The optional 'COLR' v1 glyph-specific clip box helps upstream graphics
libraries allocate a sufficiently large bitmap for a glyph without having to
traverse the glyph graph for that. See
https://github.com/googlefonts/colr-gradients-spec/issues/251
for background on the introduction of this specification change.
* include/freetype/ftcolor.h (FT_ClipBox): New structure.
(FT_Get_Color_Glyph_ClipBox): New function declaration.
* include/freetype/internal/sfnt.h (TT_Get_Color_Glyph_ClipBox_Func):
New function type.
(SFNT_Interface, FT_DEFINE_SFNT_INTERFACE): Use it.
* src/base/ftobjs.c (FT_Get_Color_Glyph_ClipBox): New function to link API
with SFNT implementation.
* src/sfnt/sfdriver.c (sfnt_interface): Updated.
* src/sfnt/ttcolr.c (Colr): New field `clip_list`.
(tt_face_load_colr): Parse global clip list offset.
(tt_face_get_color_glyph_clipbox): New function to find the clip box for a
glyph id from the clip list array.
* src/sfnt/ttcolr.h: Updated.
* src/sdf/ftbsdf.c (first_pass, second_pass): Fix range during forward pass.
Otherwise the index goes out of range for the last column.
Fixes issue #1077.
Do not generate SDF from bitmap if the `FT_GLYPH_OWN_BITMAP` flag is not
set. In some cases the bitmap buffer is freed but still points to a valid
address; to handle those cases check the flag before accessing the memory.
* src/sdf/ftsdfrend.c (ft_bsdf_render): Handle the above case.
Also, return an error message if the bitmap's rows/pitch is invalid,
otherwise `slot->buffer` might be assigned to some invalid memory location.
(ft_sdf_render): Same as above.
Plus, move the outline back to original state after rasterization and not if
any error occurs.
Signed-off-by: Anuj Verma <anujv@iitbhilai.ac.in>
* src/sfnt/ttcolr.c (read_paint): PaintSweepGradient follows other
spec changes and now has the angles specified as F2DOT14, reflect
that in the implementation.
* include/freetype/ftcolor.h (FT_PaintSweepGradient): Update
documentation.
* src/sfnt/ttcolr.c (read_paint): Implement spec change where
affine transform matrix is now referenced by offset instead of
being placed inline in the PaintTransform table.
* src/sfnt/ttcolr.c (read_paint): Implement spec changes around
PaintSkew, PaintSkewAroundCenter. Update parsing to read shorter
values as changed in the spec.
* include/freetype/ftcolor.h (FT_PaintSkew): Update documentation.
* src/sfnt/ttcolr.c (read_paint): Implement spec change where
PaintRotate and PaintRotateAroundCenter were split for a more
compact format definition. Update parsing to read shorter values
as changed in the spec.
* include/freetype/ftcolor.h (FT_PaintRotate): Update documentation.
The header file `afwrtsys.h` has two distinct functions: to include the
required writing system headers and also to generate code for each writing
system. At each current use site only one or the other is used, with
various macro trickery selecting one or the other. Split this header into
`afws-decl.h` for the required writing system declarations and `afws-iter.h`
for iterating over the writing systems to generate code.
The motivation for this change is that the Visual C++ compiler treats the
standard include guard idiom like `#pragma once` 'if no non-comment code or
preprocessor directive comes before or after the standard form of the
idiom'. It appears to check this after macro expansion, so if
`WRITING_SYSTEM` expands to empty the bottom of `afwrtsys.h` is empty and
looks like the standard include guard idiom which is treated like `#pragma
once`, so subsequent inclusion of `afwrtsys.h` is elided.
Fixes#1075.
* src/autofit/afglobal.c (af_writing_system_classes), src/autofit/aftypes.h
(AF_WritingSystem), src/autofit/rules.mk (AUTOF_DRV_H): Updated.
* src/autofit/afwrtsys.h: Split into...
* src/autofit/afws-decl.h, src/autofit/afws-iter.h: New files.
