summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* [HB] Move code aroundBehdad Esfahbod2009-08-142-19/+21
|
* [HB] Correctly sanitize LigatureAttachBehdad Esfahbod2009-08-144-9/+19
|
* [HB] Simplify MarkBase and MarkLig tooBehdad Esfahbod2009-08-141-79/+48
|
* [HB] Simplify MarkMarkBehdad Esfahbod2009-08-141-13/+16
|
* [HB] Fix possible int overflows during sanitizeBehdad Esfahbod2009-08-143-8/+30
|
* [HB] Debug output for SANITIZE_MEM()Behdad Esfahbod2009-08-141-11/+23
|
* [HB] Use inline function for SANITIZE_MEM()Behdad Esfahbod2009-08-141-13/+20
|
* [HB] Use SANITIZE_THIS() when applicableBehdad Esfahbod2009-08-141-3/+3
|
* [HB] Add note about auditing sanitize code for overflowsBehdad Esfahbod2009-08-142-1/+2
|
* [HB] Add empty hb_shape() APIBehdad Esfahbod2009-08-133-0/+95
|
* [HB] Bitfield packing *is* very delicateBehdad Esfahbod2009-08-131-1/+1
| | | | Fix the second instance.
* [basic-win32] Increase performance of Uniscribe codeDavid E. Hollingsworth2009-08-131-7/+6
| | | | | | | | | | | | | | | | A few weeks ago I reported a performance issue regarding the use of Uniscribe in the win32 basic shaper. It turns out that the issue was with the use of Uniscribe's SCRIPT_CACHE. Despite the name, this value is for caching per-font-plus-size values, not per-script values. See: http://msdn.microsoft.com/en-us/library/dd317726(VS.85).aspx Since shaping occurs on runs containing a single font, we only need one SCRIPT_CACHE in uniscribe_shape. I'm including a patch below that does this. With the patch applied, the time for one of my tests went from 20s to .3s.
* Bug 591511 – hh-blob.c does not compile using mingw on windowsBehdad Esfahbod2009-08-132-5/+5
| | | | Check for sys/mman.h, not mprotect().
* [HB] Fix buffer sizing issue with mingw gccBehdad Esfahbod2009-08-131-7/+7
| | | | | Bit field packing is a delicate field. Issue reported in comment 3 here: http://bugzilla.gnome.org/show_bug.cgi?id=591511
* Release 1.25.31.25.3Behdad Esfahbod2009-08-122-2/+18
|
* Bug 591413 – needs to link with libstdc++Behdad Esfahbod2009-08-122-4/+27
| | | | Enforce -fno-exceptions better.
* Bug 591511 – hb-blob.c does not compile using mingw on windowsBehdad Esfahbod2009-08-122-1/+10
| | | | | Only call mprotect() when available. For optimal performance, a win32 way to make memory writeable needs to be added.
* [HB] Add commentBehdad Esfahbod2009-08-121-1/+1
|
* New public API: pango_glyph_item_get_logical_widths()Behdad Esfahbod2009-08-127-38/+81
|
* Fix compiler warningsBehdad Esfahbod2009-08-122-2/+2
|
* Remove G_GNUC_PURE attr from pango_glyph_string_get_width()Behdad Esfahbod2009-08-121-1/+1
| | | | Since the glyph widths can be adjusted directly.
* [OT] Enlarge lookup arrayBehdad Esfahbod2009-08-121-2/+2
| | | | | We'll fix it properly later when ruleset logic is moved into HarfBuzz. In the mean time, IranNastaliq was hitting the 100 limit.
* [layout] clear state->log_widths after freeingBehdad Esfahbod2009-08-121-0/+2
|
* [HB] Always clear positions if dirty, before returning to userBehdad Esfahbod2009-08-121-2/+2
|
* [HB] Fix _hb_buffer_next() when positioningBehdad Esfahbod2009-08-122-0/+13
| | | | | | We were copying glyphs to output. This should not happen when in GPOS. Back then it was fine, then some optimizations broke then assumption.
* [HB] Fixed signedness warningsBehdad Esfahbod2009-08-121-2/+2
|
* [HB] Ouch. Fix compilation.Behdad Esfahbod2009-08-121-1/+1
|
* [HB] Fix invalid access / overflow on x86-64Behdad Esfahbod2009-08-122-11/+12
| | | | | Bug 591557 – [HB] crash scrolling the evolution message list Bug 591576 – crashed with SIGSEGV at pango
* Bug 591465 – Gimp crashes upon opening the font selection dialogBehdad Esfahbod2009-08-111-3/+3
| | | | Initialize count variables.
* [doc] Update doc templatesBehdad Esfahbod2009-08-111-0/+9
|
* Release 1.25.21.25.2Behdad Esfahbod2009-08-112-2/+10
|
* [doc] Update for ATSUI changeBehdad Esfahbod2009-08-111-1/+1
| | | | Bug 562100 – docs check fails
* Bug 591409 – crash in firefox. Handle non-mmapped FT_FaceBehdad Esfahbod2009-08-111-10/+40
|
* [HB] Add hb_font_funcs_make_immutable()Behdad Esfahbod2009-08-103-2/+22
|
* [HB] Add missing initializerBehdad Esfahbod2009-08-101-1/+3
|
* [OT] Hookup glib unicode funcsBehdad Esfahbod2009-08-107-16/+21
|
* [HB] Add hb-glibBehdad Esfahbod2009-08-106-10/+109
|
* [HB] Add hb_unicode_funcs_make_immutable()Behdad Esfahbod2009-08-103-5/+21
|
* [HB] Use nil unicode funcs in nil faceBehdad Esfahbod2009-08-103-4/+8
|
* [HB] Initialize unicode funcs to nil gettersBehdad Esfahbod2009-08-101-10/+19
|
* [HB] Flesh out Unicode funcsBehdad Esfahbod2009-08-104-19/+65
|
* [HB] Implement buffer UTF-X inputBehdad Esfahbod2009-08-102-6/+116
|
* [HB] Add buffer UTF-X APIBehdad Esfahbod2009-08-101-0/+21
|
* [HB] Add hb_buffer_reverse()Behdad Esfahbod2009-08-102-3/+36
|
* [HB] Reuse the positions array as alt string arrayBehdad Esfahbod2009-08-102-20/+9
|
* [HB] Start adding Unicode funcsBehdad Esfahbod2009-08-108-236/+464
|
* [HB] Put C++ inline's backBehdad Esfahbod2009-08-107-41/+41
| | | | Apparetly in C++, inline means a totally different thing.
* Add CXX check to configureBehdad Esfahbod2009-08-101-10/+1
|
* [HB] Don't override CXX compilerBehdad Esfahbod2009-08-101-1/+2
| | | | Just add couple of options to it (still assuming gcc)
* Release 1.25.11.25.1Behdad Esfahbod2009-08-102-2/+20
|
View Lorry Controller Status