organicmaps/harfbuzz
13
Fork 0
mirror of https://github.com/harfbuzz/harfbuzz.git synced 2025-04-10 15:23:35 +00:00
Code Issues 1 Projects Releases Wiki Activity Actions 7
17917 commits 25 branches 188 tags 127 MiB
Commit graph

17917 commits

This branch
This branch
All branches
Author SHA1 Message Date
Behdad Esfahbod
f858def14b Revert "[test] Exclude google-benchmark tests by default if meson "recent"" 
This reverts commit 3c975a857f.

See https://github.com/harfbuzz/harfbuzz/issues/4153#issuecomment-2646149763
 2025-02-09 13:07:20 +00:00
Behdad Esfahbod
f68193a00a [glyf] Use direct access to points array 
Not necessary faster. Prep'ing for next change.
 2025-02-09 12:38:38 +00:00
Behdad Esfahbod
19a152b47e [perf] Don't depend on libharfbuzz-subset for most benchmarks 2025-02-09 12:37:28 +00:00
Behdad Esfahbod
3c975a857f [test] Exclude google-benchmark tests by default if meson "recent" 
https://github.com/harfbuzz/harfbuzz/issues/4153#issuecomment-2646131694
 2025-02-09 08:58:55 +00:00
Behdad Esfahbod
cb6779f9e5 [test] Try running slow tests early 
Probably already happens, but this wouldn't hurt.

https://github.com/harfbuzz/harfbuzz/issues/4153#issuecomment-2646125367
 2025-02-09 08:39:03 +00:00
Behdad Esfahbod
134eef550d [test] Reorder test suites 
Makes it a bit faster.

https://github.com/harfbuzz/harfbuzz/issues/4153#issuecomment-2646120679
 2025-02-09 08:26:11 +00:00
Behdad Esfahbod
c55738fa77 [test] Set MALLOC_CHECK_ to 1 instead of 2; kill G_SLICE=always-malloc 
MALLOC_CHECK_=2 makes running tests twice slower on macOS. Value 1
is as fast as not checking at all. It's not like this ever caught
a bug for us or anything. We have so many sanitizers and fuzzers
funning on CI, which catch all kinds of memory problems.

G_SLICE is long gone in glib.

The slowest part of running the test suite now is the google-benchmark
tests, that get run by us... I'll see if I can disable. I had reported
it before at https://github.com/harfbuzz/harfbuzz/issues/4153
 2025-02-09 08:11:01 +00:00
Behdad Esfahbod
b13017e651 Use HB_ALWAYS_INLINE instead of bare __attribute__ 
The former works on MSVC too.
 2025-02-09 08:00:18 +00:00
Behdad Esfahbod
ee19666c39 [hb-shape/view] Allow --unicodes and --text if at least one is empty 
Useful for scripting.
 2025-02-09 05:14:36 +00:00
Behdad Esfahbod
5cab741929
Merge pull request #5058 from harfbuzz/ot-font-trak 
[ot-font/trak] Move trak application to ot-font instead of ot-shape
 2025-02-09 04:56:12 +00:00
Behdad Esfahbod
922a56f43a [trak] Fix configs build 2025-02-09 04:50:17 +00:00
Behdad Esfahbod
17c11ec523 [ft/trak] Apply trak in ft-font-funcs as well 
Part of https://github.com/harfbuzz/harfbuzz/issues/5049
 2025-02-09 04:46:29 +00:00
Behdad Esfahbod
ffae5b040d [ot-font/trak] Move trak application to ot-font instead of ot-shape 
This is what CoreText does.

Fixes https://github.com/harfbuzz/harfbuzz/issues/5049
 2025-02-09 04:39:32 +00:00
Behdad Esfahbod
f840480fde [trak] Remove leftover 2025-02-09 04:27:20 +00:00
Behdad Esfahbod
be287eab02 [trak] Don't shift glyphs 
That's what CoreText does and fonts are designed for. Who are we to improve on...
 2025-02-09 04:19:02 +00:00
Behdad Esfahbod
1150d4bf12 [trak] Remove support for disabling using pseudo-font-feature 
Fixes https://github.com/harfbuzz/harfbuzz/issues/5052
 2025-02-09 04:16:57 +00:00
Behdad Esfahbod
4c06c3d606
Merge pull request #5057 from harfbuzz/coretext-font-nonbmp 
[coretext-font] Support non-BMP characters
 2025-02-09 04:01:59 +00:00
Behdad Esfahbod
fb22295311 [coretext-font] Support non-BMP in variation-selector callback 
Fixes https://github.com/harfbuzz/harfbuzz/issues/5056
 2025-02-09 03:57:47 +00:00
Behdad Esfahbod
f6bf9f697e [coretext-font] Do early-return in get_nominal_glyphs() 
Related to https://github.com/harfbuzz/harfbuzz/issues/5056
 2025-02-09 03:57:47 +00:00
Behdad Esfahbod
c5e6c885c1 [coretext-font] Support non-BMP chars in get_nominal_glyph(s) 
Part of https://github.com/harfbuzz/harfbuzz/issues/5056
 2025-02-09 03:57:47 +00:00
Behdad Esfahbod
b9675d6d40
Merge pull request #5055 from harfbuzz/trak-interpolate 
[trak] Interpolate between tracks
 2025-02-09 03:01:21 +00:00
Behdad Esfahbod
a70a30ddfd [trak] Interpolate between tracks 
Fixes https://github.com/harfbuzz/harfbuzz/issues/5054

According to Ned, this is what CoreText does.

