|
fee0f41c
|
2018-08-28T18:27:41
|
|
Don't declare extern symbols as inline
clang -O3 was completely removing _get_gdef(), causing link
failure when needed from another compilation unit. Surprisingly,
"extern inline" didn't fix it.
|
|
967741e4
|
2018-08-28T18:18:02
|
|
Add explicit to hb_auto_t 1param constructors
|
|
c0a3f38f
|
2018-08-28T17:22:38
|
|
Fix distcheck
|
|
2dde6c80
|
2018-08-28T11:02:00
|
|
Explicitly pass reference in hb_auto_t constructor
Fixes clang bots as well as fuzzer issue.
|
|
a2680680
|
2018-08-26T15:18:46
|
|
[ot-face] Remove tables_t. The face_data_t is what tables_t was
|
|
3a0b3a29
|
2018-08-26T15:11:24
|
|
Free up hb-ot-face.hh from includes
There might be a better way to do this, but I couldn't find...
|
|
a84309af
|
2018-08-26T09:33:01
|
|
[face] Use saved cmap in collect_unicodes() APIs
|
|
c624e18a
|
2018-08-26T09:19:20
|
|
[OT] Move code around
|
|
39990946
|
2018-08-26T09:12:25
|
|
[OT] Use relaxed ops for recursing into GSUB/GPOS lookups again
|
|
d8c57e85
|
2018-08-26T09:03:31
|
|
[GDEF] Move code around
|
|
453e0c60
|
2018-08-26T01:30:52
|
|
Add template keyword
I swear I have no idea why this is needed, but it is...
|
|
c44b0caf
|
2018-08-26T01:29:30
|
|
Fixup
|
|
4096fbe4
|
2018-08-26T01:25:54
|
|
[ot-face] Port GDEF to unified table accelerator model as well
|
|
b929100f
|
2018-08-26T01:15:47
|
|
[ot-face] Unify GSUB/GPOS accelerators in the table accelerator framework
|
|
963413fc
|
2018-08-26T00:47:55
|
|
WIP
|
|
e6cb9380
|
2018-08-26T00:21:29
|
|
[ot-face] Unify accelerators and tables
|
|
7f8b6fe6
|
2018-08-26T00:07:12
|
|
Minor
|
|
59bca0d2
|
2018-08-26T00:06:12
|
|
[ot-font] Rename
|
|
1c48a7ef
|
2018-08-25T23:54:06
|
|
[ot-font] Cache tables on face, not font
|
|
71f8893e
|
2018-08-25T23:27:45
|
|
[ot-font] Use init0
|
|
f36ae6f8
|
2018-08-25T23:19:34
|
|
[ot-font] Make everything (cmap/metrics) lazy
|
|
62fc5472
|
2018-08-25T23:15:52
|
|
[ot-font] Fallback vertical advance to UPEM
Used to be ascent+descent. Whatever. This is more convenient for
upcoming changes.
|
|
d950022c
|
2018-08-25T23:06:37
|
|
Add hb-ot-face.cc/hh
|
|
dae39c58
|
2018-08-25T22:44:39
|
|
[face] Introduce character-map API
New API:
+hb_face_collect_unicodes
+hb_face_collect_variation_selectors
+hb_face_collect_variation_unicodes
|
|
66384db9
|
2018-08-25T22:39:17
|
|
Minor
|
|
c77ae408
|
2018-08-25T22:36:36
|
|
Rename hb-*private.hh to hb-*.hh
Sorry for the noise, downstream custom builders. Please adjust.
|
|
82b12bca
|
2018-08-25T22:07:17
|
|
[unicode] Add HB_UNICODE_MAX
New API:
HB_UNICODE_MAX
|
|
b09a0336
|
2018-08-25T22:04:39
|
|
[cmap] More hb_auto_t<>(...)
|
|
28634db0
|
2018-08-25T21:23:43
|
|
[cmap] Fix skipping gid0 in Format4 collect_unicodes()
|
|
acce1fa3
|
2018-08-25T21:15:39
|
|
Use new hb_auto_t<> constructor with Coverage::Iter
|
|
ddea4d19
|
2018-08-25T21:08:15
|
|
Add 1-param passthrough constructor to hb_auto_t<>
And use.
|
|
c8cfb702
|
2018-08-25T16:14:32
|
|
[cmap] Minor
|
|
1dcf5fb0
|
2018-08-25T16:11:26
|
|
[cmap Add hb_subset_collect_variation_unicodes()
To be moved to hb-face.h later.
|
|
4806b380
|
2018-08-25T15:56:07
|
|
[cmap] Add hb_subset_collect_variation_selectors()
To be moved to hb-face later.
