The main problems
-----------------
o If FT_STROKER_LINEJOIN_BEVEL was specified, unlimited miter
joins (not bevel joins) were generated. Indeed, the meanings of
`miter' and `bevel' were incorrectly reversed (consistently) in
both the code and comments.
o The way bevel joins were constructed (whether specified
explicitly, or created as a result of exceeding the miter limit)
did not match what is required for stroked text in PostScript or
PDF.
The main fixes
--------------
o The behaviour of FT_STROKER_LINEJOIN_BEVEL has been corrected.
o A new line join style, FT_STROKER_LINEJOIN_MITER_FIXED, has been
introduced to support PostScript and PDF miter joins.
o FT_STROKER_LINEJOIN_MITER_VARIABLE has been introduced as an
alias for FT_STROKER_LINEJOIN_MITER.
Additionally, a variety of stroking errors have been fixed. These
would cause various artifacts (including points `at infinity'),
especially when stroking poor quality fonts.
See
http://lists.gnu.org/archive/html/freetype-devel/2011-07/msg00001.html
for example documents. The FreeType stroker now produces results
very similar to that produced by GhostScript and Distiller for these
fonts.
Other problems
--------------
The following problems have been resolved:
o Inside corners could be generated incorrectly. Intersecting the
inside corner could cause a missing triangular area and other
effects.
The intersection point can only be used if the join is between
two lines and both lines are long enough. The `optimization'
condition in `ft_stroker_inside' has been corrected; this
requires the line length to be passed into various functions and
stored in `FT_StrokerRec'.
o Incorrect cubic curves could be generated. The angle
calculations in `FT_Stroker_CubicTo' have been corrected to
handle the case of the curve crossing the +/-PI direction.
o If the border radius was greater than the radius of curvature of
a curve, then the negative sector would end up outside (not
inside) the border. This situation is now recognized and the
negative sector is circumnavigated in the opposite direction.
(If round line joins are being used, this code is disabled
because the line join will always cover the negative sector.)
o When a curve is split, the arcs may not join smoothly (especially
if the curve turns sharply back on itself). Changes in
direction between adjacent arcs were not handled. A round
corner is now added if the deviation from one arc to the next is
greater than a suitable threshold.
o The current direction wasn't retained if a the outline contained
a zero length lineto or a curve that was determined to be
`basically a point'. This could cause a spurious join to be
added.
o Cubics with close control points could be mishandled. All eight
cases are now distinguished correctly.
Other improvements
------------------
o Borders for cubic curves could be too `flat'.
FT_SMALL_CUBIC_THRESHOLD has been reduced a little to prevent
this.
o The handling and use of movable points has been simplified a
little.
o Various values are now computed only if the results are actually
needed.
o The directions of the outer and inner borders have been swapped,
as recommended by Graham Asher.
* src/base/ftstroke.c: Revised.
* include/freetype/ftstroke.h: Updated.
Passing uninitialized pointer to the buffer allocator is
not problematic theoretically (as far as the returned
pointer is checked before writing), but g++4.6 dislikes
it and warns by -Wuninitialized. Initialize them by NULL.
* src/base/ftobjs.c (FT_Stream_New): Init `stream'.
(new_memory_stream): Ditto.
(FT_New_GlyphSlot): Init `slot'.
(FT_CMap_New): Init `cmap'.
(open_face_PS_from_sfnt_stream): Init `sfnt_ps'.
(Mac_Read_POST_Resource): Init `pfb_data'.
(Mac_Read_sfnt_Resource): Init `sfnt_data'.
* src/base/ftrfork.c (FT_Raccess_Get_DataOffsets):
Init `offsets_internal' and `ref'.
(raccess_guess_darwin_hfsplus): Init `newpath'.
(raccess_guess_darwin_newvfs): Ditto.
* src/base/ftbitmap.c (ft_bitmap_assure_buffer):
Init `buffer'.
* src/base/ftstroke.c (FT_Stroker_New): Init `stroker'.
Some invalid, overrunning, unrecommended non-zero values
are cared in paranoid validation mode only. There are
many lines looking like:
if ( valid->root->level >= FT_VALIDATE_PARANOID )
FT_INVALID_xxx;
To simplify them, GXV_SET_ERR_IF_PARANOID( err ) is
introduced for more paranoid validation in future.
* src/gxvalid/gxvcommn.h (IS_PARANOID_VALIDATION):
New macro to assure valid->root->level is more or
equal to FT_VALIDATE_PARANOID. (GXV_SET_ERR_IF_PARANOID):
New macro to raise an error if in paranoid validation.
