|
19802dfd
|
2022-06-16T13:50:35
|
|
Minor rename of static symbols to namespace them
|
|
|
e0a52316
|
2022-06-16T13:24:02
|
|
[draw] Lazy-allocate user-data/destroy callback vector
|
|
823f32a0
|
2022-06-15T18:50:45
|
|
[subset] Fix potential out of bounds write setting overlap flag on composite glyphs.
|
|
|
3f9226da
|
2022-06-16T09:28:47
|
|
[meta] Fix typo
|
|
1c0eeb7c
|
2022-06-15T23:05:15
|
|
Don't try to set _USE_MATH_DEFINES if already defined
|
|
|
25917c78
|
2022-06-15T23:03:22
|
|
Fix build warning when __GNUG__ isn't defined
|
|
|
e9c0a740
|
2022-06-15T16:57:16
|
|
Fix clang -Wcomma warnings
Fixes https://github.com/harfbuzz/harfbuzz/issues/3656
|
|
|
d9c18cc2
|
2022-06-15T16:50:34
|
|
[indic-table] Update
|
|
|
99a26bc1
|
2022-06-15T16:14:31
|
|
[indic-generator] Fix typo
|
|
|
2cbb7758
|
2022-06-11T08:57:21
|
|
[myanmar] Fold category P into GB
Fixes https://github.com/harfbuzz/harfbuzz/issues/3649
This actually now allows Asat after the Myanmar punctuation marks;
something I see in Wikipedia data.
|
|
|
b350e301
|
2022-06-11T08:52:11
|
|
[myanmar] Remove category D completely
Fixes https://github.com/harfbuzz/harfbuzz/issues/3651
|
|
|
8533214a
|
2022-06-11T08:49:36
|
|
[khmer] Fold category Coeng completely into category H
|
|
|
607a9fe7
|
2022-06-11T04:20:23
|
|
[indic-like] Remove category duplication
|
|
|
04851921
|
2022-06-11T03:55:23
|
|
[indic-like] Remove dependence on ot_category_t
|
|
|
14049003
|
2022-06-11T03:53:47
|
|
[indic-like] Reduce indic-dependency of khmer/myanmar even more
|
|
|
f9f0969c
|
2022-06-13T17:55:26
|
|
[subset] switch to hb_memcpy.
|
|
|
1e34852f
|
2022-06-11T01:27:33
|
|
[subset] Fix undefined behaviour.
Don't memcpy if there's nothing to copy.
|
|
|
311413f1
|
2022-06-11T01:05:57
|
|
[subset] Fix fuzzer issue.
Fixes https://oss-fuzz.com/testcase-detail/5693568490012672. new_index should be set from new_index2 when the entry is present in the map.
|
|
4ba7980b
|
2022-06-11T15:50:37
|
|
Fix the generated gobject headers
These headers are not acceptable to modern compilers.
gcc says things like:
/usr/include/harfbuzz/hb-gobject-enums.h:100:1: warning:
function declaration isn’t a prototype [-Wstrict-prototypes]
100 | hb_gobject_ot_metrics_tag_get_type () G_GNUC_CONST;
|
|
79bb3b52
|
2022-06-11T14:31:28
|
|
[docs] fix typo
|
|
|
98116e5c
|
2022-06-11T03:42:36
|
|
[myanmar] Fix comments on categories
https://github.com/harfbuzz/harfbuzz/pull/3648#discussion_r894955430
|
|
|
02016914
|
2022-06-10T17:24:19
|
|
[indic-generator] Remove unnecessary Myanmar category=D overrides
https://github.com/harfbuzz/harfbuzz/pull/3648#discussion_r894685106
|
|
|
937c8780
|
2022-06-10T17:20:15
|
|
[indic-generator] Remove unnecessary override for Myanmar U+1039
https://github.com/harfbuzz/harfbuzz/pull/3648#discussion_r894762535
|
|
|
9504037c
|
2022-06-10T17:13:16
|
|
[indic-generator] Remove three unneeded Myanmar overrides U+AA74-6
These three characters have Indic_Syllabic_Category=Consonant_Placeholder. The
original evidence that prompted these overrides says they can take tone marks.
