Commit Graph

12849 Commits

Author SHA1 Message Date
Behdad Esfahbod 42b119edb5 [hmtx] Remove other subset-specific method from hmtx accelerator 2022-02-12 14:22:35 -06:00
Behdad Esfahbod 9ba520b5d1 [hmtx] Remove one subset-specific method from hmtx accelerator
Looks ugly but is the right thing to do.
2022-02-12 14:22:35 -06:00
Behdad Esfahbod af407dd24d Add a fuzzer font 2022-02-12 14:22:35 -06:00
Garret Rieger 27a6c895ba [subset] Fix memory leak in plan creation. 2022-02-11 16:19:43 -08:00
Garret Rieger 670ef070bd [subset] Change subset plan create to be or_fail. 2022-02-11 16:01:33 -08:00
Garret Rieger ae8d373bcf [subset] add subset plan reference, set/get user data functions. 2022-02-11 14:54:23 -08:00
Garret Rieger b65e48a786 [subset] Add subset plan to public API.
Add the ability to create a subset plan which an be used to gather info on things like glyph mappings in the final subset. The plan can then be passed on to perform the subsetting operation.
2022-02-11 14:37:21 -08:00
Behdad Esfahbod 6893723879 [machinery] Allow using lazy-loader with void*
Fixes https://github.com/harfbuzz/harfbuzz/issues/3427
2022-02-11 14:28:51 -06:00
Chun-wei Fan 3d22dae5f8 meson.build: Only use get_variable() on icu_dep with pkgconfig
We might have found ICU via the .lib names, and get_variable() is only
available for pkg-config deps.
2022-02-10 14:42:08 -06:00
Chun-wei Fan 52c536bb8d hb-algs.hh: Fix build on Visual Studio 2015
The compiler does not recognize the std::hash call on
hb_decay<decltype (hb_deref (v)>, so give it a shove so that it is recognized
there.
2022-02-10 13:26:56 -06:00
Khaled Hosny e045dbf617 [ci] Upgrade pip on MSVC job
To avoid bad pre-installed version.
2022-02-10 04:06:20 +02:00
Alexis King 81754a5a96 [ot-math] Add hb_ot_math_get_glyph_kernings
closes #3396
2022-02-09 12:03:07 -06:00
Khaled Hosny 02a737e532 Revert "[ci] Downgrade pip on MSVC jobs"
This reverts commit c4cf5ddb27.
2022-02-08 21:45:42 +02:00
Alexis King 1bc4bad7a5 [ot-tag] Add HB_SCRIPT_MATH (Zmth) and map it to OT ‘math’ tag
The ISO 15924 code for mathematical notation is ‘Zmth’, but the
OpenType script is ‘math’.
2022-02-08 12:40:15 -06:00
Khaled Hosny 76cd8a4737 [arabic] Update to AMTRA revision 6
https://www.unicode.org/reports/tr53/tr53-6.html#Modifications
2022-02-08 12:38:51 -06:00
Khaled Hosny c4cf5ddb27 [ci] Downgrade pip on MSVC jobs
Turns out, pip 22.0 is the source of the breakage:
https://github.com/mesonbuild/meson/issues/9955#issuecomment-1030843844
https://github.com/pypa/pip/issues/10875
2022-02-06 18:13:16 +02:00
Khaled Hosny ac46c3248e 3.3.2 2022-02-06 15:41:18 +02:00
Behdad Esfahbod 7657bdd91f Revert "Fix failing Mac test for previous commit"
This reverts commit 070e30e6da.
2022-02-04 19:16:54 -06:00
Behdad Esfahbod 44da4e2a27 Revert "One more fix"
This reverts commit cab943a015.
2022-02-04 19:16:47 -06:00
Behdad Esfahbod 88798ee8be [GPOS] Disable split-kerning
Fixes https://github.com/harfbuzz/harfbuzz/issues/3408
Reverts https://github.com/harfbuzz/harfbuzz/pull/3235
Fixes https://github.com/harfbuzz/harfbuzz/pull/3235#issuecomment-1029814978
2022-02-04 19:01:57 -06:00
Behdad Esfahbod 9177953720 [atexit] Allow hb_atexit redefinition
Fixes https://github.com/harfbuzz/harfbuzz/issues/3412
2022-02-04 12:38:45 -06:00
Behdad Esfahbod 8670ffbf4c [buffer] Comment 2022-02-03 11:19:23 -06:00
Behdad Esfahbod e59da2871a [glyf] Don't store face in accelerator 2022-02-01 16:32:39 -06:00
Garret Rieger 0fb8e22406 [serialize] document how the serializer works. 2022-01-31 18:09:03 -06:00
Khaled Hosny 45df259538 3.3.1 2022-01-31 23:06:10 +02:00
Garret Rieger 4e2f409bce [subset] Don't hold references to members of the active_glyph_stack.
These references may get invalidated after the vector for the stack is resized. Fixes: https://oss-fuzz.com/testcase-detail/5422577634377728
2022-01-31 22:58:53 +02:00
Khaled Hosny 98da330638 3.3.0 2022-01-31 18:39:06 +02:00
Behdad Esfahbod 3e8742e376
Merge pull request #3402 from harfbuzz/language-tags
Make miscellaneous changes to hb-ot-tag-table.hh
2022-01-30 17:19:46 -08:00
David Corbett ae9afd9772 Let BCP 47 tag "mo" fall back to OT tag 'ROM ' 2022-01-30 14:32:59 -05:00
David Corbett a184c5f851 Don’t always inherit from macrolanguages
If an OpenType tag maps to a BCP 47 macrolanguage, that is presumably to
support the use of the macrolanguage as a vague stand-in for one of its
individual languages. For example, "ar" and "zh" are often used for
"arb" and "cmn". When the OpenType tag maps to a macrolanguage and some
but not all of its individual languages, that indicates that the
OpenType tag only corresponds to the listed individual languages (which
may be referred to using the macrolanguage subtag) but not the missing
individual languages. In particular, INUK (Nunavik Inuktitut) is mapped
to "ike" (Eastern Canadian Inuktitut) and "iu" (Inuktitut) but not to
"ikt" (Inuinnaqtun), so "ikt" should not inherit the INUK mapping from
its macrolanguage "iu".
2022-01-30 13:28:23 -05:00
Behdad Esfahbod b47b3b9972 [fallback-kern] Move buffer message to correct position 2022-01-29 10:24:38 -07:00
Behdad Esfahbod e986c12075 [verify] Show buffer input text when verification fails 2022-01-29 09:08:48 -07:00
Behdad Esfahbod d35f380126 [util] Change "All shapers failed." message to "Shaping failed."
Since we now emit this when verification fails as well.
2022-01-29 09:08:20 -07:00
Behdad Esfahbod 3972e0a8f1 [buffer] Whitespace 2022-01-29 08:22:19 -07:00
David Corbett 0b1bf89cc2 Replace “[family]” with “[collection]”
Not all language collections are language families.
2022-01-29 10:15:23 -05:00
David Corbett 0e31595e0d Infer tag mappings for unregistered macrolanguages
Every macrolanguage not mentioned in the OT language system tag registry
is mapped to every tag of its individual languages, if those have
registered tags.
2022-01-29 10:15:23 -05:00
David Corbett 5a6545940a Add the language system tag INUK 2022-01-28 21:29:43 -05:00
Garret Rieger bc899650c7 [subset] Fix for issue #3397.
cur_intersected_glyphs gets modified during recursion leading to incorrect filtering of sub tables in some cases. So don't use cur_intersected_glyphs. Instead just add an additional entry onto the parent_active_glyphs () stack.