==========================
Tag sources with `VER-2-11-0'.
* docs/VERSION.TXT: Add entry for version 2.11.0.
* docs/CHANGES: Updated.
* README, src/base/ftver.rc, builds/windows/vc2010/index.html,
builds/windows/visualc/index.html,
builds/windows/visualce/index.html,
builds/wince/vc2005-ce/index.html,
builds/wince/vc2008-ce/index.html, docs/freetype-config.1:
s/2.10.4/2.11.0/, s/2104/2110/.
* include/freetype/freetype.h (FREETYPE_MINOR): Set to 11.
(FREETYPE_PATCH): Set to 0.
* builds/unix/configure.raw (version_info): Set to 24:0:18.
* CMakeLists.txt (VERSION_MINOR): Set to 11.
(VERSION_PATCH): Set to 0.
* builds/toplevel.mk (dist): Ignore more git-related files.
The code assumed that if `__SSE2__` is defined, then 64-bit integer
types are available. This is not the case apparently for certain
multilib compiler targets like 'x86_32.x86' used by Gentoo.
This patch fixes the issue by disabling the special code path when
64-bit integer types are not available.
Fixes#1071.
* src/smooth/ftgrays.c (gray_render_conic): Move variable and
structure declarations to beginning of function. Inspite of C99
compliance we still do this for the sake of backward compatibility.
This also avoids a shadowing declaration of `count`.
(gray_convert_glyph_inner): Fix typo.
Guard inclusion of emmintrin.h with "#ifdef __SSE2__". The gcc version
of this header, xmmintrin.h, and mmintrin.h check that the appropriate
defines are set before defining anything (are internally guarded).
However the clang versions of these includes are not internally guarded.
As a result of this, externally guard the inclusion of these headers.
Benchmarking shows that this provides a very slighty performance
boost when rendering fonts with lots of quadratic bezier arcs,
compared to the recursive arc splitting, but only when SSE2 is
available, or on 64-bit CPUs.
On a 2017 Core i5-7300U CPU on Linux/x86_64:
./ftbench -p -s10 -t5 -cb .../DroidSansFallbackFull.ttf
Before: 4.033 us/op (best of 5 runs for all numbers)
After: 3.876 us/op
./ftbench -p -s60 -t5 -cb .../DroidSansFallbackFull.ttf
Before: 13.467 us/op
After: 13.385 us/op
This speeds up the smooth rasterizer by avoiding a
conditional branches in the hot path. Namely:
- Define a fixed "null cell" which will be pointed
to whenever the current cell is outside of the current
target region. This avoids a "ras.cell != NULL"
check in the FT_INTEGRATE() macro.
- Also use the null cell as a sentinel at the end of
all ycells[] linked-lists, by setting its x coordinate
to INT_MAX. This avoids a 'if (!cell)' check in
gray_set_cell() as well.
- Slightly change the worker struct fields to perform
a little less operations during rendering.
Example results (on a 2013 Corei5-3337U CPU)
out/ftbench -p -s10 -t5 -bc /usr/share/fonts/truetype/droid/DroidSansFallbackFull.ttf
Before: 5.472 us/op
After: 5.275 us/op
out/ftbench -p -s60 -t5 -bc /usr/share/fonts/truetype/droid/DroidSansFallbackFull.ttf
Before: 17.988 us/op
After: 17.389 us/op
This was already true (though undocumented) most of the time, but
not if `FT_NEW` inside `FT_Stream_New` failed or if the
`FT_OPEN_XXX` flags were bad.
Normally, `FT_Open_Face` calls `FT_Stream_New`, which returns the
user-supplied stream unchanged, and in case of any subsequent error
in `FT_Open_Face`, the stream is closed via `FT_Stream_Free`.
