text paths will be shared by using type3 fonts
4880 (for PS and PDF---and similarly, defs for SVG). Also, if possible,
4881 type3 and truetype fonts will be embedded in PostScript and PDF
4882 output. There are still some known bugs with this, (for example,
4883 selecting text in a cairo-generated PDF file with an embedded truetype
4884 font does not work). So there will be some more changes in this area
4885 before cairo 1.2, but do try test this feature out as it exists so
4888 Many thanks to Kristian Høgsberg for the truetype and type1 font
4893 Performance improvements by preferring GDI over pixman rendering when possible.
4894 Fixes for text rendering.
4898 Fix potentially big performance bug by making xlib's create_similar
4899 try harder to create a pixmap of a depth matching that of the screen.
4903 Among various other fixes, the following bugs listed in bugzilla have
4906 Bug 2488: Patch to fix pixman samping location bug (#2488).
4907 https://bugs.freedesktop.org/show_bug.cgi?id=2488
4909 Bug 4196: undef MIN an MAX before defining to avoid duplicate definition
4910 https://bugs.freedesktop.org/show_bug.cgi?id=4196
4912 Bug 4723: configure.in: Fix m4 quoting when examining pkg-config version
4913 https://bugs.freedesktop.org/show_bug.cgi?id=4723
4915 Bug 4882: Flag Sun's X server has having buggy_repeat.
4916 https://bugs.freedesktop.org/show_bug.cgi?id=4882
4918 Bug 5306: test/pdf2png: Add missing include of stdio.h
4919 https://bugs.freedesktop.org/show_bug.cgi?id=5306
4921 Bug 7075: Fix make clean to remove cairo.def
4922 https://bugs.freedesktop.org/show_bug.cgi?id=7075
4924 (Many thanks to Behdad Esfahbod for helping us track down and fix many
4927 Snapshot 1.1.6 (2006-05-04 Carl Worth <cworth@cworth.org>)
4928 ==========================================================
4929 This is the third in a series of snapshots working toward the imminent
4930 1.2 release of cairo. For a list of items still needing work on the
4931 cairo 1.2 roadmap, please see:
4933 http://cairographics.org/ROADMAP
4935 As can be seen in that list, there are no longer any API additions
4936 left on the roadmap. Instead, there is a feature (PDF type 3 fonts) a
4937 performance optimization (X server gradients) and a list of bug
4938 fixes. This gives us a fair amount of freedom to cut the 1.2 release
4939 at almost any point by deciding to defer remaining bug fixes to
4940 subsequent maintenance releases such as 1.2.2 and 1.2.4.
4942 Before we will do that, we must first be wiling to commit to all the
4943 new API additions. As a heads-up, there are a couple of potential API
4944 changes being considered. (Note that these are changes to new API
4945 introduced during 1.1 so these will not introduce API
4946 incompatibilities compared to the stable 1.0 series). The changes
4947 being considered are:
4949 cairo_get_group_target: may acquire x and y offset return
4950 parameters. May also be eliminated in favor of
4951 cairo_get_target assuming its role
4953 cairo_pdf_surface_set_dpi:
4954 cairo_ps_surface_set_dpi:
4955 cairo_svg_surface_set_dpi: These functions may be removed in favor
4956 of a new cairo_surface_set_fallback_resolution
4958 Additionally there is the possibility of a slight change in the
4959 semantics of cairo_set_line_width. We believe the current behavior of the sequence:
4961 cairo_set_line_width; ... change CTM ...; cairo_stroke;
4963 is buggy. It is currently behaving the same as:
4965 ... change CTM ...; cairo_set_line_width; cairo_stroke;
4967 We are considering fixing this bug before 1.2 with the hope that
4968 nobody is already relying on the buggy behavior described here. Do
4969 shout if you suspect you might be in that position.
4971 The items included in this snapshot (since the 1.1.4 snapshot) are
4976 The long-awaited group-rendering support is now available with the
4977 following function calls:
4980 cairo_push_group_with_content
4982 cairo_pop_group_to_source
4983 cairo_get_group_target
4985 This API provides a much more convenient mechanism for doing rendering
4986 to an intermediate surface without the need to manually create a
4987 temporary cairo_surface_t and a temporary cairo_t and clean them up
4990 Add the following missing get function to complement
4991 cairo_surface_set_device_offset:
4993 cairo_surface_get_device_offset
4995 PDF backend (API addition)
4996 --------------------------
4997 The PDF backend now provides for per-page size changes, (similar to
4998 what the PostScript backend got in the 1.1.4 snapshot). The new API
5001 cairo_pdf_surface_set_size
5003 Xlib backend (API additions)
5004 ----------------------------
5005 The following functions have been added to allow the extraction of
5008 cairo_xlib_surface_get_display
5009 cairo_xlib_surface_get_drawable
5010 cairo_xlib_surface_get_screen
5011 cairo_xlib_surface_get_visual
5012 cairo_xlib_surface_get_depth
5014 XCB backend (experimental)
5015 --------------------------
5016 Update backend so that it now compiles with the recent XCB 0.9 release.
5018 Bug fixes and memory leak cleanup
5019 ---------------------------------
5020 Various little things, nothing too significant though.
5022 Snapshot 1.1.4 (2006-05-03 Carl Worth <cworth@cworth.org>)
5023 ==========================================================
5024 This is the second in a series of snapshots working toward the
5025 upcoming 1.2 release of cairo. For a list of items still needing work
5026 on the cairo 1.2 roadmap, please see:
5028 http://cairographics.org/ROADMAP
5030 The items included in this snapshot (since the 1.1.2 snapshot) are
5033 PostScript backend: new printing-oriented API
5034 ---------------------------------------------
5035 We anticipate that with cairo 1.2, toolkits will begin to use cairo
5036 for printing on systems that use PostScript as the spool format. To
5037 support this use case, we have added 4 new function calls that are
5038 specific to the PostScript backend:
5040 cairo_ps_surface_set_size
5041 cairo_ps_surface_dsc_comment
5042 cairo_ps_surface_dsc_begin_setup
5043 cairo_ps_surface_dsc_begin_page_setup
5045 These functions allow variation of the page size/orientation from one
5046 page to the next in the PostScript output. They also allow the toolkit
5047 to provide per-document and per-page printer control options in a
5048 device-independent way, (for example, by using PPD options and
5049 emitting them as DSC comments into the PostScript output). This should
5050 allow toolkits to provide very fine-grained control of many options
5051 available in printers, (media size, media type, tray selection, etc.).
5053 SVG backend: builds by default, version control
5054 -----------------------------------------------
5055 The SVG backend continues to see major improvements. It is expected
5056 that the SVG backend will be a supported backend in the 1.2
5057 release. This backend will now be built by default if its dependencies
5058 (freetype and libxml2) are met.
5060 Additionally, the SVG backend now has flexibility with regard to what
5061 version of SVG it targets. It will target SVG 1.1 by default, which
5062 will require image fallbacks for some of the "fancier" cairo
5063 compositing operators. Or with the following new function calls:
5065 cairo_svg_surface_restrict_to_version
5066 cairo_svg_get_versions
5067 cairo_svg_version_to_string
5069 it can be made to target SVG 1.2 in which there is native support for
5070 these compositing operators.
5074 At least the following bugs have been fixed since the 1.1.2 snapshot:
5076 crash at XRenderAddGlyphs
5077 https://bugs.freedesktop.org/show_bug.cgi?id=4705
5079 Can't build cairo-1.1.2 on opensolaris due to " void function cannot return value"
5080 https://bugs.freedesktop.org/show_bug.cgi?id=6792
5082 Missing out-of-memory check at gfx/cairo/cairo/src/cairo-atsui-font.c:185
5083 https://bugzilla.mozilla.org/show_bug.cgi?id=336129
5085 A couple of memory leaks.
5087 Snapshot 1.1.2 (2006-04-25 Carl Worth <cworth@cworth.org>)
5088 ==========================================================
5089 This is the first in a series of snapshots working toward the upcoming
5090 1.2 release of cairo. (Subsequent snapshot will use successive even
5091 numbers for the third digit, 1.1.4, 1.1.6, etc.) This snapshot is
5092 backwards-compatible with the 1.0 series---it makes a few API
5093 additions but does not remove any API.