* src/gxvalid/gxvcommn.c: Use GXV_SET_ERR_IF_PARANOID().
* src/gxvalid/gxvfeat.c: Ditto.
* src/gxvalid/gxvjust.c: Ditto.
* src/gxvalid/gxvkern.c: Ditto.
* src/gxvalid/gxvmort.c: Ditto.
* src/gxvalid/gxvmort0.c: Ditto.
* src/gxvalid/gxvmort1.c: Ditto.
* src/gxvalid/gxvmort2.c: Ditto.
* src/gxvalid/gxvmorx1.c: Ditto.
* src/gxvalid/gxvmorx2.c: Ditto.
* src/gxvalid/gxvmort.c (gxv_mort_subtables_validate):
Conditionalize unvalidated variable `subFeatureFlags'.
(gxv_mort_chain_validate): Conditionalize unvalidated
variable `defaultFlags'.
* src/gxvalid/gxmort0.c
(gxv_mort_subtable_type0_entry_validate): Check the
conflict of the marks for the glyphs.
* src/gxvalid/gxmort1.c
(gxv_mort_subtable_type1_offset_to_subst_validate):
Local variables `min_gid', `max_gid' are replaced by
variables in the validator.
(gxv_mort_subtable_type1_entry_validate): Conditionalize
unvalidated variables; `setMark', `dontAdvance'.
(gxv_mort_subtable_type1_substTable_validate):
Validate the GID by the min/max GIDs in the validator.
* src/gxvalid/gxvmort2.c
(gxv_mort_subtable_type2_ligActionOffset_validate):
Conditionalize unvalidated variables; `last', `store'.
Checking for overrunning offset is added.
(gxv_mort_subtable_type2_entry_validate):
Conditionalize unvalidated variables; `setComponent',
`dontAdvance'.
(gxv_mort_subtable_type2_ligatureTable_validate):
Check if the GID for ligature does not exceed the
max GID in `maxp' table.
* src/gxvalid/gxvmort5.c
(gxv_mort_subtable_type5_InsertList_validate):
Conditionalize unvalidated loading of `insert_glyphID'
array. (gxv_mort_subtable_type5_entry_validate):
Conditionalize unvalidated variables; `setMark',
`dontAdvance', `currentIsKashidaLike',
`markedIsKashidaLike', `currentInsertBefore',
`markedInsertBefore'.
* src/gxvalid/gxvjust.c (gxv_just_check_max_gid):
New function to unify the checks of too large GID.
(gxv_just_wdp_entry_validate): Conditionalize unvalidated
variables; `beforeGrowLimit', `beforeShrinkGrowLimit',
`afterGrowLimit', `afterShrinkGrowLimit', `growFlags',
`shrinkFlags'. Additional check for non-zero values in
unused storage `justClass' is added.
(gxv_just_actSubrecord_type0_validate): Conditionalize
unvalidated variable `order'. GID is checked by
gxv_just_check_max_gid(). Additional check for upside-down
relationship between `lowerLimit' and `upperLimit' is added.
(gxv_just_actSubrecord_type1_validate): GID is checked by
gxv_just_check_max_gid().
(gxv_just_actSubrecord_type2_validate): Conditionalize
unvalidated variable `substThreshhold'. GID is checked by
gxv_just_check_max_gid().
(gxv_just_actSubrecord_type5_validate): GID is checked by
gxv_just_check_max_gid().
(gxv_just_classTable_entry_validate): Conditionalize
unvalidated variables; `setMark', `dontAdvance',
`markClass', `currentClass'.
* src/gxvalid/gxvcommn.h (GXV_LOAD_TRACE_VARS): New macro to
conditionalize the variable which is only used for trace messages.
Automatically set by FT_DEBUG_LEVEL_TRACE.
(GXV_LOAD_UNUSED_VARS): New macro to conditionalize the loading of
unvalidated variables. Undefined by default to calm gcc4.6 warning.
(GXV_ValidatorRec.{min_gid,max_gid}): New variables to hold defined
GID ranges, for the comparison of GID ranges in different subtables.
* src/autofit/afcjk.c (af_cjk_metrics_init_blues): Use casts and
remove unused variables.
* src/autofit/aflatin.c (af_latin_hints_compute_edges): Comment out
`up_dir'.
* src/smooth/ftsmooth.c (ft_smooth_render_generic): Use `height_org'
and `width_org' conditionalized.
