Commit Graph

13416 Commits

Author SHA1 Message Date
David Corbett 78c5ae3979 [indic] Remove remnants of Sinhala 2022-06-25 13:49:18 -04:00
Behdad Esfahbod 0cc948b96c [use] Switch Sinhala to USE
https://github.com/harfbuzz/harfbuzz/issues/1044
2022-06-25 13:48:43 -04:00
David Corbett 6059828769 [use] Reintroduce the HVM class for U+0DCA 2022-06-25 13:48:43 -04:00
David Corbett 1555b30081 Add U+25CC to lone Robatic but not after U+17D9 2022-06-25 11:02:39 -06:00
David Corbett 0f15cb12de [indic-table] Fix block headers 2022-06-25 11:02:39 -06:00
David Corbett e35cfb4bde Document the subsetter argument of record-test.sh 2022-06-25 11:02:39 -06:00
Garret Rieger 2674962cf5 [repacker] Add comment to graph class. 2022-06-24 17:16:50 -06:00
Garret Rieger 81a2dd0e80 [repacker] Update Makefile for repacker re-org. 2022-06-24 17:16:50 -06:00
Garret Rieger 7078560e33 [repacker] extract graph serialization code into a seperate file. 2022-06-24 17:16:50 -06:00
Garret Rieger 20b02a672d [repacker] Begin splitting up the repacker implementation into several files. 2022-06-24 17:16:50 -06:00
Behdad Esfahbod ad2ab1ddb4 [indic] Clear syllables at the end of GSUB 2022-06-24 11:08:35 -06:00
Behdad Esfahbod 1f462804d1 [README.mingw.md] Add link to issue with further instructions 2022-06-23 15:35:38 -06:00
Behdad Esfahbod 8bfb3e9df2 [indic] Disable vowel-constraints under uniscribe-bug-compatible 2022-06-23 13:28:07 -06:00
Behdad Esfahbod f8d052df6d [ansi-print] Remove impossible condition 2022-06-22 18:35:48 -06:00
Behdad Esfahbod 251320ea22 [ansi-print] Whitespace 2022-06-22 16:44:07 -06:00
Behdad Esfahbod 0d59d7952c [ansi-print] Precision 2022-06-22 16:41:02 -06:00
Behdad Esfahbod c695a09154 [ansi-print] Reorder cases; harmless 2022-06-22 16:33:37 -06:00
Behdad Esfahbod ae0fce31cd [ansi-print] Fix quadrants 2022-06-22 16:31:21 -06:00
Behdad Esfahbod 2ebaf0c5ba [ansi-print] Fix unicolor detection 2022-06-22 16:22:02 -06:00
Behdad Esfahbod a4db80ca0a [ansi-print] Fix color calculation
Still something's wrong.
2022-06-22 16:13:53 -06:00
Behdad Esfahbod 1abec5cd0d [CONFIG.md] Grammar 2022-06-21 13:39:16 -06:00
Behdad Esfahbod 0ab08a8bbb [doap] Update 2022-06-21 13:19:08 -06:00
Behdad Esfahbod a5cf1a8738 Another null adjustment 2022-06-20 18:01:39 -06:00
Behdad Esfahbod a7960bdfb0 [config] Add HB_NO_LANGUAGE_LONG and enable in TINY profile
Disables 3letter language tags and more complex ones.

