Commit Graph

9623 Commits

Author SHA1 Message Date
Behdad Esfahbod 16e5ba85b1 [pragma] More 2019-01-18 09:46:17 -05:00
Behdad Esfahbod 2bd530d598 [pragma] Reorg again
https://github.com/harfbuzz/harfbuzz/issues/1546
2019-01-18 09:28:22 -05:00
Behdad Esfahbod c04272fac8 [pragma] GCC diagnostic ignored "-Wunknown-warning-option"
Try at fixing https://github.com/harfbuzz/harfbuzz/issues/1546
2019-01-18 08:49:45 -05:00
Behdad Esfahbod 4d3c8f9685 [uniscribe] More warnings 2019-01-17 19:58:54 -05:00
Behdad Esfahbod 6a5d2b21d8 [uniscribe] Fix warnings 2019-01-17 19:57:17 -05:00
Behdad Esfahbod 5b13e458c3 Fix build 2019-01-17 19:52:32 -05:00
Behdad Esfahbod a77bd6bb89 [pragma] GCC diagnostic ignored "-Wclass-memaccess" 2019-01-17 19:33:41 -05:00
Behdad Esfahbod e1302842d9 [ci] Remove unnecessary warning declarations
clang -Weverything still left to be ported to pragmas
2019-01-17 19:12:07 -05:00
Behdad Esfahbod 436c7b294a [pragma] ignored -Wunsafe-loop-optimizations // TODO fix 2019-01-17 19:03:33 -05:00
Behdad Esfahbod e3b3181e9d [pragma] error -Wclass-memaccess 2019-01-17 19:02:06 -05:00
Behdad Esfahbod 1a6b97a491 [pragma] More 2019-01-17 18:55:03 -05:00
Behdad Esfahbod 06e8091661 [pragma] More 2019-01-17 18:47:32 -05:00
Behdad Esfahbod 1077e40bf1 [pragma] Flesh out more 2019-01-17 18:36:10 -05:00
Behdad Esfahbod 7e6bd510fc Use NNOffsetTo<> in rest of places 2019-01-17 18:24:18 -05:00
Behdad Esfahbod b1152d5e66 Use NNOffsetTo<> 2019-01-17 18:17:04 -05:00
Behdad Esfahbod 205d72a198 Add NNOffsetTo<> 2019-01-17 18:10:38 -05:00
Behdad Esfahbod c99d13d860 [AAT] Comment 2019-01-17 17:56:27 -05:00
Behdad Esfahbod f39b5603ff Comment 2019-01-17 17:50:01 -05:00
Behdad Esfahbod 347ad454b8 [AAT] Remove unused code 2019-01-17 17:47:29 -05:00
Behdad Esfahbod b344d4385a [AAT] Allow null in ankr lookupTable offset 2019-01-17 17:44:39 -05:00
Behdad Esfahbod 14a560a245 [pragma GCC] Ignore -Wtype-limits 2019-01-17 17:42:44 -05:00
Behdad Esfahbod 41b4312e93 [ci] Remove -Werror from Travis
hb-machinery.hh:111:37: warning: comparison of unsigned expression >= 0 is always true [-Wtype-limits]

Also ignore -Wtype-limits.
2019-01-17 16:08:30 -05:00
Behdad Esfahbod 3f2daae630 More GCC pragmas 2019-01-17 15:27:31 -05:00
Behdad Esfahbod d6c5473e1f Rename macro 2019-01-17 15:22:46 -05:00
Behdad Esfahbod a9946b7cfe Add more GCC warning pragmas 2019-01-17 15:22:28 -05:00
Behdad Esfahbod 868b0c0c2e Move -Wcast-align to hb.hh 2019-01-17 15:18:18 -05:00
Behdad Esfahbod 8874eef8ff Add pragram GCC diagnostic ignored "-Wunused-macros" 2019-01-17 15:04:44 -05:00
Behdad Esfahbod cc8e9a436f [AAT] Fully sanitize ankr table at sanitize time
Third try to fix access. Followup 6879efc2c1

Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=12532
Fixes https://bugs.chromium.org/p/chromium/issues/detail?id=922303
2019-01-17 14:55:05 -05:00
Behdad Esfahbod 6879efc2c1 [AAT] Fix anchor bound checking, again
Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=12532
Fixes https://bugs.chromium.org/p/chromium/issues/detail?id=922303
2019-01-17 14:08:02 -05:00
Behdad Esfahbod a262eb3d0b [aat] Minor 2019-01-17 14:06:32 -05:00
Behdad Esfahbod c986ca15a6 Improve overflow avoidance
Better fix for 480406cd3e
This way we behave the same on 32bit and 64bit archs.
2019-01-15 13:58:19 -05:00
Tor Arne Vestbø f401f85a5a Remove assumption about Core Text working in 96 DPI
Core Text doesn't actually have a concept of DPI internally, as it
doesn't rasterize anything by itself, it just generates vector paths
that get passed along to Core Graphics.

