90d0200ac7
[simd] Use set1 operator
2019-12-11 10:52:14 -06:00
997809ea4a
[simd] Blend 9ary search with bsearch
2019-12-11 10:52:14 -06:00
bc25a5694c
[simd] Enable ksearch for (non-range) GlyphID search
2019-12-11 10:52:14 -06:00
38c84b2246
[simd] Reduce one vector op
2019-12-11 10:52:14 -06:00
b7d01281d2
[simd] Add disabled code for preferring bsearch for shorter lists
2019-12-11 10:52:14 -06:00
41a7ec0697
[simd] Fix build
2019-12-11 10:52:14 -06:00
6013ead1f5
[simd] Use faster byteswap, using shuffle
2019-12-11 10:52:14 -06:00
343da74ec6
[simd] Add HB_SIMD_VERIFY
2019-12-11 10:52:14 -06:00
6bba9d876a
[simd] Implement 9ary search for RangeRecord
...
Is good on correctness. Have not measured performance yet.
Part of https://github.com/harfbuzz/harfbuzz/issues/566
2019-12-11 10:52:14 -06:00
291d30b1ff
[simd] Comments
2019-12-11 10:52:14 -06:00
c799742ac1
[simd] Enable GCC avx2 target if HB_SIMD_AVX2 is defined
2019-12-11 10:52:14 -06:00
0033641189
[simd] Add general design and structure
...
Part of https://github.com/harfbuzz/harfbuzz/issues/566
2019-12-11 10:52:14 -06:00
4ef597e170
Fix unary minus operator applied to unsigned int
...
Applying unary minus operator to unsigned int causes the following error on MSVS: error C4146
This patch fixes the error.
2019-12-11 09:19:28 -06:00
1770493cf5
Fix unary minus operator applied to unsigned int
...
Applying unary minus operator to unsigned int causes the following error on MSVS: error C4146
This patch fixes the error.
2019-12-11 09:19:19 -06:00
a061e47fcc
Change a few HB_INTERNAL static methods to static inline
2019-12-10 13:31:50 -06:00
dd3972a364
[GSUB] Simplify Extension is_reverse()
...
We don't allow extension lookups to chain to another extension lookup.
Simplify code for that.
2019-12-10 13:28:39 -06:00
858b627984
[machinery] Remove CastR<>()
2019-12-10 13:18:32 -06:00
b84ceb2fcf
[machinery] Remove CastP
2019-12-10 13:02:48 -06:00
85574ec287
[machinery] Minor
2019-12-10 12:52:32 -06:00
e101a67257
[perf] Add texts/fa-thelittleprince.txt
2019-12-10 12:46:14 -06:00
2c781a6701
[perf] More rename
2019-12-10 12:43:42 -06:00
d703392afe
[perf] Adjust text/fa-monologue.txt
2019-12-10 12:42:29 -06:00
19d1b9d4f3
[perf] Renames
2019-12-10 12:39:01 -06:00
6a60ca117c
[algs] Fold last other bsearch() in
...
Now truly have only one bsearch implementation.
2019-12-10 12:32:59 -06:00
53dc8d944f
Add initial shaping performance test and profiler
2019-12-10 21:53:30 +03:30
39afe608b4
[algs] Fold one more custom bsearch() in
...
One more to go.
2019-12-10 12:13:40 -06:00
b1dc676eaa
[algs] Reduce one more bsearch() impl
...
Ouch, there were three more left. Down one. Two to go.
2019-12-10 12:13:40 -06:00
6f76c325e5
[test] Update 10.15 results
...
Turned out only SFNS, which wasn't available in 10.14 anyway, needed an update
See https://crbug.com/1005969#c37 also
2019-12-10 21:43:11 +03:30
2241a676ba
[test] Add macOS 10.15 related fonts
...
breaks the test and 10.15 bot, will add the fix in next commit, also adds a broken test for f47cbade1
2019-12-10 19:50:34 +03:30
f3de3b6d3d
[ci] fix macOS 10.15 bot
...