Up to now, however, `FT_Stream_New` allocates a new stream even if
it is already given one by the user. If this allocation fails, the
user-supplied stream is not returned to `FT_Open_Face` and never
closed. Moreover, the user cannot detect this situation: all they
see is that `FT_Open_Face` returns `FT_Err_Out_Of_Memory`, but that
can also happen after a different allocation fails within the main
body of `FT_Open_Face`, when the user's stream has already been
closed by `FT_Open_Face`. It is plausible that the user stream's
`close` method frees memory allocated for the stream object itself,
so the user cannot defensively free it upon `FT_Open_Face` failure
lest it ends up doubly freed. All in all, this ends up leaking the
memory/resources used by user's stream.
Furthermore, `FT_Stream_New` simply returns an error if the
`FT_OPEN_XXX` flags are unsupported, which can mean either an
invalid combination of flags or a perfectly innocent
`FT_OPEN_STREAM` on a FreeType build that lacks stream support.
With this patch, the user-supplied stream is closed even in these
cases, so the user can be sure that if `FT_Open_Face` failed, the
stream is definitely closed.
* src/base/ftobjs.c (FT_Stream_New): Don't allocate a buffer
unnecessarily.
Move error-handling code to make the control flow more obvious.
Close user-supplied stream if the flags are unsupported.
`FT_Stream_Open` always sets `pathname.pointer`, so remove the
redundant (re)assignment. None of the `FT_Stream_Open...` functions
uses `stream->memory`, so keep just one assignment at the end,
shared among all possible control flow paths.
('Unsupported flags' that may need a stream closure can be either an
invalid combination of multiple `FT_OPEN_XXX` mode flags or a clean
`FT_OPEN_STREAM` flag on a FreeType build that lacks stream
support.)
The three modes are mutually exclusive, and the documentation of the
`FT_OPEN_XXX` constants notes this. However, there was no check to
validate this in the code, and the documentation on `FT_Open_Args`
claimed that the corresponding bits were checked in a well-defined
order, implying it was valid (if useless) to specify more than one.
Ironically, this documented order did not agree with the actual
code, so it could not be relied upon; hopefully, nobody did this and
nobody will be hurt by the new validation.
Even if multiple mode bits were allowed, they could cause memory
leaks: if both `FT_OPEN_STREAM` and `stream` are set along with
either `FT_OPEN_MEMORY` or `FT_OPEN_PATHNAME`, then `FT_Stream_New`
allocated a new stream but `FT_Open_Face` marked it as an 'external'
stream, so the stream object was never released.
* src/base/ftobjs.c (FT_Stream_New): Reject incompatible
`FT_OPEN_XXX` flags.
Move the af_sort_xxx() functions from afangles.c to afhints.c
in order to get rid of the obsolete angle-related types,
macros and function definitions.
This feature was always experimental, and probably nevery worked
properly. This patch completely removes it from the source code,
except for a documentation block describing it for historical
purpose.
This code has always been experimental and was never compiled
anyway (FT_OPTION_AUTOFIT2 does not appear in ftoption.h or even
any of our build files).
It is undefined behavior to pass `NULL` to `memcpy`. `coords' is
passed to `memcpy` but `TT_Get_MM_Blend` and `TT_Get_Var_Design`
explictly call `tt_set_mm_blend` with `coords` as `NULL`. In
addition, `TT_Set_MM_Blend` has a similar possible issue.
'COLR' v1 fonts do not necessarily need to have a layer list; for
this reason, 'fontTools' recently started generating fonts in a way
that drops the layer list if there are no layers in it. This
results in the layer list offset becoming zero, which FreeType
treated as an invalid table. Fix that and handle the case for layer
list offset being 0. This slightly changes how we need to calculate
the starting offset for paints.
* src/sfnt/ttcolr.c (tt_face_load_colr): Handle case of layer list
offset being zero without outright rejecting table.
The second pass also fixes horizontal lines through the pixel centers.
* src/raster/ftraster.c (black_TWorker): Do not use `second_pass'.
(Render_Glyph): Skip the second pass only with the appropriate flag.
* src/raster/ftraster.c (black_TWorker): Move `band_stack' from here..
(Render_Single_Pass): ... to here and accept limit arguments.