5095 PostScript and PDF backends are no longer "experimental"
5096 --------------------------------------------------------
5097 The major theme of the 1.2 release is improved PostScript and PDF
5098 backends for cairo. Unlike the 1.0 series, in the 1.2 series these
5099 backends will not be marked as experimental and will be enabled by
5100 default. We encourage people to test this snapshot and the PS/PDF
5101 backends in particular as much as possible.
5103 The PostScript and PDF output is not yet ideal.
5105 * One major problem with the PostScript output is that image
5106 fallbacks are used more often than strictly necessary, and the
5107 image fallbacks are at a lower resolution than desired, (the
5108 cairo_ps_surface_set_dpi call is ignored).
5110 * The major drawback of the current PDF backend implementation is
5111 its text support. Every glyph is represented by a filled path in
5112 the PDF file. The causes file sizes to be much larger and
5113 rendering to be much slower than desired.
5115 It is anticipated that both of these shortcomings will see some
5116 improvements before the final 1.2 release.
5118 In spite of those shortcomings, we hope that the PS and PDF backends
5119 will yield faithful results for pretty much any cairo operations you
5120 can throw at them. Please let us know if you are getting obviously
5121 "different" results from the PS/PDF backends than from the image or
5124 Other new experimental backends
5125 -------------------------------
5126 This snapshot includes three new backends that did not exist in the
5135 These are all currently marked "experimental" and are disabled by
5136 default. But the SVG backend in particular has seen a lot of recent
5137 development and is very close to passing the entire cairo test
5138 suite. It is possible that this backend will become a fully supported
5139 backend by the time of the cairo 1.2 release.
5141 Public API additions
5142 --------------------
5143 There have been a few new API functions added to cairo, including:
5145 New get_type functions for querying sub-types of object:
5147 cairo_surface_get_type
5148 cairo_pattern_get_type
5149 cairo_font_face_get_type
5150 cairo_scaled_font_get_type
5152 More convenience in working with cairo_scaled_font_t with new getter
5155 cairo_scaled_font_get_font_face
5156 cairo_scaled_font_get_font_matrix
5157 cairo_scaled_font_get_ctm
5158 cairo_scaled_font_get_font_options
5160 As well as a convenience function for setting a scaled font into a
5163 cairo_set_scaled_font
5165 and a function to allow text extents to be queried directly from a
5166 scaled font, (without requiring a cairo_surface_t or a cairo_t):
5168 cairo_scaled_font_text_extents
5170 These new scaled font functions were motivated by the needs of the
5173 Finally, a new path-construction function was added which clears the
5174 current point in preparation for a new sub path. This makes cairo_arc
5175 easier to use in some situations:
5179 Before the 1.2 release is final we do still plan a few more API
5180 additions specifically motivated by the needs of Mozilla/Firefox.
5182 Optimizations and bug fixes
5183 ---------------------------
5184 Shortly after the 1.0 maintenance series branched off the mainline
5185 there was a major rework of the cairo font internals. This should
5186 provide some good performance benefits, but it's also another area
5187 people should look at closely for potential regressions.
5189 There has not yet been any widespread, systematic optimization of
5190 cairo, but various performance improvements have been made, (and some
5191 of them are fairly significant). So if some things seem faster than
5192 1.0 then things are good. If there are any performance regressions
5193 compared to 1.0 then there is a real problem and we would like to hear
5196 There has been a huge number of bug fixes---too many to mention in
5197 detail. Again, things should be better, and never worse compared to
5198 1.0. Please let us know if your testing shows otherwise.
5200 Release 1.0.2 (2005-10-03 Carl Worth <cworth@cworth.org>)
5201 =========================================================
5202 For each bug number XXXX below, see:
5204 https://bugs.freedesktop.org/show_bug.cgi?id=XXXX
5210 * 4408 - Add support for dashing of stroked curves
5213 * 4409 - Fix dashing so that each dash is capped on both ends
5216 * 4414 - Prevent SIGILL failures (proper use of -mmmx and -msse flags)
5217 (Sebastien Bacher, Billy Biggs)
5219 * 4299 - Fix crashes with text display in multi-threaded program
5220 (Alexey Shabalin, Carl Worth)
5222 * 4401 - Do not use sincos function since it is buggy on some platforms)
5223 (Tim Mooney, Carl Worth)
5225 * 4245 - Fix several bugs in the test suite exposed by amd64 systems
5226 (Seemant Kulleen, Carl Worth)
5228 * 4321 - Add missing byteswapping on GetImage/PutImage
5229 (Sjoerd Simons, Owen Taylor)
5231 * 4220 - Make the check for rectangular trapezoids simpler and more accurate
5232 (Richard Stellingwerff, Owen Taylor)
5234 * 4260 - Add missing channel-order swapping for antialised fonts
5235 (Barbie LeVile, Owen Taylor)
5237 * 4283 - Fix compilation failure with aggressive inlining (gcc -O3)
5238 (Marco Manfredini, Owen Taylor)
5240 * 4208 - Fix some warnings from sparse
5241 (Kjartan Maraas, Billy Biggs)
5243 * 4269 - Fix to not crash when compiled with -fomit-frame-pointer
5244 (Ronald Wahl, Owen Taylor)
5246 * 4263 - Improve performance for vertical gradients
5247 (Richard Stellingwerff, Owen Taylor)
5250 * 4298 - Accomodate gentoo and Mandriva versions in X server vendor string check
5251 (Billy Biggs, Frederic Crozat, Owen Taylor)
5253 win32-specific fixes
5254 --------------------
5255 * 4599 - Fix "missing wedges" on some stroked paths (win32)
5256 (Tim Rowley, Jonathan Watt, Bertram Felgenhauer, Carl Worth, Keith Packard)
5258 * 4612 - Fix disappearing text if first character out of surface (win32)
5261 * 4602 - Fix shutdown of cairo from failing intermediate, size-0 bitmaps (win32)
5262 Aka. the "white rectangles" bug from mozilla-svg testing
5265 * Various portability improvements for win32
5266 (Hans Breuer, Owen Taylor, Carl Worth)
5268 * 4593 - Fix font sizes to match user expectations (win32)
5269 (Tor Lillqvist, Owen Taylor)
5271 * 3927 - Fix to report metrics of size 0 for glyph-not-available (win32)
5272 (Hans Breuer, Owen Taylor, Tor Lillqvist)
5274 * Add locking primitives for win32
5279 * Fix crash from size-0 pixmap due to empty clip region (xlib)
5280 (Radek Doulík, Carl Worth)
5282 Release 1.0.0 (2005-08-24 Carl Worth <cworth@cworth.org>)
5283 =========================================================
5284 Experimental backends
5285 ---------------------
5286 * The Glitz, PS, PDF, Quartz, and XCB backends have been declared
5287 experimental, and are not part of the API guarantees that accompany
5288 this release. They are not built by default, even when the required
5289 libraries are available, and must be enabled explicitly with
5290 --enable-ps, --enable-pdf, --enable-quartz or --enable-xcb.
5292 It is very painful for us to be pushing out a major release without
5293 these backends enabled. There has been a tremendous amount of work
5294 put into each one and all are quite functional to some
5295 extent. However, each also has some limitations. And none of these
5296 backends have been tested to the level of completeness and
5297 correctness that we expect from cairo backends.
5299 We do encourage people to experiment with these backends and report
5300 success, failure, or means of improving them.