They are not subjoined: Khamti Shan apparently does not use subjoined
characters at all. Therefore, PLACEHOLDER is good enough and these need not be
overridden to C.
https://www.unicode.org/L2/L2008/08276-khamti-proposal.pdf
https://github.com/harfbuzz/harfbuzz/pull/3648#discussion_r894640713
|
|
|
02eb6606
|
2022-06-10T17:10:42
|
|
[indic-generator] Remove redundant PLACEHODER characters overrides
https://github.com/harfbuzz/harfbuzz/pull/3648#discussion_r894631922
|
|
|
e16669ce
|
2022-06-10T17:05:35
|
|
[indic-generator] Remove redundant override of U+2010 / U+2011
https://github.com/harfbuzz/harfbuzz/pull/3648#discussion_r894630596
|
|
|
bb255cd9
|
2022-06-10T17:03:52
|
|
[indic-generator] Remove redundant override of U+0980
https://github.com/harfbuzz/harfbuzz/pull/3648#discussion_r894627064
|
|
|
30d8c87d
|
2022-06-10T16:56:15
|
|
[myanmar] Document Medials
https://github.com/harfbuzz/harfbuzz/pull/3648#discussion_r894532676
|
|
|
e1826c37
|
2022-06-10T16:53:11
|
|
[khmer] Reuse OT_H for OT_Coeng
https://github.com/harfbuzz/harfbuzz/pull/3648#discussion_r894517417
|
|
|
eb2f2e31
|
2022-06-10T16:47:59
|
|
[indic-generator] Update comment re U+104E
https://github.com/harfbuzz/harfbuzz/pull/3648#pullrequestreview-1002150048
|
|
|
0daafefd
|
2022-06-10T10:55:49
|
|
Merge branch 'main' into indic-cleanup
|
|
|
148283d0
|
2022-06-10T07:48:39
|
|
[syllabic] Use a buffer scratch-flag for has-broken-syllable
|
|
|
6997d10b
|
2022-06-10T07:43:33
|
|
[arabic] Remove a couple TODO items
|
|
|
e1575f23
|
2022-06-10T07:41:39
|
|
[iter] Remove a few TODO items
|
|
|
689c7753
|
2022-06-10T07:37:45
|
|
[buffer] Actually remove TODO item
In reality, one side is smaller and one side is larger. The existing code
handles that just fine.
|
|
|
d09e962b
|
2022-06-10T07:35:16
|
|
[buffer] Update a TODO item
|
|
|
b3f689e7
|
2022-06-10T07:31:47
|
|
[serializer] Remove TODO that's not gonna happen
|
|
|
10a8cc28
|
2022-06-10T07:31:06
|
|
[normalizer] Remove a TODO that's not going to happen
|
|
|
7635568f
|
2022-06-10T07:29:15
|
|
[font] Remove a TODO item that's not gonna happen
|
|
|
bd453de7
|
2022-06-10T07:25:48
|
|
[unicode] Accept a couple hacks as permanent
|
|
|
39c132a6
|
2022-06-10T07:12:39
|
|
[indic/myanmar] Simplify compare functions
|
|
|
a5bcd856
|
2022-06-10T07:03:02
|
|
[indic] Update comment re category listing
|
|
|
10a54851
|
2022-06-10T06:34:56
|
|
[indic/myanmar] Move is_consonant to .cc files
|
|
|
1c657460
|
2022-06-10T06:29:45
|
|
[indic] Expand MEDIAL_FLAGS
|
|
|
9e3917f6
|
2022-06-10T06:20:56
|
|
[indic] Move a couple of functions to .cc file
|
|
|
165ef55e
|
2022-06-10T06:20:10
|
|
[indic-generator] Move INDIC_COMBINE_CATEGORIES here
|
|
|
b030dd9e
|
2022-06-10T06:12:13
|
|
[indic-table] Minor rename
|
|
|
d414fb33
|
2022-06-10T00:50:47
|
|
[indic/khmer/myanmar] Add static_assert that categories match OT_*
|
|
|
15ea4ccb
|
2022-06-09T17:47:46
|
|
[indic-like] Add note about replicated values
|
|
|
3289e815
|
2022-06-09T17:46:15
|
|
[indic] Use categories from the machine
|
|
|
37217fc9
|
2022-06-09T16:43:50
|
|
[indic-generator/myanmar] Move most Myanmar category overrides to generator
|
|
|
c136227f
|
2022-06-09T13:36:19
|
|
[indic-generator/khmer] Move Khmer overrides to generator
|
|
|
40aa4e83
|
2022-06-09T13:30:00
|
|
[indic/khmer/myanmar] Add Khmer/Myanmar categories to indic_category_t
|
|
|
25793075
|
2022-06-09T13:11:46
|
|
[indic-generator] Move Khmer/Myanmar vowel categories to the generator
|
|
|
a6c82d4b
|
2022-06-09T13:04:28
|
|
[myanmar] Simplify to not use position info from the indic table for shaping
|
|
|
10cd8ac0
|
2022-06-09T12:27:31
|
|
[indic-generator] Move matra category overrides to generator
|
|
|
c4e4f1d3
|
2022-06-09T11:56:57
|
|
[indic-generator] Move SMVD position overrides to generator
|
|
|
2963154c
|
2022-06-09T11:49:02
|
|
[indic-generator] Add a couple comments
|
|
|
91d6f45b
|
2022-06-09T07:34:44
|
|
[indic-generator] Move some position overrides to the generator
|
|
|
0ec4dcb9
|
2022-06-09T07:33:43
|
|
[indic-generator] Ouch
Not sure how this was passing tests still.
|
|
|
f0269e0f
|
2022-06-09T07:10:47
|
|
[indic-generator] Move Ra handling to the generator
|
|
|
419d2146
|
2022-06-09T07:01:14
|
|
[indic-generator] Cap off what categories have positions
This was left off of the commit moving Indic categories to the generator.
