platform/upstream/libHarfBuzzSharp.git
5 years agoRemove F_COMBINE
Behdad Esfahbod [Mon, 24 Sep 2018 22:03:34 +0000 (18:03 -0400)]
Remove F_COMBINE

Now I wonder if any bots will be unhappy we calling | in static const initializations...
Or would that cost runtime init?  Our tests don't detect any..

5 years agoSome more
Behdad Esfahbod [Mon, 24 Sep 2018 22:01:53 +0000 (18:01 -0400)]
Some more

5 years agoMinor refactoring
Behdad Esfahbod [Mon, 24 Sep 2018 21:55:03 +0000 (17:55 -0400)]
Minor refactoring

5 years ago[circleci] Update sanitizer bots with newer clang and Ubuntu version (#1176)
Ebrahim Byagowi [Mon, 24 Sep 2018 21:30:32 +0000 (01:00 +0330)]
[circleci] Update sanitizer bots with newer clang and Ubuntu version (#1176)

5 years ago[msan] Enable again to get a build log, and try to play with suppressions
Behdad Esfahbod [Mon, 24 Sep 2018 21:22:39 +0000 (17:22 -0400)]
[msan] Enable again to get a build log, and try to play with suppressions

5 years ago[ubsan] Re-enable now that it passes locally
Behdad Esfahbod [Mon, 24 Sep 2018 17:12:15 +0000 (13:12 -0400)]
[ubsan] Re-enable now that it passes locally

5 years ago[check-static-inits.sh] Allow some if ubsan is in effect
Behdad Esfahbod [Mon, 24 Sep 2018 17:11:34 +0000 (13:11 -0400)]
[check-static-inits.sh] Allow some if ubsan is in effect

5 years ago[check-symbols.sh] Allow weak objects "V"
Behdad Esfahbod [Mon, 24 Sep 2018 17:04:48 +0000 (13:04 -0400)]
[check-symbols.sh] Allow weak objects "V"

ubsan generates these.

5 years agoLink API test programs with C++ linker
Behdad Esfahbod [Mon, 24 Sep 2018 17:01:35 +0000 (13:01 -0400)]
Link API test programs with C++ linker

Needed to make ubsan work.

Fixes https://github.com/harfbuzz/harfbuzz/issues/1174

5 years ago[msan] Add MSAN_OPTIONS=exitcode=42
Behdad Esfahbod [Mon, 24 Sep 2018 16:52:42 +0000 (12:52 -0400)]
[msan] Add MSAN_OPTIONS=exitcode=42

Default exit value is 77, which causes autotools to "skip" test.

5 years ago[circleci] Disable msan and ubsan builds
Behdad Esfahbod [Mon, 24 Sep 2018 16:43:29 +0000 (12:43 -0400)]
[circleci] Disable msan and ubsan builds

https://github.com/harfbuzz/harfbuzz/issues/1174
https://github.com/harfbuzz/harfbuzz/issues/1175

5 years agoIgnore weak symbols in check-symbols.sh
Behdad Esfahbod [Mon, 24 Sep 2018 15:52:22 +0000 (11:52 -0400)]
Ignore weak symbols in check-symbols.sh

Some clang versions leave a std::round(float) weak symbol around...

5 years ago[docs] A few improvements
Behdad Esfahbod [Mon, 24 Sep 2018 14:43:06 +0000 (10:43 -0400)]
[docs] A few improvements

If we wrote just this much every day...

5 years agoReadjust Hebrew composition again
Behdad Esfahbod [Mon, 24 Sep 2018 13:57:52 +0000 (09:57 -0400)]
Readjust Hebrew composition again

5 years agoRevert change that would decompose text if GPOS mark feature is available
Behdad Esfahbod [Mon, 24 Sep 2018 13:56:18 +0000 (09:56 -0400)]
Revert change that would decompose text if GPOS mark feature is available

https://github.com/harfbuzz/harfbuzz/issues/653#issuecomment-423905920

5 years agoFix normalization
Behdad Esfahbod [Mon, 24 Sep 2018 13:54:37 +0000 (09:54 -0400)]
Fix normalization

https://github.com/harfbuzz/harfbuzz/commit/62d1e0852a5549a1b510ad46a4b89f12730bb708#commitcomment-30613091

5 years agoSimplify ZWJ-skipping a bit
Behdad Esfahbod [Mon, 24 Sep 2018 02:33:38 +0000 (22:33 -0400)]
Simplify ZWJ-skipping a bit

Towards disabling ZWJ-skipping in certain GPOS contexts.