5304 * Prior to 0.9.0 the SOURCE, CLEAR and a number of other operators
5305 behaved in an inconsistent and buggy fashion and could affect areas
5306 outside the clip mask. In 0.9.0, these six "unbounded" operators
5307 were fixed to consistently clear areas outside the shape but within
5308 the clip mask. This is useful behavior for an operator such as IN,
5309 but not what was expected for SOURCE and CLEAR. So, in this release
5310 the behavior of SOURCE and CLEAR has been changed again. They now
5311 affect areas only within both the source and shape. We can write
5312 the new operators as:
5314 SOURCE: dest' = (mask IN clip) ? source : dest
5315 CLEAR: dest' = (mask IN clip) ? 0 : dest
5317 Behavior and API changes
5318 ------------------------
5319 * Setting the filter on a gradient pattern would change the
5320 interpolation between color stops away from the normal linear
5321 interpolation. This dubious behavior has been removed.
5323 * The CAIRO_CONTENT_VALID() and CAIRO_FORMAT_VALID() macros --
5324 implementation details that leaked into cairo.h -- have been moved
5325 into an internal header.
5327 * The cairo_show_text function now advances the current point
5328 according to the total advance values of the string.
5332 * cairo_set_dash can now detect error and can set
5333 CAIRO_STATUS_INVALID_DASH.
5337 * When compiled against recent versions of fontconfig and FreeType,
5338 artificial bold fonts can now be turned on from fonts.conf using
5339 the FC_EMBOLDEN fontconfig key.
5343 * The compositing code from the 'xserver' code tree has now been
5344 completely merged into libpixman. This includes MMX optimization of
5347 * The image transformation code in libpixman has been improved and
5348 now performs significantly faster.
5352 * Several crashes related to corruption in the font caches have been
5355 * All test cases now match pixel-for-pixel on x86 and PPC; this
5356 required fixing bugs in the compositing, stroking, and pattern
5359 * Negative dash offsets have been fixed to work correctly.
5361 * The stroking of paths with mutiple subpaths has now been fixed to
5362 apply caps to all subpaths rather than just the last one.
5364 * Many build fixes for better portability on various systems.
5366 * Lots of other bug fixes, but we're too tired to describe them in
5369 Release 0.9.2 (2005-08-13 Carl Worth <cworth@cworth.org>)
5370 =========================================================
5373 * You will notice that this release jumped from 0.9.0 to 0.9.2. We've
5374 decided to use an odd micro version number (eg. 0.9.1) to indicate
5375 in-progress development between releases. As soon as 0.9.2 is
5376 tagged, the version will be incremented in CVS to 0.9.3 where it
5377 will stay until just before 0.9.4 is built, uploaded, and tagged.
5379 So, even-micro == a released version, odd-micro == something in-between.
5381 Libpixman dependency dropped
5382 ----------------------------
5383 * As of this release, the dependency on an external libpixman has
5384 been dropped. Instead, the code from libpixman needed for cairo has
5385 been incorporated into the cairo source tree. The motivation for
5386 this change is that while cairo's API is stable and ready to be
5387 maintained after the 1.0 release, libpixman's API is not, so we do
5388 not want to expose it at this time.
5390 Also, the incorporation of libpixman into cairo also renames all
5391 previously-public libpixman symbols in order to avoid any conflict
5392 with a future release of libpixman
5396 * Macros and functions have been added so that the version of cairo
5397 can be queried at either compile-time or at run-time. The version
5398 is made available as both a human-readable string and as a single
5401 CAIRO_VERSION_STRING eg. "0.9.2"
5402 CAIRO_VERSION eg. 000902
5405 cairo_version_string (void); /* eg. "0.9.2" */
5408 cairo_version (void); /* eg. 000902 */
5410 A macro is provided to convert a three-part component version into
5411 the encoded single-integer form:
5413 CAIRO_VERSION_ENCODE(X,Y,Z)
5415 For example, the CAIRO_VERSION value of 000902 is obtained as
5416 CAIRO_VERSION_ENCODE(0,9,2). The intent is to make version
5417 comparisons easy, either at compile-time:
5419 #if CAIRO_VERSION >= CAIRO_VERSION_ENCODE(0,9,2)
5425 if (cairo_version() >= CAIRO_VERSION_ENCODE(0,9,2)) { /* ... */ }
5429 * This release adds pthread-based locking (when available) to make
5430 the caches used by cairo safe for threaded programs. Some may
5431 remember a failed experiment with this locking between the 0.5.1
5432 and 0.5.2 snapshots, (where even single-threaded programs that
5433 linked with -lpthread would deadlock). We believe that that problem
5434 has been fixed, so we are looking forward to testing and reports
5435 from users with threaded applications.
5439 * The XCB and Quartz backends failed to compiled in the 0.9.0 release
5440 due to minor syntax errors. These have now been fixed.
5442 * Various crashes in glitz and pixman due to size 0 glyphs have been
5445 Release 0.9.0 (2005-08-08 Carl Worth <cworth@cworth.org>)
5446 =========================================================
5449 * In all prior snapshots, the libtool library versioning was set to
5450 1:0:0. As this release is intended to mark the beginning of
5451 backwards-compatible releases, the versioning has been incremented
5452 to 2:0:0. You will notice that the numeric extension on the
5453 installed library filename will change similarly.
5455 This change will also require all cairo-using applications to be
5456 recompiled. We recognize that this may cause some frustration since
5457 this release is backwards-compatible with 0.6.0 and in that sense
5458 "shouldn't" require re-compilation. However, since all historical
5459 snapshots have used the same 1:0:0 version in spite of incompatible
5460 API changes between them, it was essential that the upcoming 1.0
5461 release series have distinct library versioning.
5463 All future releases will use the library versioning to properly
5464 indicate compatibility between releases. So, any application
5465 re-compiled now to work with the 0.9.0 will not need to be
5466 recompiled when a compatible 1.0 release of cairo is made in the
5471 * Add new function calls to set/get the current antialiasing mode in
5477 This call accepts the same modes recently added for font options
5478 (NONE or GRAY) but affects the rendering of geometry other than
5479 text. The intent of this call is to enable more precise control of
5480 which pixels are affected by each operation, for example to allow
5481 for full-scene antialiasing for seam-free rendering. It is not
5482 expected that non-antialiased rendering will perform better than
5483 anti-aliased rendering.
5485 * Three new functions were added to provide support for mixed cairo-
5486 and non-cairo drawing to the same surface:
5488 cairo_surface_mark_dirty
5489 cairo_surface_mark_dirty_rectangle
5492 * The return type of the several "reference" functions was change,
5493 (API compatibly), from void to the same type as the argument. The
5494 affected functions are:
5496 cairo_font_face_reference
5497 cairo_scaled_font_reference
5498 cairo_pattern_reference
5499 cairo_surface_reference
5502 This allows a convenient way to assign and reference in a single
5507 * The behavior of cairo_set_source with a pattern with a non-identity
5508 matrix was previously not well-defined. The new behavior is as
5511 The pattern's transformation matrix will be locked to the
5512 user space in effect at the time of cairo_set_source(). This means
5513 that further modifications of the CTM will not affect the source
5518 * Some portability improvements, (eg. workaround for missing stdint.h).
5522 * Updated to allow compilation with older versions of freetype.
5526 * Fix the unbounded operators to actually produce a correct result,
5527 (previously the results were artificially restricted to the
5528 bounding box of whatever shape was being drawn rather than
5529 extending out infinitely). The fixed operators are:
5531 CAIRO_OPERATOR_CLEAR
5532 CAIRO_OPERATOR_SOURCE
5535 CAIRO_OPERATOR_DEST_IN
5536 CAIRO_OPERATOR_DEST_ATOP
5538 * Fix cairo_mask and cairo_mask_surface to transform the mask by the
5539 current transformation matrix (CTM).