(Render_Glyph): Updated.
* src/raster/ftraster.c (black_TWorker): Move `arcs' from here...
(Conic_To, Cubic_To): ... to here to tighten their scope.
(Bezier_Up, Bezier_Down): ... Take the current `arc' argument.
Paint tables can appear before the `base_glyphs_v1` offset if the
font is produced with the layer list before the base glyph list. In
this case paint tables can occur after the layer list but before the
base glyph list. Checks in the 'COLR' v1 code were rejecting fonts
with this layout. Improve these checks by calculating a minimum
offset after which paint tables can occur and use that in safety
checks.
* src/sfnt/ttcolr.c (Colr, tt_face_load_colr): Declare
`paint_start_v1` and calculate that as the minimum of the end of
layer list and base glyph list.
(get_child_table_pointer, read_paint, tt_face_get_paint_layers):
Use that in safety checks.
* src/raster/ftraster.c (black_TWorker): Replace the current line
offset with the pointer and drop the increment.
(Function_Sweep_Init): Take values as arguments instead of pointers.
(Vertical_Sweep_*, Horizontal_Sweep_Init, Draw_Sweep): Updated.
We can bisect a band until it is just a single scan line. This might
be slow and cause time-outs but if we need to impose limits it should
be elsewhere.
* src/raster/ftraster.c (Render_Single_Pass): Tweak sub-banding.
FT_Render_Glyph picked up FAILURE or 1 returned from the raster
function, which became a confusing error code. Instead, return
Raster_Overflow in the unlikely event that banding does not help or
another meaningful error.
* src/smooth/ftgrays.c (gray_convert_glyph_inner, gray_convert_glyph):
Use Raster_Overflow when the rendering pool is exhausted and return it
if banding does not help.
(gray_raster_render): Use Smooth_Err_Ok.
* src/raster/ftraster.c (Render_Single_Pass): Return Raster_Overflow
if banding does not help or another error code.
* include/freetype/ftcolor.h (FT_PaintTransformed, FT_PaintFormat,
FT_COLR_Paint): Do it to make it harmonize with other names such as
'PaintTranslate'.
* src/sfnt/ttcolr.c (read_paint, tt_face_get_paint): Ditto.
We need to inverse inheritance of FT_GlyphSlot_Internal so that we
have a chance to free the rendered bitmap from the parent slot.
* src/type42/t42objs.c (T42_GlyphSlot_Init): Remove the internal parts
of the child `ttslot' and replace it with the parent structure.
(T42_GlyphSlot_Done): Updated accordingly.
* src/psaux/psintrp.c (cf2_interpT2CharString)
<cf2_escCALLOTHERSUBR>: Convert assertion into error, since the
problem can happen with invalid user input.
Test case is file
fuzzing/corpora/legacy/oss-fuzz/5754332360212480-unknown-read
in the `freetype2-testing` repository.
* src/sfnt/ttcolr.c (tt_face_get_paint_layers): In addition to the
existing sanity checks, ensure that the pointer to the layer to be
read is within the 'COLR' v1 table.
* src/sdf/ftsdf.c: Include `ftbitmap.h`.
* src/sdf/ftsdfcommon.h: Move function bodies to `ftsdfcommon.c`.
Include `ftobjs.h` to get definitions of `FT_LOCAL` and friends.
* src/sdf/ftsdfcommon.c: New file.
* src/sdf/rules.mk, src/sdf/sdf.c: Updated.
Since 8-bits is enough to represent SDF data we no longer require
16-bits for this purpose. Also, we now normalize the output data
to use the entire 8-bit range efficiently. For example: if we use
3.5 format with a spread of 1 we basically only use the starting
5-bits. By normalizing we can use the entire 8-bit range.
* include/freetype/freetype.h (FT_Render_Mode): Updated description
for `FT_RENDER_MODE_SDF` regarding this change.
* include/freetype/ftimage.h (FT_Pixel_Mode): Removed
`FT_PIXEL_MODE_GRAY16` since no longer required.