Xcode 11.3.0 image is supposed to work per https://circleci.com/docs/2.0/testing-ios/
but isn't https://circleci.com/gh/harfbuzz/harfbuzz/118693 AFAICS
Let's try Xcode 11.2.1 image
2019-12-10 16:37:48 +03:30
49434bdd49
[ci] Add a macOS 10.15 bot
...
Should update macos.tests with the fonts, for now
2019-12-10 16:28:04 +03:30
f47cbade18
[aat] Adjust fallback positioning logic
...
Fixes https://github.com/harfbuzz/harfbuzz/issues/1528
Wish could add a test...
$ ./hb-view --font-file Thonburi.ttc --unicodes U+0E17,U+0E35,U+0E48,U+0E4A --shaper ot
▃
▂▃▃▄▃▂▁▊ ▃ ▃ ▎
▗ ▅▆▆▅ ▌ ▆▆▅▆▙ ▌▗
▗ ▙ ▂ ▎▗ ▟
▅▆▆▆▇▇▇▇▇▇▇▆▆▅▄ ▇▅▅▅ ▙▆▇
▁▁▂▁ ▁▁
▖ ▖ ▂▂
▉ ▁ ▆▆ ▌ ▆ ▙▄ ▁▁
▙ ▉ ▃▖ ▟
▊ ▉ ▅▟ ▗▄
▊ ▉ ▗ ▏ ▇▆
▊ ▟ ▉ ▆ ▖
▊ ▉ ▌ ▂
▊ ▏ ▉ ▌ ▗ ▄▟
▊ ▆
2019-12-09 18:11:04 -06:00
fa7edf87c9
[bsearch] Massage API some more
2019-12-09 17:51:41 -06:00
70aa5071d8
[algs] Adjust return value of hb_ctz(0) to be 32 instead of 0
2019-12-09 17:48:10 -06:00
eefb78f674
Minor
2019-12-09 17:46:56 -06:00
9fb030585a
Rename start/end to first/last in RangeRecord
...
Because that's what they are.
2019-12-09 17:46:08 -06:00
8ac4ba14dc
Fix typo in TESTING.md ( #2066 )
2019-12-09 20:28:28 +03:30
9b1d5c4a59
[number] fix where strtod_l not available
2019-12-09 12:29:28 +03:30
5c8f960289
[number] minor, include the renamed header
2019-12-09 10:48:43 +03:30
48eef2724c
[algs/array] Consolidate the last two bsearch implementations!
...
Yay! Seems to work.
2019-12-06 05:04:11 +00:00
ed35dea8c0
Fourth try... sighs
2019-12-06 04:37:11 +00:00
9168310bb4
Fix build, third times...
2019-12-06 04:28:06 +00:00
34f5cc2cc8
Second try at fixing build
2019-12-06 04:09:33 +00:00
14ce5ab0b7
First try at fixing build errors
2019-12-06 03:54:46 +00:00
2274270c6a
[algs] Streamline bsearch some more
2019-12-06 03:42:21 +00:00
bd55d4b49f
[algs] Streamline bsearch() API more towards hb_array_t::bsearch_impl()
...
Preparing to merge the two finally!
2019-12-06 03:35:24 +00:00
fd6df520a1
[array] Isolate bsearch implementation more
2019-12-06 03:01:34 +00:00
06d3c2019f
[array] Simplify bfind() positioning
...
I had copied the old scheme from fontconfig's fccharset.c. I just
convinced myself that this change is correct and produces exact
same results. But I also am skeptical. Anyone else feel like
convincing themselves as well please?
2019-12-06 02:52:54 +00:00
d67ba649a3
Rename hb_array_t::in_range to hb_array_t::check_range
2019-12-06 02:27:23 +00:00
72d83a0280
Make hb_array_t::in_range similar to hb_sanitize_context_t::check_range
2019-12-06 02:27:23 +00:00
Behdad Esfahbod
Daeren
Ebrahim Byagowi
Evgeniy Reizner