5541 * Fix cairo_set_source to lock the CTM used to transform the pattern.
5543 * Workaround for X server Render bug involving repeating patterns
5544 with a general transformation matrix.
5546 * cairo_get_font_face fixed to return a "nil" font face object rather
5549 * cairo_set_font_face fixed to not crash if given a NULL font face,
5550 (which is the documented interface for restoring the default font
5553 * Fix xlib glyphset caching to not try to free a NULL glyph.
5555 Snapshot 0.6.0 (2005-07-28 Carl Worth <cworth@cworth.org>)
5556 ==========================================================
5559 * The prototypes of the following functions have changed:
5561 cairo_xlib_surface_create_with_xrender_format
5562 cairo_xlib_surface_create_for_bitmap
5564 A Screen* parameter has been added to each. This allows the cairo
5565 xlib backend to work correctly with multi-head X servers.
5567 * The following function has been modified:
5569 cairo_scaled_font_create
5571 to accept a cairo_font_options_t*. See below fore more details.
5573 * All opaque, reference-counted cairo objects have now been moved to a
5574 standard error-handling scheme. The new objects to receive this
5575 treatment are cairo_font_face_t, cairo_scaled_font_t, and
5576 cairo_surface_t. (Previous snapshots already provided this scheme
5577 for cairo_t, cairo_path_t, and cairo_pattern_t.)
5579 This changes two functions to have a return type of void rather than
5582 cairo_scaled_font_extent
5583 cairo_surface_finish
5585 And significantly, none of the create functions for any of the
5586 objects listed above will return NULL. The pointer returned from any
5587 function will now always be a valid pointer and should always be
5588 passed to the corresponding destroy function when finished
5590 The simplest strategy for porting code is to switch from:
5592 object = cairo_<object>_create ();
5598 cairo_<object>_destroy (object);
5602 object = cairo_<object>_create ();
5603 if (cairo_<object>_status (object))
5608 cairo_<object>_destroy (object);
5610 But significantly, it is not required to check for an error status
5611 before the "act on object" portions of the code above. All
5612 operations on an object with an error status are, by definition,
5613 no-ops without side effect. So new code might be written in an
5614 easier-to-read style of:
5616 object = cairo_<object>_create ();
5620 cairo_<object>_destroy (object);
5622 with cairo_<object>_status checks placed only at strategic
5623 locations. For example, passing an error object to another object,
5624 (eg. cairo_set_source with an in-error pattern), will propagate the
5625 error to the subsequent object (eg. the cairo_t). This means that
5626 error checking can often be deferred even beyond the destruction of
5631 * New functions for checking the status of objects that have been
5632 switched to the common error-handling scheme:
5634 cairo_font_face_status
5635 cairo_scaled_font_status
5636 cairo_surface_status
5638 * The _cairo_error function which was added in 0.5.1 has now been made
5639 much more useful. In 0.5.1 only errors on cairo_t objects passed
5640 through _cairo_error. Now, an error on any object should pass
5641 through _cairo_error making it much more reliable as a debugging
5642 mechanism for finding when an error first occurs.
5644 * Added new font options support with a myriad of functions:
5646 cairo_font_options_create
5647 cairo_font_options_copy
5648 cairo_font_options_destroy
5650 cairo_font_options_status
5652 cairo_font_options_merge
5653 cairo_font_options_equal
5654 cairo_font_options_hash
5656 cairo_font_options_set_antialias
5657 cairo_font_options_get_antialias
5658 cairo_font_options_set_subpixel_order
5659 cairo_font_options_get_subpixel_order
5660 cairo_font_options_set_hint_style
5661 cairo_font_options_get_hint_style
5662 cairo_font_options_set_hint_metrics
5663 cairo_font_options_get_hint_metrics
5665 cairo_surface_get_font_options
5667 cairo_ft_font_options_substitute
5669 cairo_set_font_options
5670 cairo_get_font_options
5672 This new font options support allows the application to have much
5673 more fine-grained control over how fonts are rendered.
5674 Significantly, it also allows surface backends to have some
5675 influence over the process. For example, the xlib backend now
5676 queries existing Xft properties to set font option defaults.
5680 cairo_xlib_surface_set_drawable
5682 which allows the target drawable for an xlib cairo_surface_t to be
5683 changed to another with the same format, screen, and display. This
5684 is necessary in certain double-buffering techniques.
5688 * Sub-pixel text antialiasing is now supported.
5692 * Fixed assertion failure in cairo_surface_create_similar when
5693 application commits an error by passing a cairo_format_t rather than
5696 * Avoid division by zero in various places (cairo-ft).
5698 * Fix infinite loop when using non-default visuals (cairo-xlib).
5700 * Eliminate segfault in cairo_image_surface_create_from_png_stream.
5702 * Prevent errant sign-extension of masks on 64-bit architectures
5703 (cairo-xlib and cairo-xcb).
5705 * Other miscellaneous fixes.
5707 Snapshot 0.5.2 (2005-07-18 Carl Worth <cworth@cworth.org>)
5708 ==========================================================
5711 * New functions for creating patterns of a single color:
5713 cairo_pattern_create_rgb
5714 cairo_pattern_create_rgba
5716 * Change cairo_surface_create_similar to accept a new type of
5717 cairo_content_t rather than cairo_format_t:
5719 typedef enum _cairo_content {
5720 CAIRO_CONTENT_COLOR = 0x1000,
5721 CAIRO_CONTENT_ALPHA = 0x2000,
5722 CAIRO_CONTENT_COLOR_ALPHA = 0x3000
5725 * Add new CAIRO_FORMAT_VALID and CAIRO_CONTENT_VALID macros.
5727 * Remove unused status value:
5729 CAIRO_STATUS_NO_TARGET_SURFACE
5731 * Add new status values:
5733 CAIRO_STATUS_INVALID_STATUS
5735 * Require libpixman >= 0.1.5 (for necessary bug fixes)
5739 * Fix cairo_surface_write_to_png for RGB24 images.
5741 * Fix broken metrics and rendering for bitmap fonts. Add mostly
5742 useless bitmap glyph transformation.
5744 * Fix glyph caches to not eject entries that might be immediately
5745 needed, (fixing intermittent crashes when rendering text).
5747 * Fix all memory leaks found by running "make check-valgrind".
5749 ATSUI backend changes
5750 ---------------------
5751 * Allow building against < 10.3 SDK.
5753 * Prevent crash on empty strings.
5755 Glitz backend changes
5756 ---------------------
5757 * Require glitz >= 0.4.4.
5759 * Use frame buffer objects instead of pbuffers for accelerated
5762 * Minor improvement to gradient pattern creation.
5764 PostScript backend fixes
5765 ------------------------
5766 * Rewrite of the PS backend to generate more interesting output that
5767 the old big-image implementation.
5771 * Implement glyph path support.
5773 * Fix swap of blue and green values in the fill_rectangles path.
5777 * Add optimization to use XCopyArea rather than XRenderComposite when
5778 transforming only with an integer translation, and using SOURCE
5779 operator or OVER with a source pattern without alpha.
5781 Snapshot 0.5.1 (2005-06-20 Carl Worth <cworth@cworth.org>)
5782 ==========================================================
5785 * Removed cairo_status_string(cairo_t*) and add
5786 cairo_status_to_string(cairo_status_t) in its place. Code using
5787 cairo_status_string can be ported forward as follows:
5791 cairo_status_to_string (cairo_status (cr));
5793 * Removed the BAD_NESTING restriction which means that two different
5794 cairo_t objects can now interleave drawing to the same
5795 cairo_surface_t without causing an error.