The native SDK on earliest Mac OS X (10.0-10.1) did not have
"AvailabilityMacros.h". To prevent the inclusion of missing
header file, ECANCELED (introduced in 10.2) in POSIX header
file <errno.h> is checked to detect the system version.
* include/freetype/config/ftconfig.h: Conditionalize the
inclusion of "AvailabilityMacros.h".
* builds/unix/ftconfig.in: Ditto.
* builds/vms/ftconfig.h: Ditto.
* builds/unix/install-sh: Removed. It is not needed to
include repository, because autogen.sh installs it.
* builds/unix/.gitignore: Register install-sh.
To remove extremas of vertical strokes of CJK Ideographs at
low resolution and make the top and bottom horizontal stems
aligned, bluezones for CJK Ideographs are calculated from
sample glyphs. At present, vertical bluezones (bluezones
to align vertical stems) are disabled by default. For detail, see
http://lists.gnu.org/archive/html/freetype-devel/2011-04/msg00070.htmlhttp://lists.gnu.org/archive/html/freetype-devel/2011-04/msg00092.htmlhttp://lists.gnu.org/archive/html/freetype-devel/2011-05/msg00001.html
* include/freetype/internal/fttrace.h: New trace component `afcjk'.
* src/autofit/afcjk.h (AF_CJK{Blue,Axis,Metric}Rec): Add CJK version
for AF_Latin{Blue,Axis,Metric}Rec.
(af_cjk_metrics_check_digits): Ditto, shared with Indic module.
(af_cjk_metrics_init_widths): Ditto.
(af_cjk_metrics_init): Take AF_CJKMetric instead of AF_LatinMetric.
(af_cjk_metrics_scale): Ditto (declaration).
(af_cjk_hints_init): Ditto (declaration).
(af_cjk_hints_apply): Ditto (declaration).
* src/autofit/afcjk.c (af_cjk_metrics_scale): Ditto (body).
(af_cjk_hints_init): Ditto (body).
(af_cjk_hints_apply): Ditto (body).
(af_cjk_metrics_init_widths): Duplicate af_latin_metrics_init_widths.
(af_cjk_metrics_check_digits): Duplicate af_latin_metrics_check_digits.
(af_cjk_metrics_init): Call CJK bluezone initializer.
(af_cjk_metrics_scale_dim): Add code to scale bluezones.
(af_cjk_hints_compute_blue_edges): New function, CJK version of
af_latin_hints_compute_blue_edges.
(af_cjk_metrics_init_blues): New function, CJK version of
af_latin_metrics_init_blues.
(af_cjk_hints_edges): Add code to align the edge stems to blue zones.
* src/autofit/afindic.c (af_indic_metrics_init): Take AF_CJKMetric
instead of AF_LatinMetric, and initialize as af_cjk_metrics_init.
However bluezones are not initialized.
(af_indic_metrics_scale): Take AF_CJKMetric instead of AF_LatinMetric.
(af_indic_hints_init): Ditto.
(af_indic_hints_apply): Ditto.
* docs/CHANGES: Note about CJK bluezone support.
* builds/unix/configure.raw: New option `--disable-mmap'
is added. It is for the developers to simulate the systems
without mmap() (like 4.3BSD, minix etc) on POSIX systems.
* src/truetype/ttobjs.c (tt_check_trickyness_sfnt_ids):
Add 8 checksum sets for NEC FA family. For the tricky fonts
without some tables (e.g. NEC FA fonts lack cvt table),
extra check is added to assure that 0-length table in the
registry is not included in the font.
Some PDF generators mangle the family name badly, prioritize
the check by the sfnt table checksums than the check by the
family name. For sample PDF, see
http://lists.gnu.org/archive/html/freetype-devel/2011-04/msg00073.html
* src/truetype/ttobjs.c (tt_check_trickyness): Exchange the order
of tt_check_trickyness_family() and tt_check_trickyness_sfnt_ids().
This makes FT_LOAD_IGNORE_GLOBAL_ADVANCE_WIDTH redundant,
deprecated, and ignored. The new behavior is what every major user
of FreeType has been requesting. Global advance is broken in many
CJK fonts. Just ignoring it by default makes most sense.
* src/truetype/ttdriver.c (tt_get_advances),
src/truetype/ttgload.c (TT_Get_HMetrics, TT_Get_VMetrics,
tt_get_metrics, compute_glyph_metrics, TT_Load_Glyph),
src/truetype/ttgload.h: Implement it.
* docs/CHANGES: Updated.
