organicmaps/harfbuzz
13
Fork 0
mirror of https://github.com/harfbuzz/harfbuzz.git synced 2025-04-04 21:15:06 +00:00
Code Issues 1 Projects Releases Wiki Activity Actions 9
18616 commits 23 branches 188 tags 126 MiB
Commit graph

348 commits

Author SHA1 Message Date
Behdad Esfahbod
4d2a362f84 [test/subset] Remove tempdir 
Was creating a tempdir in /tmp for every single test item,
and not removing it.  Create one per test file, and remove
if test successful.
 2025-03-25 15:36:39 -06:00
Behdad Esfahbod
e0aee5815a [test] Respect MESON_EXE_WRAPPER in more test runners 
It's not working for me though. Meson doesn't seem to set WINEPATH
during testing.
 2025-03-22 14:56:37 -06:00
Garret Rieger
ea1434b897 Split iup glyf partial instance tests into a separate set which ignores due to IUP rounding differences fonttools. 2025-02-25 19:02:11 +00:00
Behdad Esfahbod
9a0cf8d641 [test/subset/cff-japanese] Shrink test 
Was taking too long.
 2025-02-25 00:19:25 -07:00
Behdad Esfahbod
084fb3ab4f [test/subset/cmap] Shrink test 
Was taking more than the rest of the test suite combined.
 2025-02-25 00:16:27 -07:00
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
Garret Rieger
b0d52abe97 [subset] consider variation selectors subsetting cmap14 
cmap14 subsetting code was not considering variation selectors in the input unicode set when deciding which variant glyphs to keep. This updates subsetting to only keeps variant glyphs if their variation selector code point is in the input unicodes set.

For https://github.com/harfbuzz/harfbuzz/issues/4911
 2024-11-11 13:45:08 -07:00
Qunxin Liu
622e9c33c3 [instancer] fix for incorrect offset calculation when we have gvar padding 2024-11-06 12:38:00 -07:00
Qunxin Liu
6fd76e1f67 [subset] offset format fix in gvar table 2024-09-17 15:23:43 -06:00
Alexander Grund
70334d74dc Run subset tests with C-locale 
One problematic input is `--instance=wdth=112.5` which is parsed using
the users local (`strtof`).
So while in C or US locales it yields 112.5 in some European locales it
yields 112 as the dot is not valid in this position.

A European user might pass "112,5" instead but for the test the C locale
is assumed, so enforce it.

Fixes #4854
 2024-09-11 12:07:04 -06:00
Qunxin Liu
634778efc0 [subset] bug fix in post table 
insert retain-gid holes when retain-gids is enabled
 2024-08-21 12:06:02 -06:00
Qunxin Liu
72502ef02b [instancer] dont return false when variation data is empty after partial instancing 2024-08-12 15:51:13 -06:00
Khaled Hosny
a38f853e83 Drop unused Makefile.sources files 2024-06-27 16:33:16 +03:00
Khaled Hosny
cf1fdf1632 Drop autotools build 
Fixes https://github.com/harfbuzz/harfbuzz/issues/4720
 2024-06-27 15:28:38 +03:00
David Corbett
0db136b1b0 [subset-cff] Shorten output of encode_num_tp 2024-05-12 09:04:02 -07:00
Qunxin Liu
7975c9cabd [instancer] bug fix for COLRv1, add missing mapping to VarIdx::NO_VARIATION 2024-05-09 08:34:14 -07:00
Garret Rieger
b74a7ecc93 Fix IUP differences accoss platforms. 
- Slightly increase the tolerance to account for differences in floating point math across platforms.
- The specific issue is a fused multiply add (d = d1 + (x - x1) * scale; in hb-subset-instancer-iup.cc)
- Also fix more implicit double conversion warnings.
 2024-05-08 10:23:31 -06:00
Qunxin Liu
a5c9c82692 rename expected files from *.retain-all-codepoint.* to *.all.* 
This is to fix file name too long error
tar: harfbuzz-8.4.0/test/subset/data/expected/sync_vmetrics/playwrite_mm.subset.default.retain-all-codepoint.SPED=25,YEXT=520,wght=100-400,slnt=18.iup_optimize.ttf: file name is too long (cannot be split); not dumped
tar: Exiting with failure status due to previous errors
 2024-05-06 13:22:49 -07:00
Qunxin Liu
3b3c154bd1 add back missing tests that were disabled when wrapped by experimental 2024-05-06 12:56:23 -07:00
Qunxin Liu
6c8985724e fix bots 2024-05-06 10:42:03 -07:00
Qunxin Liu
f614a7d309 [instancer] add/update tests for COLRv1 instancing 2024-05-06 10:41:07 -07:00
Qunxin Liu
ebcdadbf54 [instancer] add tests for instantiating BASE table 2024-05-03 17:22:16 -05:00
Garret Rieger
f145ab2f1e Disable empty_region_vardata test when running on 32bit machines. 2024-05-03 18:31:18 +00:00
Garret Rieger
4cf606f674 [instancer] Promote partial instancing api out of experimental. 2024-05-02 21:22:18 +00:00
Garret Rieger
2cfa55056c [instancer] Switch to doubles for delta calculations. 
FontTools uses double's internally so match FontTool's precision so we get the same results more often for floating point calculations.
 2024-04-15 16:06:19 -06:00
Garret Rieger
6e87481731 [subset] add virtual links to keep the gdef varstore last. 2024-03-28 23:00:54 +00:00
Garret Rieger
7145607149 [subset] place the gdef var store at the end of the GDEF table. 
It appears that some font processors assume it is the last sub table.
 2024-03-28 23:00:54 +00:00
Garret Rieger
990fb14d78 [subset] Disable fonttools comparison in a few tests where rounding diffs show up. 
Context: https://github.com/harfbuzz/harfbuzz/issues/4642
 2024-03-28 17:00:12 -06:00
Garret Rieger
11f5dec75f [subset] Move post apply mvar test intance position to avoid rounding diff. 2024-03-28 17:00:12 -06:00
Garret Rieger
975bd39b15 [subset] Match hb rounding to fonttools for avg char width. 
Update Muli and Roboto input test fonts to recalc class values. FontTools recalcs class values during instancing while harfbuzz does not (like the ft subsetter).
 2024-03-28 17:00:12 -06:00
Garret Rieger
15a05b2432 [subset] Update generate-expected-outputs to drop BASE. 
This matches the behaviour of the test runner.
 2024-03-28 17:00:12 -06:00
Garret Rieger
ec46c4c2e6 [subset] Update subset test suite to match some changes made to fonttools. 
FontTools has added pruning functionality for code pages which we don't yet have in harfbuzz. Update our test runner to disable that for now. Also match fonttools and always recalc usFirst/LastCharIndex even if no prune unicode ranges is set.
 2024-03-28 17:00:12 -06:00
Qunxin Liu
cfbb6a6872
[subset] support BASE table subsetting (#4618) 
* [subset] support BASE table