5797 * The following functions which previously had a return type of
5798 cairo_status_t now have a return type of void:
5800 cairo_pattern_add_color_stop_rgba
5801 cairo_pattern_set_matrix
5802 cairo_pattern_get_matrix
5803 cairo_pattern_set_extend
5804 cairo_pattern_set_filter
5806 See discussion of cairo_pattern_status below for more details.
5810 * Improved error handling:
5813 cairo_pattern_status (cairo_pattern_t *pattern);
5815 This snapshot expands the status-based error handling scheme from
5816 cairo_t to cairo_path_t and cairo_pattern_t. It also expands the
5817 scheme so that object-creating functions, (cairo_create,
5818 cairo_pattern_create_*, cairo_copy_path_*), are now guaranteed to
5819 not return NULL. Instead, in the case of out-of-memory these
5820 functions will return a static object with
5821 status==CAIRO_STATUS_NO_MEMORY. The status can be checked with the
5822 functions cairo_status and cairo_pattern_status, or by direct
5823 inspection of the new status field in cairo_path_t.
5825 Please note that some objects, including cairo_surface_t and all of
5826 the font-related objects have not been converted to this
5827 error-handling scheme.
5829 * In addition to the above changes, a new private function has been added:
5833 This function can be used to set a breakpoint in a debugger to make
5834 it easier to find programming error in cairo-using code. (Currently,
5835 _cairo_error is called when any error is detected within a cairo_t
5836 context, but is not called for non-cairo_t errors such as for
5837 cairo_path_t and cairo_pattern_t).
5839 * Fixed cairo_path_data_t so that its enum is visible to C++ code, (as
5840 cairo_path_data_type_t).
5842 Performance improvements
5843 ------------------------
5844 * Made a minor performance improvement for clipping, (restrict clip
5845 surface to the new intersected bounds).
5847 * Optimize rendering of a solid source pattern with a pixel-aligned
5848 rectangular path to use backend clipping rather than rasterization
5849 and backend compositing.
5851 * Optimize cairo_paint_with_alpha to defer to cairo_paint when alpha
5856 * Fixed memory leak in cairo_copy_path.
5858 * A build fix for non-srcdir builds.
5862 * New support for path-based clipping.
5864 * Fix for text rotated to angles other than multiples of π/2.
5868 * Fix for text extents.
5872 * Implemented a complex workaround for X server bug[*] related to
5873 Render-based compositing with untransformed, repeating source
5874 pictures. The workaround uses core Xlib when possible for
5875 performance, (ie. with CAIRO_OPERATOR_SOURCE or CAIRO_OPERATOR_OVER
5876 with an opaque source surface), and falls back to the pixman
5877 image-based compositing otherwise.
5879 [*] https://bugs.freedesktop.org/show_bug.cgi?id=3566
5881 * Various bug fixes, particularly in the fallback paths.
5883 Snapshot 0.5.0 (2005-05-17 Carl Worth <cworth@cworth.org>)
5884 ==========================================================
5885 This is a pretty big, and fairly significant snapshot. It represents
5886 between 2 and 3 months of solid work from a lot of people on improving
5887 the API as much as possible. I'd like to express my appreciation and
5888 congratulations to everyone who has worked on the big API Shakeup,
5889 (whether in email battles over names, or fixing my silly bugs).
5891 This snapshot will require some effort on the part of users, since
5892 there are a _lot_ of API changes (ie. no cairo program ever written is
5893 safe --- they're all broken now in at least one way). But, in spite of
5894 that, we do encourage everyone to move their code to this snapshot as
5895 soon as possible. And we're doing everything we can think of to make
5896 the transition as smooth as possible.
5898 The idea behind 0.5 is that we've tried to make every good API change
5899 we could want now, and get them all done with. That is, between now
5900 and the 1.0 release of cairo, we expect very few new API changes,
5901 (though some will certainly sneak in). We will have some significant
5902 additions, but the pain of moving code from cairo 0.4 to cairo 0.5
5903 should be a one time experience, and things should be much smoother as
5904 we continue to move toward cairo 1.0.
5906 And with so many changes coming out for the first time in this 0.5
5907 release, we really do need a lot of people trying this out to make
5908 sure the ideas are solid before we freeze the API in preparation for
5911 OK, enough introduction. Here is a (not-quite-complete) description of
5912 the API removals, changes and additions in this snapshot, (compared to
5917 The following public functions have been removed:
5919 - cairo_set_target_*
5921 This is a big change. See the description of cairo_create in
5922 the API changes section for how to deal with this.
5926 Alpha blending hasn't gone away; there's just a much more
5927 unified rendering model now. Almost all uses of
5928 cairo_set_alpha will be trivially replaced with
5929 cairo_set_source_rgba and a few others will be replaced just
5930 as easily with cairo_paint_with_alpha.
5932 - cairo_show_surface
5934 Another useful function that we realized was muddling up the
5935 rendering model. The replacement is quite easy:
5936 cairo_set_source_surface and cairo_paint.
5938 - cairo_matrix_create
5939 - cairo_matrix_destroy
5941 - cairo_matrix_get_affine
5943 These functions supported an opaque cairo_matrix_t. We now
5944 have an exposed cairo_matrix_t structure, so these can be
5947 - cairo_surface_set_repeat
5948 - cairo_surface_set_matrix
5949 - cairo_surface_set_filter
5951 These properties don't belong on surfaces. If you were using
5952 them, you'll just want to instead use
5953 cairo_pattern_create_for_surface and then set these properties
5958 This was a confusing function and hopefully nobody will miss
5959 it. But if you really don't find cairo_save/restore adequate,
5960 let us know and we have another idea for a potential
5963 And while we're on the subject of removals, we carefully tightened up
5964 the cairo header files so they no longer gratuitously include header
5965 files that are not strictly necessary, (stdio.h, stdint.h, pixman.h,
5966 Xrender.h, etc. and their dependencies). This may lead to some
5967 surprising errors, so keep your eyes open for that.
5971 Here are some of the API changes that have occurred:
5973 ~ cairo_create(void) -> cairo_create(cairo_surface_t *)
5975 This is the big change that breaks every program. The ability
5976 to re-target a cairo_t was not particularly useful, but it did
5977 introduce a lot of muddy semantic questions. To eliminate
5978 that, cairo_create now requires its target surface to be
5979 passed in at creation time. This isn't too hard to cope with
5980 as the typical first operation after cairo_create was often
5981 cairo_set_target_foo. So the order of those two swap and the
5982 application instead has cairo_foo_surface_create, then
5985 ~ cairo_current_* -> cairo_get_*
5987 We had a strange mixture of cairo_get and cairo_current
5988 functions. They've all been standardized on cairo_get, (though
5989 note one is cairo_get_current_point).
5991 ~ CAIRO_OPERATOR_SRC -> CAIRO_OPERATOR_SOURCE
5992 ~ CAIRO_OPERATOR_OVER_REVERSE -> CAIRO_OPERATOR_DEST_OVER
5994 Many of the cairo_operator_t symbolic values were renamed to
5995 reduce the amount of abbreviation. The confusing "OP_REVERSE"
5996 naming was also changed to use "DEST_OP" instead which is
5997 easier to read and has wider acceptance in other
5998 libraries/languages.
6000 ~ cairo_set_pattern -> cairo_set_source
6001 ~ cairo_set_rgb_color -> cairo_set_source_rgb
6003 All of the various functions that changed the source
6004 color/pattern were unified to use cairo_set_source names to
6005 make the relation more clear.
6007 ~ cairo_transform_point -> cairo_user_to_device
6008 ~ cairo_transform_distance -> cairo_user_to_device_distance
6009 ~ cairo_inverse_transform_point -> cairo_device_to_user
6010 ~ cairo_inverse_transform_distance -> cairo_device_to_user_distance
6012 These names just seemed a lot more clear.
6014 ~ cairo_init_clip -> cairo_reset_clip
6015 ~ cairo_concat_matrix -> cairo_transform
6017 More abbreviation elimination
6019 ~ cairo_current_path -> cairo_copy_path
6020 ~ cairo_current_path_flat -> cairo_copy_path_flat
6022 The former mechanism for examining the current path was a
6023 function that required 3 or 4 callbacks. This was more
6024 complexity than warranted in most situations. The new
6025 cairo_copy_path function copies the current path into an
6026 exposed data structure, and the documentation provides a
6027 convenient idiom for navigating the path data.