|
|
3336de24
|
2018-08-25T15:41:25
|
|
[cmap] Remove unused code
|
|
1becabe0
|
2018-08-25T15:37:56
|
|
[cmap] Use bsearch to find subtables
|
|
02fe03e0
|
2018-08-25T15:33:05
|
|
[cmap] Factor out find_best_subtable
|
|
b41c43b4
|
2018-08-25T15:25:03
|
|
Minor
|
|
7d382fa2
|
2018-08-25T09:35:45
|
|
[cmap] Implement collect_unicodes() for Format0/6/10
|
|
bd0e5425
|
2018-08-25T09:33:30
|
|
[cmap] Simplify collect_unicodes()
Don't use accelerator (almost). Hooks up Format13 as well.
|
|
d60c4656
|
2018-08-25T08:47:06
|
|
Rename get_all_codepoints() to collect_unicodes()
|
|
531051b8
|
2018-08-25T08:44:18
|
|
[ot-font] Return char-not-found if mapped to gid 0
OpenType cmap table can use gid=0 to mean "not covered" to produce
more optimized cmap subtables. Return false from get_nominal_glyph()
for those. hb-ft already does this.
|
|
aadb2a91
|
2018-08-25T08:18:53
|
|
Add face-builder
New API:
+hb_face_builder_create
+hb_face_builder_add_table
|
|
6cac9dc9
|
2018-08-24T10:29:45
|
|
[blob] Add as_bytes()
|
|
29a9a088
|
2018-08-24T10:24:45
|
|
[sanitize] Document how sanitize machinery works
|
|
142ac5a6
|
2018-08-24T10:07:49
|
|
[serialize] Add copy_bytes() and copy_blob()
|
|
1c6b3693
|
2018-08-15T20:12:25
|
|
[ot] Rename hb_ot_layout_t to hb_ot_face_data_t
|
|
22defe09
|
2018-08-14T14:47:20
|
|
Add guard around free_static_C_locale
Add HB_USE_ATEXIT guard around free_static_C_locale missed by previous
commit.
|
|
7112c641
|
2018-08-14T13:50:24
|
|
Guard free_static function definitions with macro
Add HB_USE_ATEXIT macros around free_static function definitions to
avoid unused function compiler warnings/errors.
|
|
63be5dcd
|
2018-08-14T10:59:34
|
|
1.8.8
|
|
c9a22fa1
|
2018-08-13T17:02:21
|
|
[icu] Pass normalizer instance down as user_data
|
|
f0ef096b
|
2018-08-13T10:30:41
|
|
Don't add reference in get_unicode_funcs() functions
Users don't expect a new reference returned from a get() function.
Indeed, all users of that API I foud where NOT destroying the reference.
Just change the implementations to NOT return a reference.
This applies to the following APIs:
hb_unicode_funcs_get_default()
hb_glib_get_unicode_funcs()
hb_icu_get_unicode_funcs()
Fixes https://github.com/harfbuzz/harfbuzz/issues/1134
|
|
9bbd1cdf
|
2018-08-12T18:20:53
|
|
[uniscribe] Use lazy loader
Fixes https://github.com/harfbuzz/harfbuzz/issues/1129
|
|
26636899
|
2018-08-12T18:10:59
|
|
Another clang bot fix
|
|
989e71a9
|
2018-08-12T17:47:59
|
|
Silence clang
|
|
6750ec69
|
2018-08-12T17:42:16
|
|
[lazy] Use for C_locale
|
|
747d2564
|
2018-08-12T17:32:10
|
|
[lazy] Port more
|
|
7bd508a0
|
2018-08-12T17:19:55
|
|
[lazy] Rename free()
|
|
a3b97db9
|
2018-08-12T17:14:32
|
|
[lazy] Port more to it
|
|
69010909
|
2018-08-12T16:57:06
|
|
[lazy] Make hb_lazy_loader_t<> more usable
|
|
bb905e9a
|
2018-08-12T16:40:08
|
|
[lazy] Minor
|
|
3945cd78
|
2018-08-12T16:33:48
|
|
Minor
|
|
c21a1b95
|
2018-08-12T16:30:39
|
|
[icu] Use get_relaxed for normalizer access
We've had the proper barriers already. No need for more with every access.
