Commit Graph

46 Commits

Author SHA1 Message Date
Behdad Esfahbod 4f252fedc7 Remove SANITIZE macro 2010-05-06 13:30:23 -04:00
Behdad Esfahbod 41a93d2c1f Remove SANITIZE_WITH_BASE 2010-05-06 12:55:14 -04:00
Behdad Esfahbod 1cd1e117d0 Remove SANITIZE_ARRAY 2010-05-05 20:15:14 -04:00
Behdad Esfahbod 583d7f9586 Cosmetic 2010-05-05 01:49:22 -04:00
Behdad Esfahbod 39840474af Remove SANITIZE_ARG_DEF and SANITIZE_ARG 2010-05-05 00:23:19 -04:00
Behdad Esfahbod bb029af943 Remove SANITIZE_THIS 2010-05-04 15:28:52 -04:00
Behdad Esfahbod ce5694c796 [main] Recognize Apple SFNTs 2010-05-04 14:10:18 -04:00
Behdad Esfahbod 710500a93e Comment new SFNT tags 2010-05-03 23:11:16 -04:00
Behdad Esfahbod 64d3fc8d0d Cosmetic: Rename HB_LIKELY/HB_UNLIKELY to likely/unlikely 2010-05-03 22:51:19 -04:00
Behdad Esfahbod 6b84198f9d Merge remote branch 'jrmuizel/master' 2010-05-03 22:46:52 -04:00
Jeff Muizelaar 4ce578ed36 Include the tags from the Apple specification for TrueType fonts 2010-05-03 15:03:53 -04:00
Behdad Esfahbod 631d10b728 Remove unused method 2010-05-02 21:14:21 -04:00
Behdad Esfahbod 690b919461 Remove use of flexible arrays
Also remove wrong ASSERT_SIZE that would trigger only when not using
flexible arrays (which was the case on win32, but not with gcc).
2010-04-25 22:52:22 -04:00
Behdad Esfahbod efb324a46f Remove GET_FOR_DATA macros
The major-version check is now handled by sanitize.  If major
doesn't match, we reject and fall back to the Null object.
2010-04-23 16:28:27 -04:00
Behdad Esfahbod f1aaa2a436 Add TODO 2010-04-23 15:19:50 -04:00
Behdad Esfahbod ef66ebeabc Remove unused parameter 2010-04-23 14:40:36 -04:00
Behdad Esfahbod 1aa4666b91 Cleanup OpenTypeFontFile 2010-04-23 13:32:03 -04:00
Behdad Esfahbod ae4190cafe Properly define separate structs for TTCHeader and TTCHeaderVersion1 2010-04-23 12:33:02 -04:00
Behdad Esfahbod a0bb49c583 Add comment re bsearch effect on sanitize 2010-04-22 18:47:03 -04:00
Behdad Esfahbod a3263aa773 Convert the last set of cast macros to templates 2010-04-22 18:42:01 -04:00
Behdad Esfahbod 1856184b93 Fail sanitize on major version mismatch
We handle major-version differences via get_for_data(), so sanitize
should never see a major version mismatch.
2010-04-22 14:21:17 -04:00
Behdad Esfahbod df3f505dcf More sanitize cleanup 2010-04-22 14:11:33 -04:00
Behdad Esfahbod 8015a8c762 Don't sanitize raw table data
That part is performed by individual table sanitize.
2010-04-22 13:21:12 -04:00
Behdad Esfahbod 079dc40112 Avoid overflow in TableDirectory sanitize 2010-04-22 10:47:12 -04:00
Behdad Esfahbod c755cb3e3a Change header comment 2010-04-22 00:11:43 -04:00
Behdad Esfahbod 0dfcc13a46 Rename ConstCharP to CharP (overloaded now) 2010-04-21 23:41:26 -04:00
Behdad Esfahbod 62c0fd7573 Cleanup de-const-casting during sanitize 2010-04-21 23:30:48 -04:00
Behdad Esfahbod 198facdc55 Use templates for const char * casts 2010-04-21 13:35:36 -04:00
Behdad Esfahbod ffd321afd9 Simplify Tag struct 2010-04-21 01:55:09 -04:00
Behdad Esfahbod 4b8487d83e Fix the mystery bug!
A couple bugs joined forces to exhibit the mystery behavior of
crashes / infinite loops on OS X / wrong kerning / invalid memory
access.  Pooh!

The bugs were involved:

  - Wrong pointer math with ValueRecord in PairPosFormat1

  - Fallout from avoiding flex arrays, code not correctly updated
    to remove sizeof() usage.

We strictly never use sizeof() directly now.  And the PairPos code
is cleaned up.  Should fix them all.  Bugs are:

  Bug 605655 - Pango 1.26.2 introduces kerning bug
  Bug 611229 - Pango reads from uninitialized memory
  Bug 593240 - (pangoosx) Crash / infinite loop with Mac OS X

We were also doing wrong math converting Device adjustments to
hb_position_t.  Fallout from FreeType days.  Should shift 16, not
6.  Fixed that too.

There's still another bug: we don't sanitize Device records
referenced from value records.  Fixing that also.
2010-04-21 01:41:29 -04:00
Behdad Esfahbod 636f017e38 Remove obsolete TODO item 2009-11-10 12:56:35 -05:00
Behdad Esfahbod d3480ba37f Don't use zero-sized arrays
It's not part of the standard and MSVC doesn't like it.  It makes the code
a lot less elegant than it used to be, but I think it should work now.
2009-11-03 10:47:29 -05:00
Behdad Esfahbod 3e2401f6c5 [HB] Rename DEBUG to TRACE where appropriate 2009-11-02 14:40:48 -05:00
Behdad Esfahbod 20b035dad4 [HB] Put C++ inline's back
Apparetly in C++, inline means a totally different thing.
2009-11-02 14:40:36 -05:00
Behdad Esfahbod bff3c0fde5 [HB] Remove clumsy macros and improve API 2009-11-02 14:40:34 -05:00
Behdad Esfahbod e605c367c5 [HB] Rebrand XXX as TODO 2009-11-02 14:40:33 -05:00
Behdad Esfahbod 71a8344a5e [HB] Remove a few 'inline's, though the compiler mostly ignores them 2009-11-02 14:40:32 -05:00
Behdad Esfahbod 2870ac5e5d [HB] Add missing sanitize 2009-11-02 14:40:30 -05:00
Behdad Esfahbod b28815c1f6 [HB] Add sanitize debugging facilities 2009-11-02 14:40:30 -05:00
Behdad Esfahbod 7edb430f91 [HB] Rename open-types to open-type; beauty 2009-11-02 14:40:30 -05:00
Behdad Esfahbod 9bd629ccd0 [HB] Fix warnings 2009-11-02 14:40:30 -05:00
Behdad Esfahbod 2765d333e5 [HB] Add XXX 2009-11-02 14:40:30 -05:00
Behdad Esfahbod b508e5ccd5 [HB] open-file sanitize() 2009-11-02 14:40:29 -05:00
Behdad Esfahbod 196598bbcc [Hb] Use reinterpret casts instead of direct casts to char * 2009-11-02 14:40:28 -05:00
Behdad Esfahbod cd3827ee56 [HB] More sanitize() 2009-11-02 14:40:28 -05:00
Behdad Esfahbod 5f5b24f99f [OT] Rename C++ header files from *.h to *.hh 2009-11-02 14:40:27 -05:00