In practice this means Core Text operates in the classical macOS
logical DPI of 72, with one typographic point corresponding to one
point in the Core Graphics coordinate system, which for a normal
bitmap context then corresponds to one pixel -- or two pixels for
a "retina" context with a 2x scale transform.

Scaling the font point sizes given to HarfBuzz to an assumed DPI
of 96 is problematic with this in mind, as fonts with optical
features such as 'trak' tables for tracking, or color glyphs,
will then base the metrics off of the wrong point size compared
to what the client asked for.

This in turn causes mismatches between the metrics of the shaped
text and the actual rasterization, which doesn't include the 72
to 96 DPI scaling.

If a 96 DPI is needed, such as on the Web, the scaling should be
done outside of HarfBuzz, allowing the client to keep the DPI of
the shaping in sync with the rasterization.

The recommended way to do that is by scaling the font point size,
not by applying a transform to the target Core Graphics context,
to let Core Text choose the right optical features of the target
point size, as described in WWDC 2015 session 804:

  https://developer.apple.com/videos/play/wwdc2015/804/
2019-01-15 13:26:35 +01:00
Behdad Esfahbod 0d2727f4fe
fix FDSelect fuzzing bug (#1539)
Rewrote struct FDSelect3_4.ranges as ArrayOf
Updated FDSelect3_4::sanitize () to call ranges.sanitize ()
nRanges now a function to return a reference to ranges.len
2019-01-14 18:23:17 -08:00
Behdad Esfahbod 9f6172d669 Move _POSIX_SOURCE definition
Fixes https://github.com/harfbuzz/harfbuzz/issues/1308
2019-01-14 20:45:31 -05:00
Behdad Esfahbod 91d774712f [test] Add test for previous commit 2019-01-14 15:31:31 -05:00
Behdad Esfahbod 480406cd3e Fix assertion on address overflow
Fixes https://bugs.chromium.org/p/chromium/issues/detail?id=917031
2019-01-14 15:27:34 -05:00
Behdad Esfahbod 7a6686a589 [AAT] Fix mort ContextualSubtable offset access
Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=12312
2019-01-14 15:09:14 -05:00
Behdad Esfahbod a3fa7d3336 [AAT] Fix ankr table access
Fixes https://bugs.chromium.org/p/chromium/issues/detail?id=918340
2019-01-14 14:37:36 -05:00
Michiharu Ariza 760303d411 deleted a duplicate fontdicts_mod.fini() call (#1538) 2019-01-11 15:27:35 -08:00
Ebrahim Byagowi 5110f3da9a
Add an initial .clang-format config
There are things can be improved on clang-format side I guess before
the full assertion, but is the best we can get for now I guess.

Fixes https://github.com/harfbuzz/harfbuzz/issues/1536
2019-01-09 16:01:11 +03:30
Ebrahim Byagowi 099bca6d40
Minor, hack .editorconfig for vscode use
As VSCode and Atom https://github.com/Microsoft/vscode/issues/44438 don't support it
2019-01-09 15:56:32 +03:30
Michiharu Ariza 7b48641aa8 fix leak in subset_enc_supp_codes (#1537)
oss-fuzz issue 12310
2019-01-08 15:37:48 -05:00
Ebrahim Byagowi 97f67a3c3d
[ci] Enable more of GCC compile warnings (#1533) 2019-01-04 23:10:39 +03:30
Nathan Willis ad95487038 Usermanual: small formatting and wording fixes. 2019-01-04 15:43:07 +02:00
Nathan Willis ed13caddf2 Usermanual: small updates. 2019-01-04 15:43:07 +02:00
cclauss 26c5b54fb0 CircleCI: Test for Python 3 syntax errors and undefined names (#1522)
Catch missing imports and errors like #1520 and #1521

__E901,E999,F821,F822,F823__ are the "_showstopper_" [flake8](http://flake8.pycqa.org) issues that can halt the runtime with a SyntaxError, NameError, etc. Most other flake8 issues are merely "style violations" -- useful for readability but they do not effect runtime safety.
* F821: undefined name `name`
* F822: undefined name `name` in `__all__`
* F823: local variable name referenced before assignment
* E901: SyntaxError or IndentationError
* E999: SyntaxError -- failed to compile a file into an Abstract Syntax Tree
2018-12-31 07:00:43 +03:30
Behdad Esfahbod 686e6f2e40 Fix automake warnings
[skip ci]
2018-12-30 11:28:06 -05:00
cclauss b5c12b9f4d print() is a function in Python 3 (#1520) 2018-12-30 15:37:28 +03:30
cclauss f4da28b1f1 Python 3 fixes to gen-os2-unicode-ranges.py (#1521)
In Python 3, __reload()__ was moved and __sys.setdefaultencoding()__ because the default is already utf-8.  Also __Error()__ is an _undefined name_ and __Exception()__ creates a generic exception.
2018-12-30 15:28:34 +03:30
Ebrahim Byagowi e5989e0962 [cmake] Enable C++11 on CMake (#1519) 2018-12-30 01:53:47 -05:00