Part of https://github.com/flutter/flutter/issues/16886

5 years agoReinit skippy iters when auto_zwj / auto_zwnj change
Behdad Esfahbod [Mon, 24 Sep 2018 02:00:34 +0000 (22:00 -0400)]
Reinit skippy iters when auto_zwj / auto_zwnj change

Ouch.  How did we not hit this bug before...

5 years agoPrefer decomposed form if font has GPOS mark feature
Behdad Esfahbod [Mon, 24 Sep 2018 01:32:18 +0000 (21:32 -0400)]
Prefer decomposed form if font has GPOS mark feature

Fixes https://github.com/harfbuzz/harfbuzz/issues/653

5 years ago[ot-font] Fix leak
Behdad Esfahbod [Sun, 23 Sep 2018 23:12:52 +0000 (19:12 -0400)]
[ot-font] Fix leak

Fixes https://github.com/harfbuzz/harfbuzz/issues/1171

I'm glad we have leak-detector bots now.

5 years agoAdd bots with address- and thread-sanitizer
Ebrahim Byagowi [Sat, 22 Sep 2018 10:52:17 +0000 (14:22 +0330)]
Add bots with address- and thread-sanitizer

5 years agosrc/hb-blob.cc: Fix mmap functionality with UWP.
Matt Oliver [Sun, 23 Sep 2018 08:08:30 +0000 (18:08 +1000)]
src/hb-blob.cc: Fix mmap functionality with UWP.

5 years ago[test] Fix printing subprocess error messages
Khaled Hosny [Sat, 22 Sep 2018 14:41:56 +0000 (16:41 +0200)]
[test] Fix printing subprocess error messages

Decode the string as Python 3 returns bytes string, and also don’t
assume that it ends with a new line.

5 years agoFix test-subset-* leak issues
Ebrahim Byagowi [Sat, 22 Sep 2018 13:54:03 +0000 (17:24 +0330)]
Fix test-subset-* leak issues

See also https://github.com/harfbuzz/harfbuzz/pull/1169

5 years agoFix test-ot-math leak issue
Ebrahim Byagowi [Sat, 22 Sep 2018 13:35:52 +0000 (17:05 +0330)]
Fix test-ot-math leak issue

See also https://github.com/harfbuzz/harfbuzz/pull/1169

5 years agoFix test-set leak issue
Ebrahim Byagowi [Sat, 22 Sep 2018 13:19:23 +0000 (16:49 +0330)]
Fix test-set leak issue

See also https://github.com/harfbuzz/harfbuzz/pull/1169

5 years agoFix test-font leak issue
Ebrahim Byagowi [Sat, 22 Sep 2018 13:15:31 +0000 (16:45 +0330)]
Fix test-font leak issue

See also https://github.com/harfbuzz/harfbuzz/pull/1169

5 years ago[morx] Fix mark_set check
Behdad Esfahbod [Thu, 20 Sep 2018 02:34:09 +0000 (22:34 -0400)]
[morx] Fix mark_set check

5 years ago[aat] Minor
Behdad Esfahbod [Wed, 19 Sep 2018 21:32:21 +0000 (17:32 -0400)]
[aat] Minor

5 years ago[coretext] Minor
Behdad Esfahbod [Wed, 19 Sep 2018 21:10:26 +0000 (17:10 -0400)]
[coretext] Minor

5 years ago[morx] Implement forward/backward processing
Behdad Esfahbod [Wed, 19 Sep 2018 20:46:41 +0000 (16:46 -0400)]
[morx] Implement forward/backward processing

We reverse too many times. Can be optimized. But I doubt many fonts
use reverse lookups, so doesn't matter.

Other than not applying user features, this completes morx table
implementation.

