Commit Graph

10495 Commits

Author SHA1 Message Date
Behdad Esfahbod b030dd9e88 [indic-table] Minor rename 2022-06-10 06:12:13 -06:00
Behdad Esfahbod d414fb3328 [indic/khmer/myanmar] Add static_assert that categories match OT_* 2022-06-10 00:50:47 -06:00
Behdad Esfahbod 15ea4ccb04 [indic-like] Add note about replicated values 2022-06-09 17:47:46 -06:00
Behdad Esfahbod 3289e81532 [indic] Use categories from the machine 2022-06-09 17:46:15 -06:00
Behdad Esfahbod 37217fc9be [indic-generator/myanmar] Move most Myanmar category overrides to generator 2022-06-09 17:27:43 -06:00
Behdad Esfahbod c136227f57 [indic-generator/khmer] Move Khmer overrides to generator 2022-06-09 13:38:34 -06:00
Behdad Esfahbod 40aa4e8320 [indic/khmer/myanmar] Add Khmer/Myanmar categories to indic_category_t 2022-06-09 13:30:00 -06:00
Behdad Esfahbod 25793075e2 [indic-generator] Move Khmer/Myanmar vowel categories to the generator 2022-06-09 13:11:46 -06:00
Behdad Esfahbod a6c82d4b8c [myanmar] Simplify to not use position info from the indic table for shaping 2022-06-09 13:04:28 -06:00
Behdad Esfahbod 10cd8ac0e5 [indic-generator] Move matra category overrides to generator 2022-06-09 12:27:31 -06:00
Behdad Esfahbod c4e4f1d387 [indic-generator] Move SMVD position overrides to generator 2022-06-09 11:58:37 -06:00
Behdad Esfahbod 2963154c15 [indic-generator] Add a couple comments 2022-06-09 11:53:24 -06:00
Behdad Esfahbod 91d6f45bc9 [indic-generator] Move some position overrides to the generator 2022-06-09 11:52:56 -06:00
Behdad Esfahbod 0ec4dcb93d [indic-generator] Ouch
Not sure how this was passing tests still.
2022-06-09 11:52:25 -06:00
Behdad Esfahbod f0269e0f1b [indic-generator] Move Ra handling to the generator 2022-06-09 11:52:03 -06:00
Behdad Esfahbod 419d2146c2 [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.
2022-06-09 11:51:38 -06:00
Behdad Esfahbod e1d965d527 [indic-generator] Move position mapping to generator 2022-06-09 11:51:15 -06:00
Behdad Esfahbod 4907514026 [indic-generator] Move category overrides to generator 2022-06-09 11:50:30 -06:00
Behdad Esfahbod 58eeb3a180 [indic-generator] Move category mapping to generator 2022-06-09 11:49:57 -06:00
Behdad Esfahbod 899ca24387 [myanmar] Remove duplication of categories in the Myanmar shaper 2022-06-09 07:02:54 -06:00
Behdad Esfahbod ce0528c0ff [khmer] Remove duplication of categories in the Khmer shaper 2022-06-09 07:02:54 -06:00
Behdad Esfahbod 17c80035ad Revert "[cplusplus] Internally allow using hb_unique_ptr with hb_free()"
This reverts commit b5f621b08d.

A build was failing with clang 14 for Firefox apparently, sigh.

https://github.com/harfbuzz/harfbuzz/issues/3647
2022-06-09 04:17:38 -06:00
Behdad Esfahbod 9fc9b1ece4 [layout] Minor, add using Lookup to GSUB/GPOS 2022-06-08 12:37:31 -06:00
Behdad Esfahbod d4ddb3acf8 Comments typos 2022-06-08 11:45:33 -06:00
Behdad Esfahbod 4119f73c21 [subset/layout] Rename dispatch_closure_lookups_recurse_func to dispatch_recurse_func<> 2022-06-08 18:42:09 +01:00
Behdad Esfahbod c13ff39520 [layout] Rename apply_recurse_func to specialization of dispatch_recurse_func 2022-06-08 18:42:09 +01:00
Behdad Esfahbod 6a1edb8c97 [set-digest] One more rename 2022-06-08 11:38:17 -06:00
Behdad Esfahbod 6453737b0e [set-digest] Rename lowest_bits to bits_pattern 2022-06-08 11:37:12 -06:00
Behdad Esfahbod 2a061cb9cc [set-digest] Improve documentation 2022-06-08 11:35:50 -06:00
Behdad Esfahbod 42051fe18a [layout] s/inplace/always_inplace/g 2022-06-08 04:00:21 -06:00
Behdad Esfahbod 21346af01d [layout-cache] Adjust cost-function for recent change 2022-06-07 14:41:39 -06:00
Behdad Esfahbod bfee6839b0 [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
2022-06-07 11:41:05 -06:00
Behdad Esfahbod 39820af72f [layout] Add HB_NO_OT_LAYOUT_LOOKUP_CACHE to disable caching lookups
Enabled when optimize-size profile is enabled.
2022-06-07 10:27:55 -06:00
Behdad Esfahbod 845279c34c
Merge pull request #3636 from harfbuzz/classdef-cache
Classdef cache
2022-06-07 16:40:22 +01:00
Behdad Esfahbod b59e25f25e [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
2022-06-07 09:27:44 -06:00
Behdad Esfahbod c8fb048f79 [gsubgpos] Document caching 2022-06-07 09:20:27 -06:00
Behdad Esfahbod 5963cf4469 [gsubgpos] Merge cache_enter and cache_leave entry points
Saves a pointer per subtable
2022-06-07 09:17:14 -06:00
Behdad Esfahbod d4c09e9a87 [gsubgpos] Remove apply_cached() entry point
Just use a bool to apply()
2022-06-07 09:17:14 -06:00
Behdad Esfahbod b96622d15c [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
2022-06-07 09:14:38 -06:00
Xavier Claessens 4266f4e29a 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.
2022-06-06 16:53:29 +01:00
Behdad Esfahbod 15543f70e0 [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.
2022-06-05 08:15:47 -06:00
Behdad Esfahbod 104dc85a22 [buffer] Add try_allocate for buffer variables 2022-06-05 08:15:46 -06:00
Behdad Esfahbod b5f621b08d [cplusplus] Internally allow using hb_unique_ptr with hb_free()
...for arbitrary types.
2022-06-05 08:15:31 -06:00
Behdad Esfahbod f18eb000d3 [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.
2022-06-05 07:06:07 -06:00
Behdad Esfahbod e6409d3905 Revert "[layout] Use a cache for main input ClassDef of (Chain)ContextLookups"
This reverts commit 57d1c08739.