* [subset] add tests for BASE table subsetting
 2024-03-14 18:33:34 -06:00
Qunxin Liu
84771374e8 [instancer] add tests for iup delta optimization 2024-01-26 15:51:49 -08:00
Qunxin Liu
df735213a8
[instancer-solver] port solver fix (#4519) 
* [instancer-solver] port solver fix

from https://github.com/fonttools/fonttools/pull/3374
 2023-12-06 10:50:29 -07:00
Qunxin Liu
f1ac36ba7f [instancer] update hhea vertical metrics with OS/2 ones 2023-11-15 13:25:43 -07:00
Qunxin Liu
2d001a0adc [subset] OTS doesn't allow ligature attachment offset to be 0 2023-11-08 14:10:06 -07:00
Garret Rieger
9701b0024e [subset] Always place CFF1/2 char strings at the end of the packed table. 
Makes the behaviour from HB_SUBSET_FLAGS_IFTB_REQUIREMENTS the default.
 2023-11-06 16:49:09 -07:00
Garret Rieger
c6884377ec [subset] Add HB_SUBSET_FLAGS_IFTB_REQUIREMENTS flag. 
When enabled the output subset conforms to the requirements for a base font to be patched by IFTB patches. IFTB is a proposed incremental font transfer patch. This new flag is set as experimental. Currently, setting this flag causes the output subset to also use long offsets for outline data in loca/glyf, gvar, CFF, and CFF2.

This change is a version of 3ae2fe2084 rebased onto head w/ tests added.
 2023-11-06 16:49:09 -07:00
Qunxin Liu
d550e16acc [instancer] Recalc avgcharwidth after instancing 2023-11-06 13:12:04 -07:00
Qunxin Liu
6baf791e6f [instancer] add tests 2023-11-03 11:06:07 -07:00
Qunxin Liu
31b5b2b1d7 [instancer] add tests for valueformat 2023-11-03 10:27:33 -07:00
Qunxin Liu
413af0a9b2 [instancer] When partial instancing, subset LigCaretList after VarStore instantiation 2023-11-03 10:24:56 -06:00
Qunxin Liu
f6f35f73bc [instancer] add tests for feature variations partial instancing 2023-10-30 12:31:49 -06:00
Qunxin Liu
e595fe75cc fix bots 
partial instancing tests are wrapped by experimenal_api, do not enable
them in Makefile
 2023-10-19 09:38:38 -07:00
Qunxin Liu
ce46698929 [subset] update tests 
to reflect changes in GDEF.MarkGlyphSetsDef and Lookup.MarkFilteringSet
 2023-10-18 15:49:57 -07:00
Qunxin Liu
bde6fb83ea add missing file for test empty_region_vardata 2023-10-18 10:21:02 -07:00
Qunxin Liu
57dac23d79 [instancer] add tests 
Also update tests to gei rid of no-tables-with-item-variations profile
 2023-10-18 08:07:26 -06:00
Garret Rieger
bbd53fcfa4 Remove glyph_map test that uses retain gids. 
We no longer allow the use of glyph map + retain gids (see: https://github.com/harfbuzz/harfbuzz/blob/main/src/hb-subset-plan.cc#L817). Also add the test to the meson file so it will be run by default.
 2023-10-10 17:59:05 -06:00
Qunxin Liu
2415d5f239 [instancer] update tests to include GDEF/GPOS tables 2023-09-27 13:33:09 -07:00