Previously, signed integers were converted to unsigned integers, but
this can fail because of sign extension. For example, 0xa344a1eb
becomes 0xffffffffa344a1eb.
We now do the reverse which is always correct because the integer
size is the same during the cast from unsigned to signed.
* include/freetype/internal/ftstream.h, src/base/ftstream.c
(FT_Stream_Get*): Replace with...
(FT_Stream_GetU*): Functions which read unsigned integers.
Update all macros accordingly.
* src/gzip/ftgzip.c (ft_gzip_get_uncompressed_size): Updated.
This isn't the optimal solution yet, but it restores the previous
rendering quality (more or less).
* src/smooth/ftgrays.c (gray_render_conic): Do more splitting.
* include/freetype/config/ftconfig.h: Ported the FT_MulFix_i386
function from GNU inline assembly syntax (see #ifdef __GNUC__ block
above) to MASM syntax for Microsoft Visual C++.
Patch suggested by Miles.Lau <sunliang_liu@foxitsoftware.com>.
* src/cff/cffgload.c (cff_decoder_parse_charstrings): Emit tracing
message for unknown operators and continue instead of exiting with a
syntax error.
* src/truetype/interp.c: Set even more TT_Err_Invalid_Reference
error codes only if pedantic hinting is active. At the same time,
try to provide sane values which hopefully allow useful
continuation. Exception to this is CALL and LOOPCALL – due to
possible stack corruption it is necessary to bail out.
* src/truetype/ttinterp.c (TT_RunIns, Ins_FLIPPT, Ins_DELTAP,
Ins_DELTAC): Exit with error only if `pedantic_hinting' is set.
Otherwise, try to do something sane.
Thanks to Greg Hitchcock who explained the issue.
* src/truetype/ttinterp.c (Ins_MIRP): Replace a `>=' operator with
`>' since the description in the specification is incorrect.
This fixes, for example, glyph `two' in font `Helvetica Neue LT Com
65 medium' at 15ppem.
* include/freetype/config/ftconfig.h (FT_MulFix_arm):
Copy the maintained code from builds/unix/ftconfig.in.
Old GNU binutils could not accept the reduced syntax
`orr %0, %2, lsl #16'. Un-omitted syntax like RVCT,
`orr %0, %0, %2, lsl #16' is better. Reported by
Johnson Y. Yan. The bug report by Qt developers is
considered too.
http://bugreports.qt.nokia.com/browse/QTBUG-6521
Right before B/W rasterizing, the bbox gets simply rounded to
integers. This fixes, for example, glyph `three' in font `Helvetica
Neue LT Com 65 Medium' at 11ppem.
Thanks to Greg Hitchcock who explained this behaviour.
* src/raster/ftrend1.c (ft_raster1_render): Implement it.
* builds/unix/configure.raw: Consider recent gcc-standard
flags to specify architecture in CFLAGS & LDFLAGS
harmonization. Requested by Savannah bug #32114, to
support multilib feature of BuildRoot SDK correctly.
* builds/unix/configure.raw: Some important options that
included in CFLAGS but not in LDFLAGS are copied to
LDFLAGS, but the last option in CFLAGS was not checked.
This catches a rendering problem with glyph `x' from Tahoma at
10ppem. It seems that the increase of the precision in the change
from 2009-06-11 makes a larger jitter value unnecessary.
* src/raster/ftraster.c (Set_High_Precision): Implement it.
If a drop-out rule would switch on a pixel outside of the glyph's
bounding box, use the right (or top) pixel instead. This is an
undocumented feature, but some fonts like `Helvetica Neue LT Com 65
Medium' heavily rely on it.
Thanks to Greg Hitchcock who explained this behaviour.
* src/raster/ftraster.c (Vertical_Sweep_Drop,
Horizontal_Sweep_Drop): Implement it.
When a node comparator changes the cached nodes during the
search of a node matching with queried properties, the
pointers obtained before the functon should be updated to
prevent the dereference to freed or reallocated nodes.
To minimize the rescan of the linked list, the update is
executed when the comparator notifies the change of cached
nodes. This change depends previous change:
38b272ffbb
* src/cache/ftccache.h (FTC_CACHE_LOOKUP_CMP): Rescan the
top node if the cached nodes are changed.
* src/cache/ftccache.c (FTC_Cache_Lookup): Ditto.
Some node comparators (comparing the cache node content and
the properties specified by the query) can flush the cache
node to prevent the cache inflation. The change may
invalidate the pointers to the node obtained before the node
comparison, so the change should be noticed to the caller.