5 years ago[morx] Implement horiz-only/vert-only subtables
Behdad Esfahbod [Wed, 19 Sep 2018 20:24:34 +0000 (16:24 -0400)]
[morx] Implement horiz-only/vert-only subtables

5 years ago[morx] Improve buffer position in Insertion w DontAdvance and wo
Behdad Esfahbod [Wed, 19 Sep 2018 20:02:56 +0000 (16:02 -0400)]
[morx] Improve buffer position in Insertion w DontAdvance and wo

Just speculation. Needs tests to determine which is correct behavior.

5 years ago[morx] Implement InsertionChain
Behdad Esfahbod [Mon, 17 Sep 2018 17:29:34 +0000 (19:29 +0200)]
[morx] Implement InsertionChain

5 years agoDon't enforce a native direction for Old Hungarian
David Corbett [Mon, 17 Sep 2018 15:09:51 +0000 (11:09 -0400)]
Don't enforce a native direction for Old Hungarian

5 years agoFix installation of cmake config when building with cmake (#1161)
Simon Hausmann [Mon, 17 Sep 2018 08:33:34 +0000 (10:33 +0200)]
Fix installation of cmake config when building with cmake (#1161)

When building with the auto tools, the manually mantained
harfbuzz-config.cmake is installed. When building with cmake, we can use
cmake to generate the correct config files for us and install them.

5 years agoRevert forcing use of single-parameter static_assert()
Behdad Esfahbod [Sun, 16 Sep 2018 17:33:48 +0000 (19:33 +0200)]
Revert forcing use of single-parameter static_assert()

Some clang versions define static_assert as a macro apparently, so we cannot
redefine it...

This reverts commit 94bfea0ce6a7b4d5641c198d50751748a353df11.
This reverts commit 4e62627831e7457ed60ff87712570065b14b200a.

5 years agoOne more bot fix
Behdad Esfahbod [Sun, 16 Sep 2018 17:25:19 +0000 (19:25 +0200)]
One more bot fix

5 years agoFix build, another try
Behdad Esfahbod [Sun, 16 Sep 2018 17:12:19 +0000 (19:12 +0200)]
Fix build, another try

5 years agoFix bots
Behdad Esfahbod [Sun, 16 Sep 2018 16:11:46 +0000 (18:11 +0200)]
Fix bots

"Unused private member" warning turned error.  ugh.

5 years agoEnforce single-param static_assert() only
Behdad Esfahbod [Sun, 16 Sep 2018 16:09:36 +0000 (18:09 +0200)]
Enforce single-param static_assert() only

So we don't accidentally break it again.

5 years agoRemove unused try
Behdad Esfahbod [Sun, 16 Sep 2018 16:01:32 +0000 (18:01 +0200)]
Remove unused try

5 years agoFix static_assert
Behdad Esfahbod [Sun, 16 Sep 2018 15:57:12 +0000 (17:57 +0200)]
Fix static_assert

5 years agoDisallow null-enabled offsets to unsized structures...
Behdad Esfahbod [Sat, 15 Sep 2018 17:43:33 +0000 (19:43 +0200)]
Disallow null-enabled offsets to unsized structures...

...like UnsizedArrayOf<>.

This fixes a class of crasher bugs, mostly with color and AAT tables.  We
cannot use nullable offsets to varsized data that does not declare min_size,
because it's nost safe to use our fixed-size null pool for types that have
their size external.  So, use non_null'able offsets for these.

A further enhancement would be to make use of min_size in Null<> itself.
Will try that after.

5 years ago[morx] Respect default feature settings
Behdad Esfahbod [Sat, 15 Sep 2018 16:31:14 +0000 (18:31 +0200)]
[morx] Respect default feature settings

Does NOT apply user-selected features.  But at least now enables
correct subtables.

5 years ago[aat] Change version field
Behdad Esfahbod [Sat, 15 Sep 2018 12:51:50 +0000 (14:51 +0200)]
[aat] Change version field

5 years ago[morx] Add stub for InsertionChain
Behdad Esfahbod [Sat, 15 Sep 2018 12:47:18 +0000 (14:47 +0200)]
[morx] Add stub for InsertionChain

5 years ago[dfont] Disable null-processsing for offsets
Behdad Esfahbod [Fri, 14 Sep 2018 16:59:53 +0000 (18:59 +0200)]
[dfont] Disable null-processsing for offsets

An offset to unsized arrayis not safe to be redirected to our fixed-sized
null pool.  Plus, we want to reject, not repair, bad-looking dfonts.