It didn't fail any tests, but adding it back because it has implications
possibly.
|
|
|
e1d965d5
|
2022-06-09T06:48:25
|
|
[indic-generator] Move position mapping to generator
|
|
|
49075140
|
2022-06-09T06:33:51
|
|
[indic-generator] Move category overrides to generator
|
|
|
58eeb3a1
|
2022-06-09T05:34:49
|
|
[indic-generator] Move category mapping to generator
|
|
|
899ca243
|
2022-06-05T01:52:31
|
|
[myanmar] Remove duplication of categories in the Myanmar shaper
|
|
|
ce0528c0
|
2022-06-05T01:34:10
|
|
[khmer] Remove duplication of categories in the Khmer shaper
|
|
|
17c80035
|
2022-06-09T04:17:38
|
|
Revert "[cplusplus] Internally allow using hb_unique_ptr with hb_free()"
This reverts commit b5f621b08d56c15832ab4f588a0673fce03305a0.
A build was failing with clang 14 for Firefox apparently, sigh.
https://github.com/harfbuzz/harfbuzz/issues/3647
|
|
|
9fc9b1ec
|
2022-06-08T12:35:43
|
|
[layout] Minor, add using Lookup to GSUB/GPOS
|
|
|
d4ddb3ac
|
2022-06-08T11:45:14
|
|
Comments typos
|
|
|
4119f73c
|
2022-06-08T06:34:48
|
|
[subset/layout] Rename dispatch_closure_lookups_recurse_func to dispatch_recurse_func<>
|
|
|
c13ff395
|
2022-06-08T06:25:23
|
|
[layout] Rename apply_recurse_func to specialization of dispatch_recurse_func
|
|
|
6a1edb8c
|
2022-06-08T11:38:17
|
|
[set-digest] One more rename
|
|
|
6453737b
|
2022-06-08T11:37:12
|
|
[set-digest] Rename lowest_bits to bits_pattern
|
|
|
2a061cb9
|
2022-06-08T11:35:50
|
|
[set-digest] Improve documentation
|
|
|
42051fe1
|
2022-06-08T04:00:21
|
|
[layout] s/inplace/always_inplace/g
|
|
|
21346af0
|
2022-06-07T14:41:39
|
|
[layout-cache] Adjust cost-function for recent change
|
|
|
bfee6839
|
2022-06-07T11:41:05
|
|
[layout-cache] Cache lookahead, not input, classdef in ChainContextFormat2
From the commit:
+ /* For ChainContextFormat2 we cache the LookaheadClassDef instead of InputClassDef.
+ * The reason is that most heavy fonts want to identify a glyph in context and apply
+ * a lookup to it. In this scenario, the length of the input sequence is one, whereas
+ * the lookahead / backtrack are typically longer. The one glyph in input sequence is
+ * looked-up below and no input glyph is looked up in individual rules, whereas the
+ * lookahead and backtrack glyphs are tried. Since we match lookahead before backtrack,
+ * we should cache lookahead. This decisions showed a 20% improvement in shaping of
+ * the Gulzar font.
https://github.com/harfbuzz/harfbuzz/pull/3636
|
|
|
39820af7
|
2022-06-07T10:18:38
|
|
[layout] Add HB_NO_OT_LAYOUT_LOOKUP_CACHE to disable caching lookups
Enabled when optimize-size profile is enabled.
|
|
|
845279c3
|
2022-06-07T16:40:22
|
|
Merge pull request #3636 from harfbuzz/classdef-cache
Classdef cache
|
|
|
b59e25f2
|
2022-06-07T09:27:09
|
|
[cff] Try fixing Heap-buffer-overflow in CFF::Charset::collect_glyph_to_sid_map
Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=47790
|
|
|
c8fb048f
|
2022-06-07T09:20:27
|
|
[gsubgpos] Document caching
|
|
|
5963cf44
|
2022-06-07T09:12:45
|
|
[gsubgpos] Merge cache_enter and cache_leave entry points
Saves a pointer per subtable
|
|
|
d4c09e9a
|
2022-06-07T09:03:30
|
|
[gsubgpos] Remove apply_cached() entry point
Just use a bool to apply()
|
|
|
b96622d1
|
2022-06-05T02:45:41
|
|
[layout] Use a cache for main input ClassDef of (Chain)ContextLookupFormat2
This commit adds a per-lookup caching infrastructure to GSUB/GPOS, and
uses it to cache input ClassDef.get_class value for (Chain)ContextLookupFormat2.