6033 A generalized version of the painting operators cairo_stroke
6034 and cairo_fill. The cairo_paint call applies the source paint
6035 everywhere within the current clip region. Very useful for
6036 clearing a surface to a solid color, or painting an image,
6037 (see cairo_set_source_surface).
6039 + cairo_paint_with_alpha
6041 Like cairo_paint but applying some alpha to the source,
6042 (making the source paint translucent, eg. to blend an image on
6047 A more generalized version of cairo_paint_with_alpha which
6048 allows a pattern to specify the amount of translucence at each
6049 point rather than using a constant value everywhere.
6051 + cairo_mask_surface
6053 A convenience function on cairo_mask for when the mask pattern
6054 is already contained within a surface.
6056 + cairo_surface_set_user_data
6057 + cairo_surface_get_user_data
6058 + cairo_font_face_set_user_data
6059 + cairo_font_face_get_user_data
6061 Associate arbitrary data with a surface or font face for later
6062 retrieval. Get notified when a surface or font face object is
6065 + cairo_surface_finish
6067 Allows the user to instruct cairo to finish all of its
6068 operations for a given surface. This provides a safe point for
6069 doing things such as flushing and closing files that the
6070 surface may have had open for writing.
6072 + cairo_fill_preserve
6073 + cairo_stroke_preserve
6074 + cairo_clip_preserve
6076 One interesting change in cairo is that the path is no longer
6077 part of the graphics state managed by
6078 cairo_save/restore. This allows functions to construct paths
6079 without interfering with the graphics state. But it prevents
6080 the traditional idiom for fill-and-stroke:
6082 cairo_save; cairo_fill; cairo_restore; cairo_stroke
6084 Instead we know have alternate versions cairo cairo_fill,
6085 cairo_stroke, and cairo_clip that preserve the current path
6086 rather than consuming it. So the idiom now becomes simply:
6088 cairo_fill_preserve; cairo_stroke
6090 + cairo_surface_write_to_png
6091 + cairo_surface_write_to_png_stream
6093 In place of a single PNG backend, now a surface created
6094 through any backend (except PDF currently) can be written out
6097 + cairo_image_surface_create_from_png
6098 + cairo_image_surface_create_from_png_stream
6100 And its just as easy to load a PNG image into a surface as well.
6104 With the new, exposed path data structure, it's now possible
6105 to append bulk path data to the current path, (rather than
6106 issuing a long sequence of cairo_move_to/line_to/curve_to
6109 Xlib and XCB backends
6110 ---------------------
6112 Any cairo_format_t and Colormap arguments have been dropped from
6113 cairo_xlib_surface_create. There are also two new
6114 cairo_xlib|xcb_surface_create functions:
6116 cairo_xlib|xcb_surface_create_for_bitmap
6117 (Particular for creating A1 surfaces)
6118 cairo_xlib|xcb_surface_create_with_xrender_format
6119 (For any other surface types, not described by a Visual*)
6121 All of these surface create functions now accept width and height. In
6122 addition, there are new cairo_xlib|xcb_surface_set_size functions
6123 which must be called each time a window that is underlying a surface
6126 Print backends (PS and PDF)
6127 ---------------------------
6128 The old FILE* based interfaces have been eliminated. In their place we
6129 have two different functions. One accepts a simple const char
6130 *filename. The other is a more general function which accepts a
6131 callback write function and a void* closure. This should allow the
6132 flexibility needed to hook up with various stream object in many
6135 In addition, when specifying the surface size during construction, the
6136 units are now device-space units (ie. points) rather than inches. This
6137 provides consistency with all the other surface types and also makes
6138 it much easier to reason about the size of the surface when drawing to
6139 it with the default identity matrix.
6141 Finally, the DPI parameters, which are only needed to control the
6142 quality of fallbacks, have been made optional. Nothing is required
6143 during surface_create (300 DPI is assumed) and
6144 cairo_ps|pdf_surface_set_dpi can be used to set alternate values if
6149 Owen very graciously listened to feedback after the big font rework he
6150 had done for 0.4, and came up with way to improve it even more. In 0.4
6151 there was a cairo_font_t that was always pre-scaled. Now, there is an
6152 unscaled cairo_font_face_t which is easier to construct, (eg. no
6153 scaling matrix required) and work with, (it can be scaled and
6154 transformed after being set on the graphics state). And the font size
6155 manipulation functions are much easier. You can set an explicit size
6156 and read/modify/write the font matrix with:
6159 cairo_get_font_matrix
6160 cairo_set_font_matrix
6162 (Previously you could only multiply in a scale factor or a matrix.) A
6163 pleasant side effect is that we can (and do) now have a default font
6164 size that is reasonable, as opposed to the old default height of one
6165 device-space unit which was useless until scaled.
6167 Of course, the old pre-scaled font had allowed some performance
6168 benefits when getting many metrics for a font. Those benefits are
6169 still made available through the new cairo_scaled_font_t. And a
6170 cairo_font_face_t can be "promoted" to a cairo_scaled_font_t by
6171 suppling a font_matrix and the desired CTM.
6175 Tim Rowley put in the work to bring the Quartz backend back after it
6176 had been disabled in the 0.4.0 snapshot. He was not able to bring back
6177 the function that allows one to create a cairo_font_t from an ATSUI
6181 cairo_atsui_font_create (ATSUStyle style);
6183 because he didn't have a test case for it. If you care about this
6184 function, please provide a fairly minimal test and we'll try to bring
6185 it back in an upcoming snapshot.
6187 Snapshot 0.4.0 (2005-03-08 Carl Worth <cworth@cworth.org>)
6188 ==========================================================
6191 Owen Taylor has converted cairo's documentation system to gtk-doc and
6192 has begun some long-needed work on the documentation, which can now be
6195 http://cairographics.org/manual/
6199 This is the first snapshot to include a functional win32 backend,
6200 (thanks to Owen Taylor). The interface is as follows:
6202 #include <cairo-win32.h>
6205 cairo_set_target_win32 (cairo_t *cr,
6209 cairo_win32_surface_create (HDC hdc);
6212 cairo_win32_font_create_for_logfontw (LOGFONTW *logfont,
6213 cairo_matrix_t *scale);
6216 cairo_win32_font_select_font (cairo_font_t *font,
6220 cairo_win32_font_done_font (cairo_font_t *font);
6223 cairo_win32_font_get_scale_factor (cairo_font_t *font);
6225 And see also the documentation at:
6227 http://cairographics.org/manual/cairo-Microsoft-Windows-Backend.html
6229 Disabled backend: quartz
6230 ------------------------
6231 Unfortunately, the quartz backend code is currently out of date with
6232 respect to some recent backend interface changes. So, the quartz
6233 backend is disabled in this snapshot.
6235 If the quartz backend is brought up-to-date before the next snapshot,
6236 we would be glad to make a 0.4.1 snapshot that re-enables it, (we do
6237 not expect many more big backend interface changes).
6241 The font system has been revamped, (as Owen Taylor's work with
6242 integrating pango and cairo gave us the first serious usage of the
6245 One fundamental, user-visible change is that the cairo_font_t object
6246 now represents a font that is scaled to a particular device
6247 resolution. Further changes are described below.