5 years ago[unicode] Move Fitzpatrick hack from ot-layout into unicode.hh
Behdad Esfahbod [Fri, 14 Sep 2018 12:23:09 +0000 (14:23 +0200)]
[unicode] Move Fitzpatrick hack from ot-layout into unicode.hh

5 years agoMinor
Behdad Esfahbod [Fri, 14 Sep 2018 10:15:53 +0000 (12:15 +0200)]
Minor

5 years agoFix builds
Behdad Esfahbod [Fri, 14 Sep 2018 10:14:42 +0000 (12:14 +0200)]
Fix builds

5 years ago[aat] Add enums for pre-defined state and classes
Behdad Esfahbod [Fri, 14 Sep 2018 09:31:33 +0000 (11:31 +0200)]
[aat] Add enums for pre-defined state and classes

Not sure how I didn't add before...

5 years agoDon't dereference offset before check_struct()
Behdad Esfahbod [Fri, 14 Sep 2018 08:58:00 +0000 (10:58 +0200)]
Don't dereference offset before check_struct()

5 years ago[dfont] Also check dataLen range in sanitize
Behdad Esfahbod [Thu, 13 Sep 2018 18:35:21 +0000 (20:35 +0200)]
[dfont] Also check dataLen range in sanitize

Just to disagree with myself re being done with this code...

5 years ago[dfont] Solve the mystery +2 offset thing!
Behdad Esfahbod [Thu, 13 Sep 2018 18:30:04 +0000 (20:30 +0200)]
[dfont] Solve the mystery +2 offset thing!

Previously, ResourceForkHeader was defined as 30 bytes, having the typeCountM1 as last member.
There was a mysterious offset-by-2 in the code, derived from FontTools and JDK code this was
ported from.

In testing, I observed that typeListZ offset is actually 28.  Suggesting that the typeCountM1
does NOT actually belong to ResourceForkHeader, but belongs to the array itself.  Adjusting for
that resolves the mystery +2 offset hack, so everything is clean and good now.

This, concludes my dfont hacking.  The code looks great now, and I'm happy to leave it.
Fuzzers might disagree though, we will see!

5 years agoRename HeadlessArrayOf::len to lenP1
Behdad Esfahbod [Thu, 13 Sep 2018 18:21:54 +0000 (20:21 +0200)]
Rename HeadlessArrayOf::len to lenP1

So it doesn't accidentally match our templates, etc.

5 years ago[dfont] Some more
Behdad Esfahbod [Thu, 13 Sep 2018 17:19:57 +0000 (19:19 +0200)]
[dfont] Some more

5 years ago[dfont] Push methods further down
Behdad Esfahbod [Thu, 13 Sep 2018 17:13:01 +0000 (19:13 +0200)]
[dfont] Push methods further down

5 years ago[dfont] Minor
Behdad Esfahbod [Thu, 13 Sep 2018 17:08:22 +0000 (19:08 +0200)]
[dfont] Minor

5 years ago[dfon]t Sanitize OpenTypeFontFace
Behdad Esfahbod [Thu, 13 Sep 2018 17:05:59 +0000 (19:05 +0200)]
[dfon]t Sanitize OpenTypeFontFace

5 years ago[dfont] Minor
Behdad Esfahbod [Thu, 13 Sep 2018 16:49:16 +0000 (18:49 +0200)]
[dfont] Minor

5 years ago[dfont] Make test pass
Behdad Esfahbod [Thu, 13 Sep 2018 16:47:53 +0000 (18:47 +0200)]
[dfont] Make test pass

Offset 0 is not null in this context.

5 years agoAllow Offset<>'s that have no 0==null
Behdad Esfahbod [Thu, 13 Sep 2018 16:45:35 +0000 (18:45 +0200)]
Allow Offset<>'s that have no 0==null

5 years ago[dfont] Minor
Behdad Esfahbod [Thu, 13 Sep 2018 16:27:20 +0000 (18:27 +0200)]
[dfont] Minor

5 years ago[bytes] Rename content to arrayZ
Behdad Esfahbod [Thu, 13 Sep 2018 15:39:09 +0000 (17:39 +0200)]
[bytes] Rename content to arrayZ

5 years ago[dfont] Clean up sanitize()
Behdad Esfahbod [Thu, 13 Sep 2018 14:49:26 +0000 (16:49 +0200)]
[dfont] Clean up sanitize()

I don't think I broke anything.  Fuzzers will let me know..