Err. This was an accident.
2022-06-05 06:57:37 -06:00
Behdad Esfahbod 57d1c08739 [layout] Use a cache for main input ClassDef of (Chain)ContextLookups 2022-06-05 06:54:19 -06:00
Behdad Esfahbod 92e81ab1df [indic/khmer/myanmar] Shift category numbers around to avoid overlap
Fixes https://github.com/harfbuzz/harfbuzz/issues/3632
2022-06-05 01:14:32 -06:00
Behdad Esfahbod d6dbc0d175 [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)
2022-06-05 00:58:53 -06:00
Behdad Esfahbod f78a25098a [gi] Remove Xconstructor annotations 2022-06-05 00:55:35 -06:00
Behdad Esfahbod cb961eac5c [configs] Fix builds with HB_NO_BUFFER_SERIALIZE but not HB_NO_BUFFER_VERIFY
e986c12075 (commitcomment-75339317)
2022-06-05 00:45:29 -06:00
Behdad Esfahbod 67852504f6 [indic/myanmar] No-op update to machines 2022-06-04 09:27:45 -06:00
Behdad Esfahbod 20a61ca447 [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
2022-06-04 09:27:37 -06:00
Behdad Esfahbod f9b643f6b2 [layout] s/hb_get_subtables_context_t/hb_accelerate_subtables_context_t/g 2022-06-04 09:27:31 -06:00
Behdad Esfahbod d4dfb8c1f1 [arabic] Free up buffer variable earlier 2022-06-04 09:27:23 -06:00
Behdad Esfahbod cc7ebb0ffa Remove remaining mention to complex shapers in the code
https://github.com/harfbuzz/harfbuzz/pull/3628#issuecomment-1146248037
2022-06-04 05:42:58 -06:00
Behdad Esfahbod b39b5f2f31 [name] Implement approximate language matching
Very rudimentary.

