For GPOS pair positioning format 1 the description of ValueRecords
in the OpenType specification (1.8.2, from today) is wrong – the
offset has to be taken from the parent structure; in this case the
`PairSet' table.
* src/otvalid/otvgpos.c (otv_PairSet_validate): Set `extra3'.
(otv_PairPos_validate): Adjust.
This is just very basic stuff and just a little bit tested on
GNU/Linux only. I won't delve into this since I'm not a Jam user.
* Jamfile: Call `HDRMACRO' for `ftserv.h' also.
(DEFINES): Replace with...
(CCFLAGS): ... this.
* src/Jamfile: Don't call `HDRMACRO' for `internal.h'; this is
already handled in the top-level Jamfile.
* src/autofit/Jamfile (DEFINES): Replace with...
(CCFLAGS): ... this.
(_sources): Add missing files.
* src/cache/Jamfile: Don't call `HDRMACRO' for `ftcache.h'; it no
longer contains macro header definitions.
* src/base/Jamfile, src/cff/Jamfile, src/sfnt/Jamfile,
src/truetype/Jamfile (_sources): Add missing files.
Assure that FreeType's internal include directories are found before
`CPPFLAGS' (which might be set by the user in the environment), and
`CPPFLAGS' before `CFLAGS'.
* builds/freetype.mk (FT_CFLAGS): Don't add `INCLUDE_FLAGS'.
(FT_COMPILE): Make this a special variable for compiling only the
files handled in `freetype.mk'.
(.c.$O): Removed, unused.
* src/*/rules.mk (*_COMPILE): Fix order of include directories.
Some prototypes in ftvalid.h use `valid' for the variables
typed as FT_Validator. Their implementations in src/base/
ftobjs.c and utilizations in src/sfnt/ttcmap.c do similar.
Some macros in otvcommn.h assume the exist of the variable
`valid' typed as OTV_Validator in the caller.
Mixing these two conventions cause invalid pointer conversion
and unexpected SEGV in longjmp. To prevent it, all variables
typed as OTV_Validator are renamed to `otvalid'.
* src/otvalid/otvcommn.h: Replace `valid' by `otvalid'.
* src/otvalid/{otvcommn.c, otvbase.c, otvgdef.c, otvgpos.c,
otvgsub.c, otvjstf.c, otvmath.c}: Replace `valid' by `otvalid'
if it is typed as OTV_Validator.
This is essentially a mechanical conversion, adding inclusion of
`FT_INTERNAL_DEBUG_H' where necessary, and providing the macros for
stand-alone compiling modes of the rasterizer modules.
To convert the remaining occurrences of FT_Err_XXX and friends it is
necessary to rewrite the code. Note, however, that it doesn't harm
if some cases are not handled since FT_THROW is a no-op.
The idea is to always have FT_ERR_PREFIX available internally.
* include/freetype/fterrors.h: Use FT2_BUILD_LIBRARY to guard
undefinition of FT_ERR_PREFIX
* src/gxvalid/gxverror.h, src/otvalid/otverror.h,
src/sfnt/sferrors.h: Updated.
extern const FT_Module_Class
(or similar for C++). However, the actual types of the variables
being declared are often different, e.g., FT_Driver_ClassRec or
FT_Renderer_Class. (Some are, indeed, FT_Module_Class.)
This works with most C compilers (since those structs begin with an
FT_Module_Class struct), but technically it's undefined behavior.
To quote the ISO/IEC 9899:TC2 final committee draft, section 6.2.7
paragraph 2:
All declarations that refer to the same object or function shall
have compatible type; otherwise, the behavior is undefined.
(And they are not compatible types.)
Most C compilers don't reject (or even detect!) code which has this
issue, but the GCC LTO development branch compiler does. (It
outputs the types of the objects while generating .o files, along
with a bunch of other information, then compares them when doing the
final link-time code generation pass.)
Patch from Savannah bug #25133.
* src/base/ftinit.c (FT_USE_MODULE): Include variable type.
* builds/amiga/include/freetype/config/ftmodule.h,
include/freetype/config/ftmodule.h, */module.mk: Updated to declare
pass correct types to FT_USE_MODULE.
targets).
* src/base/ftobjs.c (FT_Open_Face): Don't define a variable in block
which is crossed by a `goto'.
* src/otvalid/otvalid.h (otv_MATH_validate): Add prototype.
(otv_Coverage_validate): Add fourth argument to pass an expected
count value. Update all users.
Check glyph IDs.
(otv_ClassDef_validate): Check `StartGlyph'.
* src/otvalid/otvgsub.c (otv_SingleSubst_validate): More glyph ID
checks.
* src/otvalid/otvmath.c (otv_MathConstants_validate): There are only
56 constants.
(otv_GlyphAssembly_validate, otv_MathGlyphConstruction_validate):
Check glyph IDs.
Fix for previous commit, which caused many compiler warnings/errors
about addresses of volatile objects passed as function arguments
as non-volatile pointers.
* freetype2/include/freetype/internal/ftvalid.h: Make
FT_Validator typedef a pointer to a volatile object.
* freetype2/src/gxvalid/gxvmod.c (gxv_load_table): Make function
argument `table' a pointer to a volatile object.
* freetype2/src/otvalid/otvmod.c (otv_load_table): Make function
argument `table' a pointer to a volatile object.
* freetype2/src/gxvalid/gxvmod.c (GXV_TABLE_DECL): Mark local
variable `_sfnt' as volatile since it must keep its value across
a call to ft_setjmp.
(gxv_validate): Same for local variables `memory' and `valid'.
(classic_kern_validate): Same for local variables `memory',
`ckern', and `valid'.
* freetype2/src/otvalid/otvmod.c (otv_validate): Same for function
parameter `face' and local variables `base', `gdef', `gpos', `gsub',
`jstf', and 'valid'.
* freetype2/src/sfnt/ttcmap.c (tt_face_build_cmaps): Same for
local variable `cmap'.