5 years agoFix OffsetTo::sanitize() after reshuffling
Behdad Esfahbod [Thu, 13 Sep 2018 14:47:33 +0000 (16:47 +0200)]
Fix OffsetTo::sanitize() after reshuffling

5 years agoOffsetTo::sanitize() Add version with three user_data
Behdad Esfahbod [Thu, 13 Sep 2018 14:39:30 +0000 (16:39 +0200)]
OffsetTo::sanitize() Add version with three user_data

5 years agoOffsetTo::sanitize() more shuffling
Behdad Esfahbod [Thu, 13 Sep 2018 14:31:31 +0000 (16:31 +0200)]
OffsetTo::sanitize() more shuffling

5 years agoOffsetTo::sanitize() reshuffling
Behdad Esfahbod [Thu, 13 Sep 2018 14:29:49 +0000 (16:29 +0200)]
OffsetTo::sanitize() reshuffling

5 years ago[dfont] Some renaming, plus add link to reference doc
Behdad Esfahbod [Tue, 11 Sep 2018 16:12:26 +0000 (18:12 +0200)]
[dfont] Some renaming, plus add link to reference doc

5 years ago[dfont] Sanitize only sfnt resources as OpenTypeFontFile
Behdad Esfahbod [Tue, 11 Sep 2018 15:56:03 +0000 (17:56 +0200)]
[dfont] Sanitize only sfnt resources as OpenTypeFontFile

5 years ago[dfont] Minor
Behdad Esfahbod [Tue, 11 Sep 2018 15:26:24 +0000 (17:26 +0200)]
[dfont] Minor

5 years ago[dfont] Fix test expecatation and minor touch up
Behdad Esfahbod [Tue, 11 Sep 2018 15:18:21 +0000 (17:18 +0200)]
[dfont] Fix test expecatation and minor touch up

I have no way to authoritatively know, but looks like test font only has one
face.  So, adjust test expectation instead.

5 years ago[dfont] Add test
Ebrahim Byagowi [Mon, 2 Jul 2018 12:56:43 +0000 (17:26 +0430)]
[dfont] Add test

5 years ago[dfont] Re-enable and fix offset handling
Behdad Esfahbod [Tue, 11 Sep 2018 14:41:26 +0000 (16:41 +0200)]
[dfont] Re-enable and fix offset handling

Fixes https://github.com/harfbuzz/harfbuzz/pull/1085

5 years agoWhitespace
Behdad Esfahbod [Tue, 11 Sep 2018 12:45:23 +0000 (14:45 +0200)]
Whitespace

5 years ago[ft] Invalidate advance cache if font size changed
Behdad Esfahbod [Tue, 11 Sep 2018 12:41:19 +0000 (14:41 +0200)]
[ft] Invalidate advance cache if font size changed

5 years ago[ft] Cache advances
Behdad Esfahbod [Tue, 11 Sep 2018 12:35:26 +0000 (14:35 +0200)]
[ft] Cache advances

I decided to always use the cache, instead of my previous sketch direction
that was to only allocate and use cache if fast advances are not available.
The cache is a mere 1kb, so just use it...

TODO: Invalidate cache on font size change.

Fixes https://github.com/harfbuzz/harfbuzz/issues/651
Fixes https://github.com/harfbuzz/harfbuzz/pull/1082

5 years ago[util] Add --ft-load-flags
Behdad Esfahbod [Tue, 11 Sep 2018 12:21:15 +0000 (14:21 +0200)]
[util] Add --ft-load-flags

Useful for performance testing.

Not hooked to cairo yet.  Just changes shaping, not rasterization.

5 years ago[util] Add -n shorthand for --num-iterations
Behdad Esfahbod [Tue, 11 Sep 2018 12:11:10 +0000 (14:11 +0200)]
[util] Add -n shorthand for --num-iterations

Meh.

