Log message:
Update to 0.9.34:

Overview of changes leading to 0.9.34
Saturday, August 2, 2014
=====================================

- hb_feature_from_string() now accepts CSS font-feature-settings format.
- As a result, hb-shape / hb-view --features also accept CSS-style strings.
  Eg, "'liga' off" is accepted now.
- Add old-spec Myanmar shaper:
  https://bugs.freedesktop.org/show_bug.cgi?id=81775
- Don't apply 'calt' in Hangul shaper.
- Fix mark advance zeroing for Hebrew shaper:
  https://bugs.freedesktop.org/show_bug.cgi?id=76767
- Implement Windows-1256 custom Arabic shaping.  Only built on Windows,
  and requires help from get_glyph().  Used by Firefox.
  https://bugzilla.mozilla.org/show_bug.cgi?id=1045139
- Disable 'liga' in vertical text.
- Build fixes.
- API changes:

  * Make HB_BUFFER_FLAG_BOT/EOT easier to use.

    Previously, we expected users to provide BOT/EOT flags when the
    text *segment* was at paragraph boundaries.  This meant that for
    clients that provide full paragraph to HarfBuzz (eg. Pango), they
    had code like this:

      hb_buffer_set_flags (hb_buffer,
                           (item_offset == 0 ? HB_BUFFER_FLAG_BOT : 0) |
                           (item_offset + item_length == paragraph_length ?
                            HB_BUFFER_FLAG_EOT : 0));

      hb_buffer_add_utf8 (hb_buffer,
                          paragraph_text, paragraph_length,
                          item_offset, item_length);

    After this change such clients can simply say:

      hb_buffer_set_flags (hb_buffer,
                           HB_BUFFER_FLAG_BOT | HB_BUFFER_FLAG_EOT);

      hb_buffer_add_utf8 (hb_buffer,
                          paragraph_text, paragraph_length,
                          item_offset, item_length);

    Ie, HarfBuzz itself checks whether the segment is at the beginning/end
    of the paragraph.  Clients that only pass item-at-a-time to HarfBuzz
    continue not setting any flags whatsoever.

    Another way to put it is: if there's pre-context text in the buffer,
    HarfBuzz ignores the BOT flag.  If there's post-context, it ignores
    EOT flag.

Log message:
Update to 0.9.33:

Overview of changes leading to 0.9.33
Tuesday, July 22, 2014
=====================================

- Turn off ARabic 'cswh' feature that was accidentally turned on.
- Add HB_TAG_MAX_SIGNED.
- Make hb_face_make_immutable() really make face immutable!
- Windows build fixes.

Log message:
Update to 0.9.32:

Overview of changes leading to 0.9.32
Thursday, July 17, 2014
=====================================

- Apply Arabic shaping features in spec order exactly.
- Another fix for Mongolian free variation selectors.
- For non-Arabic scripts in Arabic shaper apply 'rlig' and 'calt'
  together.
- Minor adjustment to U+FFFD logic.
- Fix hb-coretext build.

Overview of changes leading to 0.9.31
Wednesday, July 16, 2014
=====================================

- Only accept valid UTF-8/16/32; we missed many cases before.
- Better shaping of invalid UTF-8/16/32.  Falls back to
  U+FFFD REPLACEMENT CHARACTER now.
- With all changes in this release, the buffer will contain fully
  valid Unicode after hb_buffer_add_utf8/16/32 no matter how
  broken the input is.  This can be overriden though.  See below.
- Fix Mongolian Variation Selectors for fonts without GDEF.
- Fix minor invalid buffer access.
- Accept zh-Hant and zh-Hans language tags.  hb_ot_tag_to_language()
  now uses these instead of private tags.
- Build fixes.
- New API:
  * hb_buffer_add_codepoints().  This does what hb_buffer_add_utf32()
    used to do, ie. no validity check on the input at all.  add_utf32
    now replaces invalid Unicode codepoints with the replacement
    character (see below).
  * hb_buffer_set_replacement_codepoint()
  * hb_buffer_get_replacement_codepoint()
    Previously, in hb_buffer_add_utf8 and hb_buffer_add_utf16, when
    we detected broken input, we replaced that with (hb_codepoint_t)-1.
    This has changed to use U+FFFD now, but can be changed using these
    new API.

Overview of changes leading to 0.9.30
Wednesday, July 9, 2014
=====================================

- Update to Unicode 7.0.0:
  * New scripts Manichaean and Psalter Pahlavi are shaped using
    Arabic shaper.
  * All the other new scripts to through the generic shaper for
    now.
- Minor Indic improvements.
- Fix graphite2 backend cluster mapping [crasher!]
- API changes:
  * New HB_SCRIPT_* values for Unicode 7.0 scripts.
  * New function hb_ot_layout_language_get_required_feature().
- Build fixes.

Log message:
Update to 0.9.29:

Overview of changes leading to 0.9.29
Thursday, May 29, 2014
=====================================

- Implement cmap in hb-ot-font.h.  No variation-selectors yet.
- Myanmar: Allow MedialYa+Asat.
- Various Indic fixes:
  * Support most characters in Extended Devanagary and Vedic
    Unicode blocks.
  * Allow digits and a some punctuation as consonant placeholders.
- Build fixes.

Log message:
Don't use buggy CoreText in MacOS X 10.5

CTLineCreateWithAttributedString() seemingly fails when trying to use a
font which isn't installed into the system standard path.

Log message:
Fix build under Mac OS X.

Log message:
Recursive revbump from x11/pixman
Fix PR pkg/48777

Log message:
Update to 0.9.28:

Overview of changes leading to 0.9.28
Monday, April 28, 2014
=====================================

- Unbreak old-spec Indic shaping. (bug 76705)
- Fix shaping of U+17DD and U+0FC6.
- Add HB_NO_MERGE_CLUSTERS build option.  NOT to be enabled by default
  for shipping libraries.  It's an option for further experimentation
  right now.  When we are sure how to do it properly, we will add
  public run-time API for the functionality.
- Build fixes.

Log message:
recursive bump from icu shlib major bump.

Log message:
Update to 0.9.27:

Overview of changes leading to 0.9.27
Tuesday, March 18, 2014
=====================================

- Don't use "register" storage class specifier
- Wrap definition of free_langs() with HAVE_ATEXIT
- Add coretext_aat shaper and hb_coretext_face_create() constructor
- If HAVE_ICU_BUILTIN is defined, use hb-icu Unicode callbacks
- Add Myanmar test case from OpenType Myanmar spec
- Only do fallback Hebrew composition if no GPOS 'mark' available
- Allow bootstrapping without gtk-doc
- Use AM_MISSING_PROG for ragel and git
- Typo in ucdn's Makefile.am
- Improve MemoryBarrier() implementation