Fixes https://github.com/harfbuzz/harfbuzz/issues/3664
2022-06-20 17:55:28 -06:00
Behdad Esfahbod 0d03123350 Mark a null variable as const 2022-06-20 16:51:35 -06:00
Behdad Esfahbod 91d00ab722 [ucd] Update 2022-06-20 13:36:01 -06:00
Khaled Hosny a15ad778fe [arabic-fallback] Generate PUA table from data
Uses packtab for more compact arrays.
2022-06-19 12:29:10 -06:00
Behdad Esfahbod abc0685749
Merge pull request #3063 from harfbuzz/arabic-pua
Arabic PUA shaping
2022-06-19 11:05:17 -06:00
Behdad Esfahbod 7ec4a556d9 [normalize] Cosmetic
I didn't know this syntax is allowed in old C++.
2022-06-19 11:01:45 -06:00
Behdad Esfahbod 8c27c51c27 [arabic-pua] Rename symbols 2022-06-19 10:47:38 -06:00
Behdad Esfahbod 7698962911 [arabic-fallback] Disable PUA shaping under HB_NO_OT_SHAPER_ARABIC_FALLBACK 2022-06-19 10:42:06 -06:00
Behdad Esfahbod 4520911429 [arabic-fallback] Fix warning 2022-06-19 10:42:06 -06:00
Behdad Esfahbod 55350377b0 [cmap/ft] Only map 0xF000 range if font_page is NONE 2022-06-19 10:42:06 -06:00
Behdad Esfahbod 41a079bdec [arabic-fallback] Make win1256 code build again
Humm. Untested.
2022-06-19 10:42:06 -06:00
Behdad Esfahbod 1db6fddb24 [arabic-fallback.hh] Hook up 3-letter ligatures 2022-06-19 10:42:06 -06:00
Behdad Esfahbod 20e9f0b1d2 [arabic-fallback] Add the component loop
Should be able to support 3-letter ligatures now.  Hooking up next.
2022-06-19 10:42:06 -06:00
Behdad Esfahbod 9684d2d8aa [arabic-fallback] More baby steps 2022-06-19 10:42:06 -06:00
Behdad Esfahbod 08715d75e0 [arabic-fallback] Another baby-step 2022-06-19 10:42:06 -06:00
Behdad Esfahbod 15dd34b515 [arabic-fallback] Another minor rename towards supporting 3-letter ligatures 2022-06-19 10:42:06 -06:00
Behdad Esfahbod d86effa4a6 [arabic-fallback] Rename; towards supporting 3-letter ligatures 2022-06-19 10:42:06 -06:00
Behdad Esfahbod 8978a18f31 [arabick-fallback] Apply mark ligatures 2022-06-19 10:42:06 -06:00
Behdad Esfahbod 16c2371352 [arabick-fallback] Apply mark ligatures 2022-06-19 10:42:06 -06:00
Khaled Hosny 7f362196c5 [arabic] Split ligature array
Generate marks and 3-component ligatures in separate arrays. The new
arrays are unused currently.
2022-06-19 10:42:06 -06:00
Khaled Hosny 6e29060af2 [arabic] Add mapping files for the PUA encoding
Unused right now.
2022-06-19 10:42:06 -06:00
Khaled Hosny dfc5e5a27d [test] Skip glyph positions for test failing on CI
We are not interested in glyph positioning for this test, and the
FreeType version on some CI machines is giving some different glyph
advances here.
2022-06-19 10:42:06 -06:00
Khaled Hosny 189e8c326e [ft] Remap legacy Arabic PUA codepoints
Similar to what ot font finctions does, to support Support legacy
pre-OpenType Windows 3.1-era fonts.
2022-06-19 10:42:05 -06:00
Khaled Hosny 69cbd365cc WIP: manually add data entries for PUA ligatures
Just a test, I’m not entirely convinced we should do this.
2022-06-19 10:42:05 -06:00
Khaled Hosny c3f590bb1e [arabic] Support legacy PUA shaping
Support legacy pre-OpenType Windows 3.1-era fonts, by remapping PUA code
points in cmap table and letting our fallback shaper build the GSUB
table.

Uniscribe applies also mset-like substitution, but our fallback mark
positioning gives better results, so this is not implemented.
2022-06-19 10:42:05 -06:00
Behdad Esfahbod b172f88c7d Fix a warning
D:\a\harfbuzz\harfbuzz\src\hb-bit-set-invertible.hh(83): warning C4805: '^': unsafe mix of type 'uint32_t' and type 'const bool' in operation
[870/1075] Compiling C++ object src/harfbuzz.dll.p/hb-face.cc.obj
cl : Command line warning D9025 : overriding '/EHs' with '/EHs-'
cl : Command line warning D9025 : overriding '/EHc' with '/EHc-'
2022-06-18 14:00:59 -06:00
Garret Rieger af74ab452f [repack] always run the sort in repack.
This is needed to ensure virtual link ordering constraints are respected when repack is being called from fontTools. For subset usage, repack won't be called if the graph doesn't already overflow so this change doesn't cause any extra work.
2022-06-16 15:17:58 -06:00