Behdad Esfahbod
db3aeeb547
[subset] Fix cast-align issue
...
Fixes https://github.com/harfbuzz/harfbuzz/issues/4316
1 year ago
Behdad Esfahbod
fc24ffbfb3
[priority-queue] Minor simplify
...
Tail-recursion definitely was being optimized by compiler,
but I prefer writing it this way.
1 year ago
Behdad Esfahbod
557653abce
[cff] Minor; add tableTag
1 year ago
Behdad Esfahbod
6df8ce7b41
[bit-page] Remove disabled assertion
1 year ago
Behdad Esfahbod
9b9a9c6f4a
[bit-page] Speed up is_empty()
1 year ago
Behdad Esfahbod
2f4ed5e939
[bit-page] Short-circuit is_subset()
1 year ago
Behdad Esfahbod
347c1f7013
[set] Keep (lazy) population per bit-page
...
36% speedup in:
BM_subset/subset_glyphs/merged.devalast.ttf/10
1 year ago
Behdad Esfahbod
04ee306b9a
[subset/hmtx] Another TODO
1 year ago
Behdad Esfahbod
caef35b3ef
[subset/hmtx] TODO
1 year ago
Behdad Esfahbod
982554c71c
[subset/cff1] Speed up plan_subset_charset more
...
16% speed up in retaingid subsetting NotoSansCJKkr-Regular.otf.
1 year ago
Khaled Hosny
15d8f0ca7c
[doc] Fix warning: Section has no title and no file
...
Move the private macros to the hb-common section, instead of a file-less
one.
1 year ago
Behdad Esfahbod
e60ec9dcd3
[subset/cff2] Speedup retaingids serialize
...
10% speedup in subset_glyphs/SourceHanSans-VF.otf/retaingids/10
1 year ago
Behdad Esfahbod
b583978bdf
[subset/DeltaSetMapIndex] Speedup retaingids serialize()
1 year ago
Behdad Esfahbod
3505486a0d
[subset/DeltaSetMapIndex] Micro-optimize
1 year ago
Behdad Esfahbod
c8f6ae1df7
[subset/hvar/retaingids] Synthesize a DeltaSetIndexMap if needed
...
Before we were emitting the entire VarStore rows if advMap
was Null. Instead, synthesize an advMap.
20% speed up in subset_glyphs/SourceHanSans-VF.otf/retaingids benchmark.
1 year ago
Behdad Esfahbod
1f4645c6fc
[subset/hvar] Speed up
1 year ago
Khaled Hosny
8d99db5c3b
[wasm] Typo [ci skip]
1 year ago
Behdad Esfahbod
8115033ac3
[hvar] Speed up retaingids planning
...
35% speedup in:
BM_subset/subset_glyphs/SourceHanSans-VF.otf/retaingids/10
1 year ago
Behdad Esfahbod
1fe1a49790
[bimap] Minor vector pre-alloc
1 year ago
Behdad Esfahbod
7a7001dfcf
Typo
1 year ago
Khaled Hosny
b4305532a7
8.0.0
1 year ago
Khaled Hosny
5b8ba51251
[doc] Fix warning about missing hb-gobject section
...
We don’t build hb-gobject docs anymore, lets not bother gtk-doc with it.
1 year ago
Khaled Hosny
0ed946c379
[wasm] Fix warning
...
In file included from src/harfbuzz.cc:62:
src/hb-wasm-shape.cc:27:9: warning: 'HB_DEBUG_WASM' macro redefined [-Wmacro-redefined]
^
src/hb-debug.hh:393:9: note: previous definition is here
^
1 warning generated.
1 year ago
Khaled Hosny
afe0910295
[font] Actually deprecate get_glyph_shape()
...
We previously marked it as deprecated in the documentation but didn’t
actually deprecate it in code. Now the only known users have migrated to
draw_glyph(), lets deprecate o=it for good.