Should add tests some time...
 2025-02-09 02:56:20 +00:00
Behdad Esfahbod
1bf0a5bc17 [trak] Streamline a bit 
Also use float math.
 2025-02-09 02:43:09 +00:00
Behdad Esfahbod
b1677e76aa [trak] Handle "out-of-range" values better 
If requested size < min-size, use the value for min-size.
If requested size > max-size, use the value for max-size.

This is the only way that makes sense. Extrapolating as we were
doing, is just wrong...

This also seems to match what CoreText does.

Adjacent to fixing https://github.com/harfbuzz/harfbuzz/issues/5054
 2025-02-09 02:05:36 +00:00
Behdad Esfahbod
691cc93976 [trak] Fix build without STAT 2025-02-06 20:08:16 +00:00
Behdad Esfahbod
b8327fe605 Fix compiler errors 2025-02-06 20:01:12 +00:00
Behdad Esfahbod
f394206b7c [ot-shape] Only apply trak'ing if STAT table present 2025-02-06 19:56:33 +00:00
Behdad Esfahbod
6e0dd8110e [trak] Default ptem to 12 
That's what CoreText does.
https://developer.apple.com/documentation/coretext/ctfontcreatewithgraphicsfont(_:_:_:_:)

Fixes https://github.com/harfbuzz/harfbuzz/issues/5048
 2025-02-06 19:56:30 +00:00
Behdad Esfahbod
ef00ac8c84
Merge pull request #5051 from harfbuzz/trak-choose 
[trak] Choose closest trak track to 0.0
 2025-02-06 17:44:55 +00:00
Behdad Esfahbod
6a37a65276 [trak] Choose closest trak track to 0.0 
Seems to be what CT does. Needs more testing to adjust better, but
for now it's better than before.
 2025-02-06 17:41:36 +00:00
Behdad Esfahbod
8efd2d85c7 [coretext] Fix CGFont leak 2025-02-06 14:38:09 +00:00
Behdad Esfahbod
b5327faf99 Fix leak 
From recent work.
 2025-02-06 14:26:37 +00:00
Behdad Esfahbod
24fe2be9c1 [subset] Fix undefined-behavior 2025-02-06 11:55:23 +00:00
Behdad Esfahbod
97db7bb2a0 [set-digest] Cosmetic 2025-02-06 11:01:59 +00:00
Behdad Esfahbod
f0d6a36612 [face] Use a macro 2025-02-06 02:51:50 +00:00
Behdad Esfahbod
7bdc6c9a5a Whitespace 2025-02-06 02:47:13 +00:00
Behdad Esfahbod
6091abcc9b [buffer] Avoid a copy in collect_codepoints() 2025-02-06 02:40:24 +00:00
Behdad Esfahbod
e5e3bc4080 [buffer] Merge two functions into a template 2025-02-06 02:37:10 +00:00
Behdad Esfahbod
403f12f766 [buffer] A few always-inline's 2025-02-06 02:33:41 +00:00
Behdad Esfahbod
4818ba9ece [set-digest] Yet another tuning 
Saves 5% on SF Pro while regressing 2% on Roboto.
 2025-02-06 01:50:12 +00:00
Behdad Esfahbod
c737829449 [OT] Remove likely() from coverage check 
Not that it matters, but because of the set-digest, this path
is not very likely() anymore.
 2025-02-06 01:42:25 +00:00
Behdad Esfahbod
3577992a5b [gsubgpos] Simplify a destruction 2025-02-06 01:30:59 +00:00
Behdad Esfahbod
d0d256bb66 [bit-page] Move a couple methods around 2025-02-06 01:26:33 +00:00
Behdad Esfahbod
a4cd11dea9 [bit-page] Micro-optimize 2025-02-06 01:13:58 +00:00
Behdad Esfahbod
0a678a9233 Fix leak 2025-02-06 01:02:44 +00:00
Behdad Esfahbod
cad264b2c9 Indent 2025-02-06 00:56:16 +00:00
Behdad Esfahbod
63c92d5a0d [set-digest] always-inline 2025-02-06 00:52:41 +00:00
Behdad Esfahbod
d712a02845 [set-digest] Back to a 3-entry, with different parameters now 
We used to use 4,0,9 for years. Now 3,0,6. It is slightly faster
and shouldn't regress common fonts.  The rationale is that OT
GSUB/GPOS are split by script, so any particular script is not
likely to use more than a few thousands of (consecutive mostly)
glyphs. The exception is CJK of course, but those have less
GSUB/GPOS work.
 2025-02-06 00:45:49 +00:00
Behdad Esfahbod
f5fe66cf5f [set-digest] Minor sanity 2025-02-06 00:36:58 +00:00
Behdad Esfahbod
4e75dfae84 Hide unused member 2025-02-06 00:14:29 +00:00