* include/freetype/fttypes.h (FT_F6Dot10): Removed since no longer
required.
* src/sdf/ftsdfrend.c (ft_sdf_render, ft_bsdf_render): Allocate 8-bit
bitmap instead of 16-bit buffer.
* src/sdf/ftsdfcommon.h (map_fixed_to_sdf): Added function to convert
16.16 distance value to our desired format.
* src/sdf/ftsdf.c (sdf_generate_with_overlaps,
sdf_generate_bounding_box): Use the new `map_fixed_to_sdf` function
and also use 8-bit output buffer.
* src/sdf/ftbsdf.c (finalize_sdf): Output to a 8-bit buffer instead
of 16-bit buffer.
The autohinter could not access the base (unscaled) outline in the
child TrueType glyph slot. We now share the internal parts between
the parent and child glyph slots. Fixes#1057.
* src/type42/t42objs.c (T42_GlyphSlot_Init): Remove the internal parts
of `T42_GlyphSlot' and replace it with the child TrueType structure.
(T42_GlyphSlot_Done): Updated accordingly.
While a function pointer may be cast to another function pointer
type, it is required to cast the function pointer back to the
original function pointer type before calling it. If a parameter is
a pointer the exact pointer type is required. Using a pointer to a
different underlying type is technically undefined behavior. The
wrapper functions `ft_gzip_alloc` and `ft_gzip_free` took
`FT_Memory` (a `FT_MemoryRec_*`) instead of `voidpf` (`void*`), so
when gzip calls these callbacks through `alloc_func` or `free_func`
it invokes undefined behavior. On most platforms this works out as
expected, but newer undefined behavior detectors and targets like
wasm can detect this and will produce an error.
* src/gzip/ftgzip.c (ft_gzip_alloc, ft_gzip_free): Update signatures
to exactly match `alloc_func` and `free_func`, respectively.
Internally, cast the `void*` opaque pointer to `FT_Memory`.
Fixes timeout (#1055) analyzed by Ben Wagner, reported as
https://crbug.com/1194092
* src/sfnt/ttload.c (tt_face_load_post): Check POST format.
* src/sfnt/sfobjs.c (sfnt_load_face): Synthesize the missing unicode
charmap only if the glyph names exist.
* src/psnames/psmodule.c (ps_unicode_value): Short cut ".notdef" and
".null".
Fixes timeout reported as
https://crbug.com/1206181
* src/psaux/psarrst.c (cf2_arrstack_{push,init}): Implement it.
* src/psaux/psarrst.h (CF2_ArrStackiRec): Drop `chunk'.
Selecting the fill rule or checking the direct mode each time we call
`gray_hline' is sub-optimal. This effectively splits the direct mode
into a separate code path while inlining `gray_hline' and saving 5-7%
of rendering time.
* src/smooth/ftgrays.c (gray_hline): Eliminated in favor of...
(FT_FILL_RULE, FT_GRAY_SET): ... these new macros...
(gray_sweep): ... inlined here.
(gray_sweep_direct): New function that handles the direct span buffer.
(gray_TWorker): Remove the span buffer.
(gray_raster_render, gray_convert_glyph): Updated.
Add argument to return the actual number of bytes that were decoded.
The actual number of bytes decoded can be quite variable depending
on the number of ignored 'whitespace' bytes or early termination
with `>`.
(cid_face_open): Updated to use this calculated value. This avoids
trusting `parser->binary_length` is always be correct and reading
uninitialized bits if fewer are actually decoded.
First reported as
https://crbug.com/1203240
To reduce memory allocations, we read an entire Pascal-string buffer
and convert it to a C-string buffer. We also reject tables with
Postscript glyph names exceeding 63 bytes.
* src/sfnt/ttpost.c (load_format20): Implement it.
(load_post_names): Check the minimal POST table size.
(load_format25, tt_face_free_ps_names): Updated accordingly.