6251 Removed cairo_font_set_transform and cairo_font_current_transform.
6253 Added cairo_font_extents and cairo_font_glyph_extents. See
6254 documentation for details:
6256 http://cairographics.org/manual/cairo-cairo-t.html#cairo-font-extents
6260 The cairo_ft_font API changed considerably. Please see the
6261 documentation for details:
6263 http://cairographics.org/manual/cairo-FreeType-Fonts.html
6267 Make the fast-path clipping (pixel-aligned rectangles) faster.
6269 Add optimization for applying a constant alpha to a pattern.
6271 Optimize gradients that are horizontal or vertical in device space.
6273 Xlib: When RENDER is not available, use image surfaces for
6274 intermediate surfaces rather than xlib surfaces.
6276 Backend-specific changes
6277 ------------------------
6280 Major update to glitz backend. The output quality should now be just
6281 as good as the image and xlib backends.
6283 Track changes to glitz 0.4.0.
6287 Various improvements to produce more conformant output.
6291 David Reveman contributed a large re-work of the cairo_pattern_t
6292 implementation, providing cleaner code and more optimization
6295 Backend interface changes
6296 -------------------------
6297 Rework backend interface to accept patterns, not surfaces for source
6300 Remove set_matrix, set_filter, and set_repeat functions.
6302 More sophisticated backend interface for image fallbacks,
6303 ({acquire,release}_{source,dest}_image() and clone_similar).
6307 Only install header files for backends that have been compiled.
6309 Fixed some rounding errors leading to incorrectly placed glyphs.
6311 Many other minor fixes.
6313 Snapshot 0.3.0 (2005-01-21 Carl Worth <cworth@cworth.org>)
6314 ==========================================================
6317 1) The public header files will no longer be directly installed into
6318 the system include directory. They will now be installed in a
6319 subdirectory named "cairo", (eg. in /usr/include/cairo rather than
6322 As always, the easiest way for applications to discover the
6323 location of the header file is to let pkg-config generate the
6324 necessary -I CFLAGS and -L/-l LDFLAGS. For example:
6326 cc `pkg-config --cflags --libs cairo` -o foo foo.c
6328 IMPORTANT: Users with old versions of cairo installed will need to
6329 manually remove cairo.h and cairo-features.h from the
6330 system include directories in order to prevent the old
6331 headers from being used in preference to the new ones.
6333 2) The backend-specific portions of the old monolithic cairo.h have
6334 been split out into individual public header files. The new files
6347 Applications will need to be modified to explicitly include the new
6348 header files where appropriate.
6350 3) There are two new graphics backends in this snapshot, a PDF
6351 backend, and a Quartz backend. There is also one new font backend,
6356 Kristian Høgsberg has contributed a new backend to allow cairo-based
6357 applications to generate PDF output. The interface for creating a PDF
6358 surface is similar to that of the PS backend, as can be seen in
6362 cairo_set_target_pdf (cairo_t *cr,
6364 double width_inches,
6365 double height_inches,
6366 double x_pixels_per_inch,
6367 double y_pixels_per_inch);
6370 cairo_pdf_surface_create (FILE *file,
6371 double width_inches,
6372 double height_inches,
6373 double x_pixels_per_inch,
6374 double y_pixels_per_inch);
6376 Once a PDF surface has been created, applications can draw to it as
6377 any other cairo surface.
6379 This code is still a bit rough around the edges, and does not yet
6380 support clipping, surface patterns, or transparent gradients. Text
6381 only works with TrueType fonts at this point and only black text is
6382 supported. Also, the size of the generated PDF files is currently
6385 Kristian is still actively developing this backend, so watch this
6386 space for future progress.
6390 Calum Robinson has contributed a new backend to allow cairo
6391 applications to target native Mac OS X windows through the Quartz
6392 API. Geoff Norton integrated this backend into the current
6393 configure-based build system, while Calum also provided Xcode build
6394 support in the separate "macosx" module available in CVS.
6396 The new interface, available in cairo-quartz.h, is as follows:
6399 cairo_set_target_quartz_context (cairo_t *cr,
6400 CGContextRef context,
6405 cairo_quartz_surface_create (CGContextRef context,
6409 There is an example program available in CVS in cairo-demo/quartz. It
6410 is a port of Keith Packard's fdclock program originally written for
6411 the xlib backend. A screenshot of this program running on Mac OS X is
6414 http://cairographics.org/~cworth/images/fdclock-quartz.png
6418 This new font backend complements the Quartz backend by allowing
6419 applications to use native font selection on Mac OS X. The interface
6420 is a single new function:
6423 cairo_atsui_font_create (ATSUStyle style);
6427 Prototype for non-existent function "cairo_ft_font_destroy" removed.
6429 Now depends on libpixman 0.1.2 or newer, (0.1.3 is being released
6430 concurrently and has some useful performance improvements).
6432 Default paint color is now opaque black, (was opaque white). Default
6433 background color is transparent (as before).
6435 Renamed "struct cairo" to "struct _cairo" to free up the word "cairo"
6436 from the C++ identifier name space.
6438 Functions returning multiple return values through provided pointers,
6439 (cairo_matrix_get_affine, cairo_current_point, and
6440 cairo_current_color_rgb), will now accept NULL for values the user
6443 CAIRO_HAS_FREETYPE_FONT has now been renamed to CAIRO_HAS_FT_FONT.
6445 Performance improvements
6446 ------------------------
6447 Alexander Larsson provided some fantastic performance improvements
6448 yielding a 10000% performance improvement in his application, (when
6449 also including his performance work in libpixman-0.1.3). These include
6451 * Fixed handling of cache misses.
6453 * Creating intermediate clip surfaces at the minimal size required.
6455 * Eliminating roundtrips when creating intermediate Xlib surfaces.
6459 Major re-work of font metrics system by Keith Packard. Font metrics
6460 should now be much more reliable.
6464 Updated for glitz-0.3.0.
6465 Bug fixes in reference counting.
6469 New tests for cache crashing, rotating text, improper filling of
6470 complex polygons, and leaky rasterization.
6474 Fixed assertion failure when selecting the same font multiple times in
6477 Fixed reference counting so cache_destroy functions work.
6479 Remove unintended copyright statement from files generated with
6482 Fixed to eliminate new warnings from gcc 3.4 and gcc 4.
6484 Snapshot 0.2.0 (2004-10-27 Carl Worth <cworth@cworth.org>)
6485 ===========================================================
6486 New license: LGPL/MPL
6487 ---------------------
6488 The most significant news with this release is that the license of
6489 cairo has changed. It is now dual-licensed under the LGPL and the
6490 MPL. For details see the COPYING file as well as COPYING-LGPL-2.1 and
6493 I express my thanks to everyone involved in the license change process
6494 for their patience and support!
6496 New font and glyph internals
6497 ----------------------------
6498 Graydon Hoare has put a tremendous amount of work into new internals
6499 for handling fonts and glyphs, including caches where appropriate.
6500 This work has no impact on the user-level API, but should result in
6501 great performance improvements for applications using text.
6505 This snapshot of cairo includes a (small) test suite in
6506 cairo/test. The tests can be run with "make check". The test suite was
6507 designed to make it very easy to add new tests, and we hope to see
6508 many contributions here. As you find bugs, please try adding a minimal
6509 test case to the suite, and submit it with the bug report to the
6510 cairo@cairographics.org mailing list. This will make it much easier
6511 for us to track progress in fixing bugs.
6513 New name for glitz backend
6514 --------------------------
6515 The gl backend has now been renamed to the glitz backend. This means
6516 that the following names have changed:
6518 CAIRO_HAS_GL_SURFACE -> CAIRO_HAS_GLITZ_SURFACE
6519 cairo_set_target_gl -> cairo_set_target_glitz
6520 cairo_gl_surface_create -> cairo_glitz_surface_create
6522 This change obviously breaks backwards compatibility for applications
6523 using the old gl backend.
6525 Up-to-date with latest glitz snapshots
6526 --------------------------------------
6527 This snapshot of cairo is now up to date with the latest glitz
6528 snapshot, (currently 0.2.3). We know that the latest cairo and glitz
6529 snapshots have been incompatible for a very long time. We've finally
6530 fixed that now and we're determined to not let that happen again.
