Behdad Esfahbod
|
0935b32795
|
[vector] Speed up shrink_vector for trivial destructors
|
1 year ago |
|
Behdad Esfahbod
|
da2e2c8c25
|
[subset/cff] Speed up offset writing
|
1 year ago |
|
Behdad Esfahbod
|
c85ca75eff
|
[buffer-verify] Simplify a couple ifs
|
1 year ago |
|
Behdad Esfahbod
|
e527c17e23
|
[buffer-verify] Don't consider shaping failure as error
|
1 year ago |
|
Behdad Esfahbod
|
cb516075b6
|
[buffer-verify] Error handling
|
1 year ago |
|
Behdad Esfahbod
|
39dd777a12
|
Merge pull request #4271 from googlefonts/change_axis_limits
[instancer] change tuple variations' axis limits
|
1 year ago |
Qunxin Liu
|
3669a62710
|
fix bot
|
1 year ago |
|
Qunxin Liu
|
9fd367663a
|
[instancer] add testing code for change_tuple_variation_axis_limits()
Also fixed a small bug
|
1 year ago |
|
Qunxin Liu
|
8057661f07
|
[instancer] use axis tag as hashmap key instead of axis index
This makes remove_axis() and set_tent() faster, which are used by
change_axis_limits ()
|
1 year ago |
|
Qunxin Liu
|
389446c563
|
[instancer] add merge_tuple_variations ()
|
1 year ago |
|
Behdad Esfahbod
|
fea47dd3f2
|
[vector] Fix vector error handling when allocation used to be 0
|
1 year ago |
|
Behdad Esfahbod
|
11308c4d1f
|
[graph] Remove manual destruction
Happens automatically by destructor.
|
1 year ago |
|
Behdad Esfahbod
|
d08aee5a7e
|
Add fuzzing test
|
1 year ago |
|
Behdad Esfahbod
|
dbdeb2649d
|
[vector] Keep allocated size when in error
|
1 year ago |
|
Behdad Esfahbod
|
9df07c3c30
|
[gsubgpos] Prevent a leak in closure
If the push wasn't successful we were constructing an
hb_set_t on the Crap data... At least that's my reading
of the code.
|
1 year ago |
|
Behdad Esfahbod
|
0f0b3bee21
|
[map] Fix use-after-move issue
|
1 year ago |
|
Behdad Esfahbod
|
a67a7867d2
|
[cff] Tweak CFFIndex accessors again
Faster; avoiding multiply.
|
1 year ago |
|
Behdad Esfahbod
|
04c5e46ac6
|
[subset/cff] Fix comment
|
1 year ago |
|
Behdad Esfahbod
|
8832da8309
|
[subset/cff] Comment
|
1 year ago |
|
Behdad Esfahbod
|
092373f351
|
[subset/cff] Comment
|
1 year ago |
|
Behdad Esfahbod
|
e0b60bd08d
|
[subset/cff1] Speed up remap_sids
|
1 year ago |
|
Behdad Esfahbod
|
67b1624727
|
[set] Simplify a few set iterations as range loop
|
1 year ago |
|
Behdad Esfahbod
|
988e4f068e
|
[iter] Comment
|
1 year ago |
|
Behdad Esfahbod
|
0364c69e6d
|
[iter] Add has_fast_len
Set iterators are not random_access, but have fast len().
|
1 year ago |
|
Behdad Esfahbod
|
2d1589221d
|
[subset/hdmx] Remove an unintended vector copy
|
1 year ago |
|
Behdad Esfahbod
|
81b2a56d54
|
[set] Micro-optimize iteration
|
1 year ago |
|
Behdad Esfahbod
|
ad620af6f6
|
Revert "[iter] Another try at writing some for loops as range loops"
This reverts commit 69c6928289.
This had code size increase, and slows down non-random-access
iterators since it accesses __end__ which is O(n).
|
1 year ago |
|
Behdad Esfahbod
|
80d6e996d7
|
[subset/cff1] Fix wrong comma location!
|
1 year ago |
|
Behdad Esfahbod
|
69c6928289
|
[iter] Another try at writing some for loops as range loops
|
1 year ago |
|
Behdad Esfahbod
|
138461beb4
|
[susbet/cff] Minor remove one indirection
|
1 year ago |
|
Behdad Esfahbod
|
41369b6617
|
Revert "[subset/cff1] Optimize writing of consecutive names"
This reverts commit 3b25a63040.
Not worth it. Shows 0.5% speedup only on SourceSansPro/retaingids/10
Doesn't have code size increase though.
|
1 year ago |
|
Behdad Esfahbod
|
3b25a63040
|
[subset/cff1] Optimize writing of consecutive names
|
1 year ago |
|
Behdad Esfahbod
|
78082357c8
|
Revert "[iter] Rewrite some loops as range loops"
This reverts commit 7a5242a30f.
This times out test-set. Obviously broke something (in hb_all
I think). Not bothering to figure out right now.
|
1 year ago |
|
Behdad Esfahbod
|
7a5242a30f
|
[iter] Rewrite some loops as range loops
|
1 year ago |
|
Behdad Esfahbod
|
cc9651d55d
|
[map] Micro-optimize
|
1 year ago |
|
Behdad Esfahbod
|
79113ec3b1
|
[subset/cff1] Micro-optimize String writing
|
1 year ago |
|
Behdad Esfahbod
|
5fe96213aa
|
[subset/cff] Micro-optimize CFFIndex writing
|
1 year ago |
|
Behdad Esfahbod
|
70b13ef00b
|
Merge pull request #4267 from googlefonts/decompile_tuple_varstore
[instancer] decompile tuple varstore
|
1 year ago |
|
Behdad Esfahbod
|
b80b6a4f5f
|
[cff] Remove unused types
|
1 year ago |
|
Behdad Esfahbod
|
26ac1d4b48
|
[cff] Remove unused method
|
1 year ago |
|
Behdad Esfahbod
|
c6ce1f81ec
|
[cff] Micro-optimize CFFIndex::operator[]
|
1 year ago |
|
Behdad Esfahbod
|
a55d0b8866
|
[subset/cff] Inline type only used once
|
1 year ago |
|
Behdad Esfahbod
|
2960d13f1b
|
[subset/cff1] Micro-optimize string writing
|
1 year ago |
|
Behdad Esfahbod
|
35e152b9f4
|
[subset/cff] Micro-optimize CFFIndex for empty strings
|
1 year ago |
|
Behdad Esfahbod
|
acae5ed25c
|
[subset/cff1] Speed up sid mapping
Don't need a full inc_bimap.
|
1 year ago |
|
Behdad Esfahbod
|
9e80f6b1b6
|
[subset/cff1] Speed up plan_subset_charset
25% speedup in --benchmark_filter=subset_glyphs/SourceSansPro'.*retaingids/10
|
1 year ago |
|
Behdad Esfahbod
|
fd13aa9eff
|
[subset/cff1] Optimize CFF::CFF1StringIndex::serialize
|
1 year ago |
|
Behdad Esfahbod
|
ada1e9a924
|
[graph/serialize] Handle empty blob
Fixes https://oss-fuzz.com/testcase-detail/4877513265119232
|
1 year ago |
|
Behdad Esfahbod
|
a92b288e65
|
[serializer] Handle snapshotting when current is nullptr
Happens with memory failure / fuzzing.
Fixes https://oss-fuzz.com/testcase-detail/6292420615340032
|
1 year ago |
|
Qunxin Liu
|
01f9b79777
|
address review comments
|
1 year ago |