* src/sfnt/sfdriver.c (get_win_string, get_apple_string,
sfnt_get_var_ps_name): Do not zero out the buffer.
* src/sfnt/sfobjs.c (sfnt_init_face): Ditto.
* src/sfnt/sfwoff.c (woff_open_font): Ditto.
* src/sfnt/sfwoff2.c (woff2_open_font): Ditto.
* src/pcf/pcfread.c (pcf_interpret_style): Do not zero out the buffer.
* src/bdf/bdfdrivr.c (bdf_interpret_style): Ditto.
* src/winfonts/winfnt.c (FNT_Face_Init, FNT_Load_Glyph): Ditto.
* src/base/ftobjs.c (open_face_PS_from_sfnt_stream,
Mac_Read_sfnt_Resource): Do not zero out the buffer.
* src/base/ftmac.c (FT_New_Face_From_SFNT, read_lwfn): Ditto.
* src/base/ftrfork.c (raccess_make_file_name,
raccess_guess_darwin_hfsplus, raccess_guess_darwin_newvfs): Ditto.
* src/sfnt/ttcolr.c (tt_face_get_paint_layers): Do not output
layer pointer to iterator if it is outside the 'COLR' table.
(read_paint): Do not attempt to read layers that are outside the
table.
* src/base/ftobjs.c (ft_glyphslot_clear): This function is intended
to reset all the values of a glyph slot. However, it was not
resetting the values of the advances and `glyph_index`. Reset the
advances and `glyph_index` to zero.
In commit 531d463aed365b
[truetype] Allocate TT_ExecContext in TT_Size instead of TT_Driver.
the `TT_ExecContext` was moved from being on the driver to being on the size
to make it easier to use FreeType in a multi-threaded environment. However,
the documentation for `TT_New_Context` was not updated and still reflects
the old behavior and parameter list.
This change updates `TT_New_Context` documentation to reflect the current
parameters and usage.
`FDEF` instructions are specified as allowed only in 'prep' or
'fpgm'. FreeType has attempted to prevent their use in the glyph
program, but they were still allowed in glyph programs if defined in
a function defined in 'prep' or 'fpgm' and called from the glyph
program.
Similarly, `IDEF` instructions are specified not to be able to
modify any existing instruction. FreeType has attempted to prevent
their use in the glyph program, but they can still be used like
`FDEF`.
This change stores the initial bytecode range type and disallows the
use of `FDEF` and `IDEF` while running the glyph program.
Most other state is copied from the `TT_Size` into the execution
context. However, it is possible for a glyph program to use `WS` to
write to the storage area or `WCVTP`, `WCVTF`, and `DELTAC[123]` to
write to the control value table.
Allowing any change to the global state from the glyph program is
problematic as the outlines of any given glyph may change based on
the order the glyphs are loaded or even how many times they are
loaded. There exist fonts that write to the storage area or the
control value table in the glyph program, so their use should not be
an error.
Possible solutions to using these in the glyph program are
* ignore the writes;
* value-level copy on write, discard modified values when finished;
* array-level copy on write, discard the copy when finished;
* array-level copy up-front.
Ignoring the writes may break otherwise good uses. A full copy
up-front was implemented, but was quite heavy as even well behaved
fonts required a full copy and the memory management that goes along
with it. Value-level copy on write could use less memory but
requires a great deal more record keeping and complexity. This
change implements array-level copy on write. If any attempt is made
to write to the control value table or the storage area when the
initial bytecode range was in a glyph program, the relevant array
will be copied to a designated storage area and the copy used for
the rest of the glyph program's execution.
* src/truetype/ttinterp.h (TT_ExecContextRec): New fields
`iniRange`, `glyfCvtSize`, `glyfCvt`, `origCvt`, `glyfStoreSize`,
`glyfStorage`, and `origStorage`.
* src/truetype/ttinterp.c (Modify_CVT_Check): New function to handle
`exc->glyfCvt`.
(Write_CVT, Write_CVT_Stretched, Move_CVT, Move_CVT_Stretched): Use
it.