6532 Revert some tessellation regression bugs
6533 ----------------------------------------
6534 People that have been seeing some tessellation bugs, (eg. leaked
6535 fills), in the CVS version of cairo may have better luck with this
6536 release. A change since the last snapshot was identified to trigger
6537 some of these bugs and was reverted before making the snapshot. The
6538 behavior should be the same as the previous (0.1.23) snapshot.
6540 Miscellaneous changes
6541 ---------------------
6542 Changed CAIRO_FILTER_DEFAULT to CAIRO_FILTER_BEST to make gradients
6545 Track XCB API change regarding iterators.
6549 Fix calculation of required number of vertices for pen.
6551 Fix to avoid zero-dimensioned pixmaps.
6553 Fix broken sort of pen vertices.
6555 Fix bug when cairo_show_text called with a NULL string.
6559 Fix bug in computing image length with XCB.
6561 Fix infinite loop bug in cairo_arc.
6563 Fix memory management interactions with libpixman.
6565 Snapshot 0.1.23 (2004-05-11 Carl Worth <cworth@isi.edu>)
6566 ========================================================
6569 Fix prototype mismatches so that cairo can be built by gcc 3.4.
6571 Updates to track glitz
6572 ----------------------
6573 Various fixes to support the latest glitz snapshot (0.1.2).
6577 Radial gradients now support both inner and outer circles.
6578 Transformed linear gradients are now properly handled.
6579 Fixes for extend type reflect.
6583 Converted shading routines to use fixed point values and introduced a
6584 shading operator structure for more efficient shading calculations.
6585 Support compositing with mask surface when mask is solid or
6586 multi-texturing is available.
6588 PNG backend cleanups
6589 --------------------
6590 Fix output to properly compensate for pre-multiplied alpha format in cairo.
6591 Add support for A8 and A1 image formats.
6595 Avoid crash or infinite loop on null strings and degeneratively short
6598 New? bugs in cairo_clip
6599 -----------------------
6600 There are some fairly serious bugs in cairo_clip. It is sometimes
6601 causing an incorrect result. And even when it does work, it is
6602 sometimes so slow as to be unusable. Some of these bugs may not be
6603 new, (indeed cairo_clip has only ever had a braindead-slow
6604 implementation), but I think they're worth mentioning here.
6606 Snapshot 0.1.22 (2004-04-16 Carl Worth <cworth@isi.edu>)
6607 ========================================================
6608 Cairo was updated to track the changes in libpixman, and now depends
6609 on libpixman version 0.1.1.
6611 Snapshot 0.1.21 (2004-04-09 David Reveman <c99drn@cs.umu.se>)
6612 =============================================================
6615 The OpenGL backend provides hardware accelerated output for
6616 X11 and OS X. The significant new functions are:
6619 cairo_gl_surface_create
6621 Automatic detection of available backends
6622 -----------------------------------------
6623 The configure script now automatically detect what backends are
6624 available, (use ./configure --disable-`backend' to prevent
6625 compilation of specific backends).
6627 Snapshot 0.1.20 (2004-04-06 Carl Worth <cworth@isi.edu>)
6628 ========================================================
6631 David Reveman has contributed a new pattern API which enable linear
6632 and radial gradient patterns in addition to the original surface-based
6633 patterns. The significant new top-level functions are:
6635 cairo_pattern_create_linear
6636 cairo_pattern_create_radial
6637 cairo_pattern_create_for_surface
6638 cairo_pattern_add_color_stop
6641 Any code using the old cairo_set_pattern, (which accepted a
6642 cairo_surface_t rather than a cairo_pattern_t), will need to be
6645 Update to XCB backend
6646 ---------------------
6647 The XCB backend is now enabled by default, (use ./configure
6648 --disable-xcb to turn it off).
6652 Graydon Hoare has added optimizations that make cairo_clip much faster
6653 when the path is a pixel-aligned, rectangular region.
6657 Snapshot 0.1.19 (2004-02-24 Carl Worth <cworth@isi.edu>)
6658 ========================================================
6661 Olivier Andrieu contributed a new PNG backend. It builds on the
6662 existing image backend to make it easy to render "directly" to a
6663 .png file. The user never needs to deal with the actual image
6664 buffer. The significant new functions are:
6666 cairo_set_target_png
6667 cairo_png_surface_create
6669 The PNG backend is not enabled by default so that by default there is
6670 not a new dependency on libpng. Use ./configure --enable-png to enable
6673 Snapshot 0.1.18 (2004-02-17 Carl Worth <cworth@isi.edu>)
6674 ========================================================
6675 Path query functionality
6676 ------------------------
6677 It's now possible to query the current path. The two new functions
6681 cairo_current_path_flat
6683 Each function accepts a number of callback functions that will be
6684 called for each element in the path (move_to, line_to, curve_to,
6685 close_path). The cairo_current_path_flat function does not accept a
6686 curve_to callback. Instead, all curved portions of the path will be
6687 converted to line segments, (within the current tolerance value). This
6688 can be handy for doing things like text-on-path without having to
6689 manually interpolate Bézier splines.
6693 Jamey Sharp has contributed a second X backend that uses the new, lean
6694 XCB library rather than Xlib. It cannot currently be compiled at the
6695 same time as the Xlib backend. See ./configure --enable-xcb.
6697 Build fixes for cygwin.
6701 Snapshot 0.1.17 (2003-12-16 Carl Worth <cworth@isi.edu>)
6702 ========================================================
6706 This snapshot provides much better text support by implementing the
6707 following four functions:
6714 The text/glyph_extents functions can be used to determine the bounding
6715 box (and advance) for text as if drawn by show_text/glyphs.
6717 The text/glyph_path objects functions place text shapes on the current
6718 path, where they can be subsequently manipulated. For example,
6719 following these functions with cairo_stroke allows outline text to be
6720 drawn. Calling cairo_clip allows clipping to a text-shaped region.
6722 Combined dependencies
6723 ---------------------
6724 The cairo core now depends only on the libpixman library. This single
6725 library replaces the three previous libraries libic, libpixregion, and
6726 slim. Thanks to Dave Beckett <dave.beckett@bristol.ac.uk> for all of
6727 the heavy lifting with this renaming effort.
6729 Conditional compilation of backends
6730 -----------------------------------
6731 Cairo now allows optional backends to be disabled at compile time. The
6732 following options may now be passed to the configure script:
6737 Note that the first option is a change from the old --without-x option
6738 which will no longer have any effect.
6740 OS X supported - several byte-order issues resolved
6741 ---------------------------------------------------
6742 Cairo has now been successfully compiled under OS X. Testing revealed
6743 that there were some byte-order problems in the PostScript backend and
6744 the PNG generation in the demos. These have now been resolved.
6748 Graydon Hoare <graydon@redhat.com> implemented the first real text
6749 support using Freetype/fontconfig, (previous versions of cairo used
6750 Xft and could only draw text when using an X backend).
6754 Graydon Hoare <graydon@redhat.com> added the first real support for
6755 running cairo with a non-render-aware X server.
6757 Jamey Sharp <jamey@minilop.net> virtualized the backend font and
6758 surface interfaces in September, 2003.
6762 Xr is renamed cairo to avoid confusion since it no longer had a strict
6767 A new image surface backend is added to Xr. Keith Packard
6768 <keithp@keithp.com> wrote the image compositing code in libic that is
6769 used for the image_surface backend. This code was originally written
6770 as the software fallback for the render extension within the X
6775 Carl Worth <cworth@isi.edu> wrote the first lines of Xr, after Keith
6776 Packard <keithp@keithp.com> proposed the plan for a stateful drawing
6777 library in C providing a PostScript-like rendering model.
6779 LocalWords: mutex BeOS extraordinaire distro's URL lcd bool tarball