The problem caused by the cache node changing is reported by
Harsha, see Savannah bug #31923.
* src/cache/ftccache.h (FTC_Node_CompareFunc): Add new
argument `FT_Bool* list_changed' to indicate the change of
the cached nodes to the caller.
(FTC_CACHE_LOOKUP_CMP): Watch the change of the cached nodes
by `_list_changed'.
(FTC_CACHE_TRYLOOP_END): Take new macro argument `_list_changed'
and update it when FTC_Manager_FlushN() flushes any nodes.
* src/cache/ftccback.h (ftc_snode_compare): Updated to fit
with new FTC_Node_CompareFunc type. (ftc_gnode_compare): Ditto.
* src/cache/ftcbasic.c: Include FT_INTERNAL_OBJECTS_H to
use TRUE/FALSE macros. (ftc_basic_gnode_compare_faceid):
New argument `FT_Bool* list_changed' to indicate the change
of the cache nodes, anyway, it is always FALSE.
* src/cache/ftccmap.c: Include FT_INTERNAL_OBJECTS_H to
use TRUE/FALSE macros. (ftc_cmap_node_compare):
New argument `FT_Bool* list_changed' to indicate the change
of the cache nodes, anyway, it is always FALSE.
(ftc_cmap_node_remove_faceid): Ditto.
* src/cache/ftccache.c (FTC_Cache_NewNode): Pass a NULL
pointer to FTC_CACHE_TRYLOOP_END(), because the result is
not needed. (FTC_Cache_Lookup): Watch the change of the cache
nodes by `list_changed'. (FTC_Cache_RemoveFaceID): Ditto.
* src/cache/ftcglyph.c: Include FT_INTERNAL_OBJECTS_H to
use TRUE/FALSE macros. (ftc_gnode_compare): New argument
`FT_Bool* list_changed' to indicate the change of the cache
nodes, anyway, it is always FALSE. (FTC_GNode_Compare):
New argument `FT_Bool* list_changed' to be passed to
ftc_gnode_compare().
* src/cache/ftcglyph.h (FTC_GNode_Compare): Ditto.
* src/cache/ftcsbits.c (ftc_snode_compare): New argument
`FT_Bool* list_changed' to indicate the change of the cache
nodes, anyway. It is updated by FTC_CACHE_TRYLOOP().
(FTC_SNode_Compare): New argument `FT_Bool* list_changed'
to be passed to ftc_snode_compare().
* src/cache/ftcsbits.h (FTC_SNode_Compare): Ditto.
* src/cache/ftcglyph.h (FTC_GNode_Compare): Add the 3rd
argument `FTC_Cache cache' to fit FTC_Node_CompareFunc
prototype.
* src/cache/ftcglyph.c (FTC_GNode_Compare): Ditto. Anyway,
`cache' is not used by its child ftc_gnode_compare().
There are several duplicated codes getting the top node
from a cache by a given hash, like:
idx = hash & cache->mask;
if ( idx < cache->p )
idx = hash & ( cache->mask * 2 + 1 );
pnode = cache->buckets + idx;
To deduplicate them, a cpp-macro to do same work
FTC_NODE__TOP_FOR_HASH( cache, hash ) is introduced.
For non-inlined config, non-ftc_get_top_node_for_hash() is
also introduced.
* src/cache/ftccache.h (FTC_NODE__TOP_FOR_HASH): Declare
and implement inlined version.
(FTC_CACHE_LOOKUP_CMP): Use FTC_NODE__TOP_FOR_HASH().
* src/cache/ftccache.c (ftc_get_top_node_for_hash): Non-
inlined version.
(ftc_node_hash_unlink): Use FTC_NODE__TOP_FOR_HASH().
(ftc_node_hash_link): Ditto.
(FTC_Cache_Lookup): Ditto.
* src/cache/ftcglyph.c (FTC_GNode_Compare): Conditionalized
for inlined config. This function is a thin wrapper of
ftc_gnode_compare() for inlined FTC_CACHE_LOOKUP_CMP()
(see `nodecmp' argument). Under non-inlined config,
ftc_gnode_compare() is invoked by FTC_Cache_Lookup(),
via FTC_Cache->clazz.node_compare().
* src/cache/ftcglyph.h (FTC_GNode_Compare): Ditto.
* src/cache/ftcsbits.c (FTC_SNode_Compare): Ditto,
for ftc_snode_compare().
* src/cache/ftcsbits.h (FTC_SNode_Compare): Ditto.
Just Fill Bugs