`switch' statement to make it run faster.
(ah_outline_compute_segments): Reset `segment->score' and
`segment->link'.
(ah_outline_link_segments): Provide alternative code which does
the same but runs much faster.
Handle major direction also.
(ah_outline_compute_edges): Scale `edge_distance_threshold' down
after rounding instead of scaling comparison value in loop.
* src/autohint/ahhint.c (ah_hinter_align_stong_points): Provide
alternative code which runs faster.
Handle `before->scale == 0'.
* src/autohint/ahtypes.h (AH_SegmentRec): Move some fields down.
(AH_EdgeRec): Move some fields in structure.
New field `scale'.
* src/sfnt/ttcmap0.c (tt_cmap4_char_next): Use binary search.
`freetype.in'.
* builds/unix/configure: Regenerated.
* builds/unix/install.mk (install, uninstall): Handle `freetype.pc'.
* builds/unix/freetype2.in: New file. Contains building information
for the `pkg-config' package.
argument.
(bdf_get_font_property): Use `const' for third argument.
Updated all callers.
* src/bdf/bdfdrivr.c (BDF_Face_Init): Set pixel width and height
similar to the PCF driver.
* src/bdf/bdf.h (_hashnode): Use `const' for `key'.
Updated.
* src/gzip/ftgzip.c: C++ doesn't like that the array `inflate_mask'
is declared twice. It is perhaps better to modify the zlip source
files directly instead of this hack.
(zcalloc, zfree, ft_gzip_stream_close, ft_gzip_stream_io): Add casts
to make build with g++ successful.
* README.UNX: removed (now replaced by docs/INSTALL.UNX)
* src/pshinter/pshalgo3.c: the hinter now performs as in 2.1.3 and
will ignore stem quantization only when FT_LOAD_TARGET_SMOOTH is used
* src/base/ftobjs.c: changed the default computations to include rounding
in all cases, this is required to provide accurate kerning data when
native TrueType hinting is enabled.
* src/type1/t1load.c: the Type 1 loader now accepts more general names
according to the Postscript spec (the previous one was too restrictive)
src/autohint/ahhint.c: fixed blue-scale problem
* src/cache/ftccache.c: fixed small bug that could crash the cache
in rare circumstances (mostly with broken fonts)
* include/freetype/internal/ftobjs.h, src/base/ftobjs.c, src/bdf/bdfdrivr.c,
src/pcf/pcfdriver.c, src/pfr/pfrsbit.c, src/sfnt/ttsbit.c,
src/type42/t42objs.c, src/winfonts/winfnt.c: introduced three new functions
to deal with glyph bitmaps within FT_GlyphSlot objects. these are:
ft_glyphslot_free_bitmap
ft_glyphslot_alloc_bitmap
ft_glyphslot_set_bitmap
these are much more convenient to use than managing the FT_GLYPH_OWN_BITMAP
flag manually. the font drivers have been modified to use them as well.
* src/cache/ftlru.c: fixed an invalid assertion check
to control memory debugging with FreeType. See the description of
"FT2_DEBUG_MEMORY", "FT2_ALLOC_TOTAL_MAX" and "FT2_ALLOC_COUNT_MAX"
in DEBUG.TXT
* src/cache/ftccache.c, src/cache/ftccmap.c, src/cache/ftcsbits.c,
ftlru.c: fixed the cache sub-system to correctly deal with out-of-memory
conditions.
* src/pfr/pfrobjs.c, src/pfr/pfrsbits.c: fixing compiler warnings and a
small memory leak
* src/psaux/psobjs.c (t1_reallocate_table): fixed a bug (memory leak) that
only happened when trying to resize an array would end in an OOM.
* src/smooth/ftgrays.c: removed compiler warnings / volatile bug
* src/truetype/ttobjs.c: removed segmentation fault that happened in
tight memory environments.
when available in Asian fonts
* src/sfnt/ttload.c, src/base/ftobjs.c: changed "asian" to "Asian" in
comments
* src/truetype/ttdriver.c (Set_Char_Sizes): fixed a rounding bug when
computing the scale factors for a given character size in points with
resolution.
when trying to read certain compressed gzip files. The following test
could be used to reveal the bug:
touch 0123456789 ; gzip 0123456789 ; ftdump 0123456789.gz
* src/pfr/pfrobjs.c, src/pfr/pfrload.c, src/pfr/pfrtypes.h: several
fixes to the PFR font driver:
- the list of available embedded bitmaps was not correctly set
in the root FT_FaceRec structure describing the face
- the glyph loader always tried to load the outlines when
FT_LOAD_SBITS_ONLY was specified
- the table loaded now scans for *undocumented* elements of a
physical font's auxiliary data record, this is necessary to
retrieve the "real" family and style names.
NOTE THAT THIS CHANGES THE FAMILY NAME OF MANY PFR FONTS !!
Werner Lemberg