(Ins_WS): Handle `exc->glyfStorage`.
(Ins_FDEF, Ins_IDEF): Updated.
(TT_RunIns): Updated.
(TT_Done_Context): Free 'glyf' CVT working and storage area.
(TT_Load_Context): Fix/add casts.
* src/truetype/ttgload.c (TT_Load_Simple_Glyph): Fix cast.
* src/sfnt/ttcolr.c (get_child_table_pointer): New function to fetch
child table pointer early for all paint formats that compute a child
table pointer.
(read_color_line, read_paint): Updated.
(tt_face_get_colorline_stops): Check `colr->table`.
Reported as
https://bugs.chromium.org/p/chromium/issues/detail?id=1182552
Memory is allocated and the pointer assigned to `rows` inside a
'setjmp' scope. This memory must be freed outside the 'setjmp'
scope after a 'longjmp'. Since `rows` is a local and modified
inside the 'setjmp' scope it must be marked volatile or it will have
an indeterminate value after the 'longjmp'.
* src/sfnt/pngshim.c (Load_SBit_Png): Fix memory leak of `rows`.
We now record `cover' and `area' directly into the linked list. This
makes rendering faster by 10% or even more at larger sizes.
* src/smooth/ftgrays.c (FT_INTEGRATE): Write directly.
(gray_TWorker): Add direct cell reference and remove unused fields.
(gray_set_cell): Consolidate the linked list management and pointers.
(gray_convert_glyph, gray_convert_glyph_inner): Updated.
This change fixes a crash that occurs in `Load_SBit_Png` when
running on a 64-bit Windows OS. A memory access violation exception
would be raised by `setjmp` if the `jmp_buf` is not aligned to a
16-byte memory boundary. This is due to setjmp executing `movdqa`
instructions to store 128-bit XMM registers to memory, which require
correct memory alignment. This problem occurs because
`png_create_read_struct` uses `malloc` and `free` for memory
management, which only guarantees 8-byte alignment on Windows.
Instead, to fix the problem, `png_create_read_struct_2` is used on
64-bit Windows, which allows for user-defined memory allocation and
deallocation callbacks to be specified. These callbacks forward the
allocation and deallocation requests to `_aligned_alloc` and
`_aligned_free`, ensuring that the allocated `png_struct` and
internal `jmp_buf` have the requisite 16-byte alignment.
* src/sfnt/pngshim.c <_WIN64>: Include `malloc.h`.
(malloc_callback, free_callback) <_WIN64>: New functions.
(Load_SBit_Png) <_WIN64>: Use `png_create_read_struct_2` instead of
`png_create_read_struct`
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=28148
* src/sfnt/sfwoff2.c (woff2_open_font): Reject fonts that have
multiple tables with the same tag. While not explicitly forbidden
in the OpenType specification, it is implicitly forbidden by
describing a binary search algorithm for tables that only works
reliably if table tags are unique.
This is for future changes with Meson, which doesn't allow a
different name for its `subprojects` directory. Having both a
`submodules` and a `subprojects` directory is confusing.
* .gitmodules, autogen.sh (copy_submodule_files, DLG_INC_DIR,
DLG_SRC_DIR): Updated.
* builds/toplevel.mk (<top-level>, do-dist),
builds/windows/vc2010/script.bat: Updated.
* src/tools/no-copyright: Updated.
* freetype.h (FT_PaintFormat): Update paint format identifiers after
a specification change. The specification was updated to have
sibling formats, variable and non-variable variants for each.
Reflect that here.
* sfnt/ttcolr.c (read_paint): Remove parsing of variable indices as
the non-variable formats no longer have them.
* include/freetype/internal/compiler-macros.h (FT_COMPARE_DEF):
Add new macro.
* src/base/ftrfork.c, src/bdf/bdflib.c, src/gxvalid/gxvcommn.c,
src/psaux/afmparse.c, src/psnames/psmodule.c, src/type1/t1afm.c,
src/sfnt/sfwoff.c, src/sfnt/sfwoff2.c: Update qsort callbacks.