|
|
125fefa2
|
2018-08-12T16:29:41
|
|
[icu] Fix for breakage after recent change
|
|
53442be1
|
2018-08-12T16:20:11
|
|
[lazy] Use for ft_library
|
|
7a8d4803
|
2018-08-12T16:00:13
|
|
[lazy] Add .free()
|
|
c7ca30a5
|
2018-08-12T13:46:53
|
|
[ot/ft] Port font_funcs statis to lazy-loader
|
|
cb3fc368
|
2018-08-12T13:39:01
|
|
[ucdn/glib/icu] Port unicode_funcs statics to lazy-loader
|
|
1b6b4812
|
2018-08-12T13:16:40
|
|
[lazy] Allow calling fini() multiple times
|
|
918ad9f5
|
2018-08-12T13:12:29
|
|
[lazy] More
|
|
5abdf5ee
|
2018-08-12T12:40:24
|
|
[lazy] More shuffle
|
|
5d9863be
|
2018-08-12T12:27:47
|
|
Minor
|
|
f9a3eab8
|
2018-08-12T12:21:56
|
|
Add explicit_operator
Fixes https://github.com/harfbuzz/harfbuzz/issues/1127
|
|
470acb6c
|
2018-08-12T12:09:20
|
|
Rename
|
|
39607dc9
|
2018-08-09T15:16:32
|
|
Support _ in feature ids
|
|
652cd45c
|
2018-08-09T01:02:02
|
|
[coretext] Another try
|
|
3f3585ca
|
2018-08-09T00:58:17
|
|
Fix coretext build (hopefully)
|
|
aa3b6017
|
2018-08-09T00:56:28
|
|
Revert "[iter] Make operator bool explicit"
This reverts commit 66920a6bace7c54c8166c4ed938b6ffc5fabcf2b.
Some of our bots (Oracle Studio and Apple gcc 4.2) do not allow
explicit except for constructors.
https://github.com/harfbuzz/harfbuzz/issues/1127
|
|
e1a23542
|
2018-08-09T00:53:25
|
|
[atomic] More
|
|
f1f4b45a
|
2018-08-09T00:28:40
|
|
Fix lazy_loader fini()
Was creating object even if wasn't there.
|
|
1f738094
|
2018-08-09T00:22:37
|
|
[atomic] Add hb_atomic_ptr_t<> and port all uses
Found and fixed a couple bugs.
Found a couple multithreading issues. Marked them with "XXX-MT-bug".
|
|
6e42f4c5
|
2018-08-08T22:51:35
|
|
[atomic] Remove hb_atomic_int_impl_t now that it's always int
|
|
1227a9e5
|
2018-08-08T22:45:49
|
|
[atomic] Unify typedef
Removes volatile from fallback implementation. That was handwavy
anyway.
|
|
5cd1888c
|
2018-08-08T22:43:53
|
|
[atomic] Change Mac atomic int type to int
If ever int is not 32bit, we'll fix this...
|
|
e1de86ea
|
2018-08-08T22:42:33
|
|
[atomic] Change Windows atomic int type to int
|
|
aee2d10b
|
2018-08-08T22:40:10
|
|
[atomic] Change Solaris atomic int to signed
|
|
b6fdcf4f
|
2018-08-08T21:54:08
|
|
1.8.7
|
|
51ffc3e6
|
2018-08-08T22:01:04
|
|
Fix previous commit to use atomic operations
|
|
88395768
|
2018-08-08T21:42:18
|
|
Fix assertion failure
https://bugzilla.mozilla.org/show_bug.cgi?id=1476334
|
|
e49a38b2
|
2018-08-07T09:55:42
|
|
1.8.6
|
|
b0ff79ac
|
2018-08-07T09:52:06
|
|
Add +hb_font_get_glyph_advances_for_direction
New API:
+hb_font_get_glyph_advances_for_direction
|
|
28e2f467
|
2018-08-07T09:47:00
|
|
Actually add hb_font_get_glyph_[hv]_advances
New API:
+hb_font_get_glyph_h_advances
+hb_font_get_glyph_v_advances
|
|
99677519
|
2018-08-07T09:44:01
|
|
Minor
|
|
3d7c555a
|
2018-08-07T09:37:52
|
|
Remove get_advances_nil and alias it to get_advances_default
|
|
66920a6b
|
2018-08-06T15:32:13
|
|
[iter] Make operator bool explicit
We would need to write:
for (Iter<T> it (...); bool (t); t++)
instead of:
for (Iter<T> it (...); t; t++)
But I think it's an improvement in code readability in the long term.
Fixes https://github.com/harfbuzz/harfbuzz/issues/1127
|
|
79e21984
|
2018-08-06T09:45:17
|
|
Add batch advance width callback function
New API:
+hb_font_funcs_set_glyph_h_advances_func
+hb_font_funcs_set_glyph_v_advances_func
+hb_font_get_glyph_h_advances
+hb_font_get_glyph_h_advances_func_t
+hb_font_get_glyph_v_advances
+hb_font_get_glyph_v_advances_func_t
|
|
9533364c
|
2018-08-01T13:01:08
|
|
batchwidth
|
|
b912fbea
|
2018-08-06T06:30:12
|
|
Remove most uses of direct comparison to Null objects
|
|
3506672c
|
2018-08-06T06:17:48
|
|
Port _nil objects to Null() machinery
Finally, unified!
|
|
1abd427a
|
2018-08-06T05:53:35
|
|
[BASE] Rename horzi/vert to h/v
|