5 years ago[ft] Towards caching slow get_h_advance results
Behdad Esfahbod [Tue, 11 Sep 2018 12:05:16 +0000 (14:05 +0200)]
[ft] Towards caching slow get_h_advance results

Related to https://github.com/harfbuzz/harfbuzz/pull/1082

5 years ago[ft] Add advances() callback
Behdad Esfahbod [Tue, 11 Sep 2018 11:05:47 +0000 (13:05 +0200)]
[ft] Add advances() callback

5 years ago[ot-font] Rename
Behdad Esfahbod [Tue, 11 Sep 2018 10:56:17 +0000 (12:56 +0200)]
[ot-font] Rename

5 years ago[ot-font] Add advances() callbacks
Behdad Esfahbod [Tue, 11 Sep 2018 10:55:54 +0000 (12:55 +0200)]
[ot-font] Add advances() callbacks

5 years ago[cache] Minor
Behdad Esfahbod [Tue, 11 Sep 2018 10:22:42 +0000 (12:22 +0200)]
[cache] Minor

5 years agoRevert "Remove unused hb_cache_t"
Behdad Esfahbod [Tue, 11 Sep 2018 10:02:34 +0000 (12:02 +0200)]
Revert "Remove unused hb_cache_t"

This reverts commit 473b17af4d421f4ce7ac18c769731bb2aa4088f8.

Updates to recent changes.

5 years ago[random] Switch to 32bit RNG
Behdad Esfahbod [Tue, 11 Sep 2018 08:57:48 +0000 (10:57 +0200)]
[random] Switch to 32bit RNG

5 years ago[random] Shuffle
Behdad Esfahbod [Tue, 11 Sep 2018 08:51:19 +0000 (10:51 +0200)]
[random] Shuffle

5 years agoMake --features rand=1 available to the user
Behdad Esfahbod [Mon, 10 Sep 2018 20:37:19 +0000 (22:37 +0200)]
Make --features rand=1 available to the user

Use rand=255 to mean "randomize".

Part of https://github.com/harfbuzz/harfbuzz/pull/803

5 years agoRespect user's wish if they set rand feature manually
Behdad Esfahbod [Mon, 10 Sep 2018 14:36:05 +0000 (16:36 +0200)]
Respect user's wish if they set rand feature manually

Except if the set it to 1, which would mean "randomize"... Ugly.

5 years agoMinor clean up of 'rand' patchset
Behdad Esfahbod [Mon, 10 Sep 2018 14:24:52 +0000 (16:24 +0200)]
Minor clean up of 'rand' patchset

5 years agoDon't seed the RNG from the contents of the buffer
David Corbett [Fri, 23 Feb 2018 17:22:32 +0000 (12:22 -0500)]
Don't seed the RNG from the contents of the buffer

5 years agoTest 'rand'
David Corbett [Sun, 28 Jan 2018 00:48:38 +0000 (19:48 -0500)]
Test 'rand'

5 years agoAllow requesting a specific glyph for 'rand'
David Corbett [Sat, 27 Jan 2018 02:36:15 +0000 (21:36 -0500)]
Allow requesting a specific glyph for 'rand'

Randomization only happens by default. If the user specifies a value for
'rand', that value is respected.

5 years agoImplement 'rand'
David Corbett [Thu, 25 Jan 2018 19:22:03 +0000 (14:22 -0500)]
Implement 'rand'

5 years ago[uniscribe] Fix build
Behdad Esfahbod [Mon, 10 Sep 2018 23:39:23 +0000 (01:39 +0200)]
[uniscribe] Fix build

5 years agoAdd HB_FEATURE_GLOBAL_START/END
Behdad Esfahbod [Mon, 10 Sep 2018 23:26:18 +0000 (01:26 +0200)]
Add HB_FEATURE_GLOBAL_START/END

Fixes https://github.com/harfbuzz/harfbuzz/issues/1141

New API:
HB_FEATURE_GLOBAL_START
HB_FEATURE_GLOBAL_END

5 years agoAdd (unused) hb_array_t<>
Behdad Esfahbod [Mon, 10 Sep 2018 23:07:06 +0000 (01:07 +0200)]
Add (unused) hb_array_t<>

5 years agoMinor
Behdad Esfahbod [Mon, 10 Sep 2018 23:04:50 +0000 (01:04 +0200)]
Minor