For fonts heavy on use of heave class-based2 context matching, this shows
a good speedup. For NotoNastaliqUrdu for example, I observe 17% speedup.
Unfortunately not many other lookups can use a cache like this :(.
https://github.com/harfbuzz/harfbuzz/pull/3636
|
|
4266f4e2
|
2022-06-03T12:06:56
|
|
Fix check-* scripts when harfbuzz is a subproject
When harfbuzz is a subproject paths are in the form
"subprojects/harfbuzz/src/...". Instead of removing "src/" prefix, take
the absolute path and make it relative to current source dir.
This fix regression introduced in
https://github.com/harfbuzz/harfbuzz/pull/3394.
|
|
|
15543f70
|
2022-06-04T10:55:50
|
|
[indic-like] Move allocation of syllable() buffer var to shapers that use it
In indic, we don't have a pause location release the var.
|
|
|
104dc85a
|
2022-06-04T06:56:35
|
|
[buffer] Add try_allocate for buffer variables
|
|
|
b5f621b0
|
2022-06-05T07:52:51
|
|
[cplusplus] Internally allow using hb_unique_ptr with hb_free()
...for arbitrary types.
|
|
|
f18eb000
|
2022-06-05T04:17:05
|
|
[buffer] Mark a variable as unused
Weird. Somehow our HB_TINY bot started erring as this var being unused
in non-debug builds. Not sure why now...
In file included from src/hb-ot-map.hh:32:0,
from src/hb-ot-shape.hh:32,
from src/hb-aat-layout.hh:32,
from src/hb-aat-layout.cc:30,
from src/harfbuzz.cc:1:
src/hb-buffer.hh: In member function ‘void hb_buffer_t::assert_var(unsigned int, unsigned int)’:
src/hb-buffer.hh:192:18: error: unused variable ‘bits’ [-Werror=unused-variable]
unsigned int bits = (1u<<end) - (1u<<start);
^~~~
cc1plus: some warnings being treated as errors
Error: Process completed with exit code 1.
|
|
|
e6409d39
|
2022-06-05T06:57:37
|
|
Revert "[layout] Use a cache for main input ClassDef of (Chain)ContextLookups"
This reverts commit 57d1c08739d0acd94b96da2f9d5dd6c0ff3b3722.
Err. This was an accident.
|
|
|
57d1c087
|
2022-06-05T02:45:41
|
|
[layout] Use a cache for main input ClassDef of (Chain)ContextLookups
|
|
|
92e81ab1
|
2022-06-05T01:14:32
|
|
[indic/khmer/myanmar] Shift category numbers around to avoid overlap
Fixes https://github.com/harfbuzz/harfbuzz/issues/3632
|
|
|
d6dbc0d1
|
2022-06-05T00:58:53
|
|
[gi/ft] Silence warnings
gi doesn't understand that FT_Face is a pointer, and not bare struct.
So it skips these APIs anyway. Mark skip to silense the warning.
../src/hb-ft.cc:242: Warning: HarfBuzz: hb_ft_font_get_face: return value: Invalid non-constant return of bare structure or union; register as boxed type or (skip)
../src/hb-ft.cc:264: Warning: HarfBuzz: hb_ft_font_lock_face: return value: Invalid non-constant return of bare structure or union; register as boxed type or (skip)
|
|
|
f78a2509
|
2022-06-05T00:55:35
|
|
[gi] Remove Xconstructor annotations
|
|
|
cb961eac
|
2022-06-05T00:44:07
|
|
[configs] Fix builds with HB_NO_BUFFER_SERIALIZE but not HB_NO_BUFFER_VERIFY
https://github.com/harfbuzz/harfbuzz/commit/e986c12075a69300a5e114fe139ae5acd762ef1b#commitcomment-75339317
|
|
|
67852504
|
2022-06-04T08:45:03
|
|
[indic/myanmar] No-op update to machines
|
|
|
20a61ca4
|
2022-06-04T08:35:22
|
|
[myanmar] Fold D category into GB
Trying to free a byte the buffer vars, for caching of lookups.
Part of https://github.com/harfbuzz/harfbuzz/issues/3633
|
|
|
f9b643f6
|
2022-06-04T07:29:40
|
|
[layout] s/hb_get_subtables_context_t/hb_accelerate_subtables_context_t/g
|
|
|
d4dfb8c1
|
2022-06-04T06:50:38
|
|
[arabic] Free up buffer variable earlier
|