1 year ago
Behdad Esfahbod
c6a01441d9
[sanitize] Always-inline OffsetTo::sanitize()
...
8% speed up in sanitizing Gulzar.
1 year ago
Behdad Esfahbod
f60dbd906a
Fix thinko
...
Fixes https://oss-fuzz.com/testcase-detail/4787105656864768
1 year ago
Behdad Esfahbod
d84c5f29b7
Revert "[gsubgpos] Make (Chain)Context funcs templatized"
...
This reverts commit 03ac08d267eb7ad2f1ac039115fc0030ea475fae.
1 year ago
Behdad Esfahbod
e73223b9e5
[gsubgpos] Make (Chain)Context funcs templatized
...
Unfortunately they still won't be inlined because the matcher_t
needs function pointers. So, no speed up. Going to revert.
1 year ago
Behdad Esfahbod
3e3820badd
Simplify a few bool returns
...
Unnecessary. Cast to bool does the job.
1 year ago
Behdad Esfahbod
1c8d54deef
[shape] Another always-inline
1 year ago
Behdad Esfahbod
df45067fbb
[shape] More always-inline
1 year ago
Behdad Esfahbod
c3a28be509
[shape] One more always-inline
1 year ago
Behdad Esfahbod
7288123481
Allow overriding HB_ALWAYS_INLINE
1 year ago
Behdad Esfahbod
9fea19ebff
[shape] Another always-inline
1 year ago
Behdad Esfahbod
5b45c56c9e
[shape] Always-inline more
...
Another 10% speedup in BM_Shape Gulzar-Regular.
1 year ago
Behdad Esfahbod
bb9692e8a3
[shape] Always-inline match_input
...
10% speedup in BM_Shape Gulzar-Regular.
1 year ago
Behdad Esfahbod
6c451054fb
[sanitize] Sprinkle a few unlikely's
1 year ago
Behdad Esfahbod
4ecc62c28d
[sanitize] Simplify ChainRule::sanitize again
...
check_struct does the same now.
1 year ago
Behdad Esfahbod
253ec08b3d
[sanitize] Speed up check_struct on x64
1 year ago
Behdad Esfahbod
fb02f4a897
[sanitize] Micro-optimize ChainRule::sanitize
1 year ago
Khaled Hosny
c650858c63
[graphite] Fix cluster advance width
...
Based on https://github.com/harfbuzz/harfbuzz/issues/4309#issuecomment-1624730406
Fixes https://github.com/harfbuzz/harfbuzz/issues/4309
1 year ago
Behdad Esfahbod
73ee9c346f
Revert "[sanitize] Only check trailing edge of ranges"
...
This reverts commit e4856cf809
.
This is wrong on at least 32-bit systems. Might add a condition
version later.
1 year ago
Behdad Esfahbod
0b879afb5a
[sanitize] Inline Coverage::sanitize
1 year ago
Behdad Esfahbod
92448910a7
[sanitize] Minor inline a few more short functions
1 year ago
Behdad Esfahbod
d84504206c
[sanitize] Optimize away an overflow check when not needed
...
When the length argument is 16bit...
1 year ago
Behdad Esfahbod
e4856cf809
[sanitize] Only check trailing edge of ranges
...
Assumptions...
Speeds up Gulzar load_face_and_shape benchmark by 7%.
1 year ago
Behdad Esfahbod
6a683ea6eb
[sanitize] Enlighten check_range() some more
...
No need to check for len=0 arrays. They must still be in range.
1 year ago
Behdad Esfahbod
c24ea1036b
[sanitize] Minor tweak to len=0 case which is handled otherwise anyway
1 year ago
Behdad Esfahbod
403bc7caa4
[sanitize] Simplify Rule::sanitize()
1 year ago
Behdad Esfahbod
385896ff66
[sanitize] Speed up ChainRule::sanitize()
...
10% speedup in sanitizing Gulzar-Regular.
1 year ago