Fixes https://github.com/harfbuzz/harfbuzz/issues/3354
2022-06-03 13:00:33 +01:00
Behdad Esfahbod 40d7d56e53 [subset-input] Minor move 2022-06-03 05:40:18 -06:00
Behdad Esfahbod 40193adbfc [subset] Update default features list 2022-06-03 05:35:19 -06:00
Behdad Esfahbod 5bfb0b721c Rename s/shape-complex/shaper/g 2022-06-03 10:30:34 +01:00
Behdad Esfahbod 44be1e5dfb s/SHAPE_COMPLEX/SHAPER/g 2022-06-03 10:30:34 +01:00
Behdad Esfahbod 6fbb552156 s/FLAG_COMPLEX/FLAG_SHAPER/g 2022-06-03 10:30:34 +01:00
Behdad Esfahbod 6d9e94d2b8 s/hb_ot_shape_complex_categorize/hb_ot_shaper_categorize/g 2022-06-03 10:30:34 +01:00
Behdad Esfahbod a560182cb3 s/complex_var/ot_shaper_var/g 2022-06-03 10:30:34 +01:00
Behdad Esfahbod 13fbed29e4 s/HB_OT_SHAPE_COMPLEX_MAX_COMBINING_MARKS/HB_OT_SHAPE_MAX_COMBINING_MARKS/g 2022-06-03 10:30:34 +01:00
Behdad Esfahbod 44a7b3b773 s/ot_complex_shaper/ot_shaper/g 2022-06-03 10:30:34 +01:00
Behdad Esfahbod e5161977a4 s/COMPLEX_SHAPER/OT_SHAPER/g 2022-06-03 10:30:34 +01:00
Behdad Esfahbod f3a8b7f36b [algs] Test hb_hash() 2022-06-03 02:21:46 -06:00
Behdad Esfahbod 215a0afad1 [algs] Remove unused hb_coerce() 2022-06-03 01:48:46 -06:00
Behdad Esfahbod 5dc12d7d8d [cmap] Rewrite set_for() slightly 2022-06-03 01:37:02 -06:00
Behdad Esfahbod 9552955e08 Add an unlikely 2022-06-03 01:33:01 -06:00
Behdad Esfahbod 88f00ecb84 [map] Fix iter_ref () and test it 2022-06-03 01:30:27 -06:00
Behdad Esfahbod a42c624fca Convert one final use of hashmap to unique_ptr 2022-06-03 01:22:34 -06:00
Behdad Esfahbod f13a79548f [subset] Convert another use of hashmap to unique_ptr 2022-06-03 01:17:20 -06:00
Behdad Esfahbod 25f57230d5 [map] Return references from new iter_ref() 2022-06-03 01:14:35 -06:00
Behdad Esfahbod a7a688616a [cmap] Convert another map use to unique_ptr 2022-06-02 18:59:15 -06:00
Behdad Esfahbod 997d9cc466 [map] Make unique_ptr hashable 2022-06-02 18:47:42 -06:00
Behdad Esfahbod 8bb2a3326e [map] Remove unneeded assignment 2022-06-02 15:18:23 -06:00
Behdad Esfahbod d7785a6da0 [cplusplus] Add unique_ptr 2022-06-02 12:43:04 -06:00
Behdad Esfahbod bca710e8ad [gsubgpos] Use map has() instead of get() when appropriate 2022-06-02 12:06:25 -06:00
Behdad Esfahbod e9407a2bd2 Use shared_ptr<hb_set_t> in one place
See if valgrind is happy...
2022-06-02 12:00:43 -06:00
Behdad Esfahbod a42a703cb6 [shared_ptr] Clear p in destructor 2022-06-02 12:00:43 -06:00
Behdad Esfahbod f0a0dcad70 [test-map] Test hashing shared_ptr 2022-06-02 11:25:56 -06:00
Behdad Esfahbod 4c1b5d9ece Whitespace 2022-06-02 11:25:11 -06:00
Behdad Esfahbod b9230c5425 [map] Fix has() 2022-06-02 11:23:38 -06:00
Behdad Esfahbod 97ea10a63a Remove old nullptr_t hacks
Were used for hashmap before.
2022-06-02 11:23:38 -06:00
Behdad Esfahbod 3f78a71ca0 [map] Finally! Just can usd hb_hashmap_t<obj_t, obj_t>
Yay!
2022-06-02 11:23:37 -06:00
Behdad Esfahbod 0ccab339f9 [map] Remove invalid-key template arguments since unused 2022-06-02 11:23:37 -06:00
Behdad Esfahbod 3f6a8f69a0 [map] Remove invalid-key special-casing
Can override invalid-key value now.
2022-06-02 11:23:37 -06:00
Behdad Esfahbod 5328b73fba [map] Reduce map item size again 2022-06-02 11:23:37 -06:00
Behdad Esfahbod 4f58ae60eb [map] Keep is_used, is_tombstone as booleans 2022-06-02 11:23:37 -06:00
Behdad Esfahbod 7ec3aad20f [shared_ptr] Fix hb_hash() crash on nullptr 2022-06-02 18:22:05 +01:00
Behdad Esfahbod 4d646773cf [cplusplus] Make .reference() return T* 2022-06-02 18:22:05 +01:00
Behdad Esfahbod 7e7a4a8f05 [cplusplus] Fix build 2022-06-02 18:22:05 +01:00
Behdad Esfahbod a089d91fda [hash] Adjust hash for shared_ptr, implement it for std::hash 2022-06-02 18:22:05 +01:00
Behdad Esfahbod e037325efb [hash] Remove custom hash, rely on std::hash 2022-06-02 18:22:05 +01:00
Behdad Esfahbod 0d3d5b62ae [cplusplus] Adjustments 2022-06-02 18:22:05 +01:00
Behdad Esfahbod 0b35940a72 Make hb::shared_ptr hashable 2022-06-02 18:22:05 +01:00
Behdad Esfahbod 3817bdfd7f [hb.hh] Include hb-cplusplus.hh 2022-06-02 18:22:05 +01:00
Behdad Esfahbod e0f3cab246 [cplusplus] Add hb-cplusplus.hh
Fixes https://github.com/harfbuzz/harfbuzz/issues/2152
2022-06-02 18:22:05 +01:00
Garret Rieger 98aaecd397 [subset] fix data race touching Crap() in cff subsetting. 2022-06-01 22:50:35 +01:00