Additionaly expands NotoNastaliqUrdu tests to include coverage of the issue from #3397.
2022-01-28 14:20:33 -08:00
Garret Rieger f3c1f4f0da [subset] convert active_glyphs_stack to be a vector of hb_set_t instead of hb_set_t*. 2022-01-28 14:20:33 -08:00
Behdad Esfahbod 476a6377a5 [buffer] Document HB_BUFFER_FLAG_VERIFY 2022-01-28 15:05:10 -07:00
Behdad Esfahbod 61856359cb [fuzz] Disable verification for now. 2022-01-28 14:07:29 -07:00
Behdad Esfahbod 6596e42d16 [fuzz] Verify shape results 2022-01-28 13:55:24 -07:00
Behdad Esfahbod 5b1d813b69 [config] Enable HB_NO_BUFFER_VERIFY in HB_LEAN 2022-01-28 13:53:44 -07:00
Behdad Esfahbod 61823838f9 [buffer] Add HB_BUFFER_FLAG_VERIFY
Move buffer verification code inside the library, from util/.

Part of https://github.com/harfbuzz/harfbuzz/issues/3010
2022-01-28 13:53:44 -07:00
Behdad Esfahbod 7c704d8982 [buffer] Make hb_buffer_append() take a const argument 2022-01-28 13:53:44 -07:00
Behdad Esfahbod 24650624c8 [font] Fix build with no-var configs 2022-01-28 13:53:21 -07:00
Behdad Esfahbod 91b779e23c Avoid redefinition of HB_NO_SETLOCALE in certain configs 2022-01-28 13:52:15 -07:00
Khaled Hosny fa29006bfa [doc] Fix generation of hb_glyph_flags_t docs
GTK-Doc does not like the empty lines here, and interprets everything
after the first empty line as the description of the enum itself not a
specific member and the generated text makes no sense.

Removing the empty lines makes the text harder to read (both in source
and HTML), but at least it is correctly organized.
2022-01-26 19:10:23 -08:00
Behdad Esfahbod e9cc5f6cdb
Merge pull request #3297 from harfbuzz/unsafe-to-concat
Implement UNSAFE_TO_CONCAT flag
2022-01-26 12:23:29 -08:00
Behdad Esfahbod 540af548dc [unsafe-to-concat] Clarify documentation as per feedback
Fixes https://github.com/harfbuzz/harfbuzz/pull/3297#discussion_r754395825
2022-01-25 09:10:56 -07:00