Fixes#1026 when compiling FreeType with an unusual calling convention
while the C library qsort still expects cdecl.
* include/freetype/freetype.h (FT_Get_Color_Glyph_Paint):
Additional function argument root_transform to control whether
root transform should be returned.
(FT_OpaquePaint): Additional tracking field to denote whether
root transform is to be returned.
* include/freetype/internal/sfnt.h
(TT_Get_Color_Glyph_Paint_Func): Propagate additional argument.
* src/base/ftobjs.c (FT_Get_Color_Glyph_Paint): Ditto.
* src/sfnt/ttcolr.c (tt_face_get_colr_glyph_paint): Return root
transform reflecting the size and tranform configured on
FT_Face.
(read_paint): Initialize and track status of insert_root_transform
flag.
This ensures good logging output, with all lines having a proper
prefix (if requested).
This is a continuation of a similar patch from 2020-12-02, which
missed some locations.
It's easier to have stubs if FT_DEBUG_LOGGING is undefined than to
modify `ftexport.sym` conditionally.
Problem reported by Alexei.
* src/base/ftdebug.c: Include `ftlogging.h`.
(FT_Trace_Set_Level, FT_Trace_Set_Default_Level, FT_Set_Log_Handler,
FT_Set_Default_Log_Handler) [!FT_DEBUG_LOGGING]: Provide stubs.
* sfnt/ttcolr.c (tt_face_get_colr_glyph_paint,
tt_face_get_colorline_stops, tt_face_get_paint): Additional checks
for whether colr table is present. Prevents crashes when these
methods are called on non-COLR fonts.
* src/base/ftobjs.c (FT_Get_Color_Glyph_Paint, FT_Get_Paint_Layers,
FT_Get_Paint, FT_Get_Colorline_Stops): Add basic sanity checks,
check for existence of `FT_Face`, check arguments and delegate calls
for the respective 'COLR' v1 API to the SFNT driver.
* include/freetype/internal/sfnt.h (TT_Get_Color_Glyph_Paint_Func,
TT_Get_Paint_Layers_Func, TT_Get_Colorline_Stops_Func,
TT_Get_Paint_Func): New function pointer types.
(SFNT_Interface): Add them.
(FT_DEFINE_SFNT_INTERFACE): Updated.
* src/sfnt/sfdriver.c (PUT_COLOR_LAYERS_V1): New macro.
(sfnt_interface): Add new function pointers.
* src/sfnt/ttcolr.c (tt_face_get_paint_layers): New function to get
the layers of a `PaintColrLayers` table in the font, using an
`FT_LayerIterator` from an `FT_PaintColrLayers` object retrieved via
`tt_face_get_paint`.
* src/sfnt/ttcolr.h: Updated.
* src/sfnt/ttcolr.c (tt_face_get_colorline_stops): New function to
return the current `FT_ColorStop` object from `FT_ColorStopIterator`.
Also increment the iterator.
* src/sfnt/ttcolr.h: Updated.
* src/sfnt/ttcolr.c (tt_face_get_paint): New function to resolve an
`FT_OpaquePaint` paint reference into an `FT_COLR_Paint` object of a
certain format, which contains the detailed information stored in a
paint of the respective format.
(read_paint): New function to provide the format specific parsing
and to populate the data members of each specific `FT_COLR_Paint`
subtype.
(read_color_line): New function to parse retrieved color line
information into an `FT_ColorLine` object, which has information
about the color line extend mode as well as an
`FT_ColorStopIterator` object.
* src/sfnt/ttcolr.h: Updated.
* src/sfnt/ttcolr.c (BaseGlyphV1Record): New structure.
(tt_face_load_colr): Handle version 1 table header.
(find_base_glyph_v1_record): New auxiliary function.
(tt_face_get_colr_glyph_paint): New function to find the root
`FT_OpaquePaint` object for a given glyph ID.
* src/sfnt/ttcolr.h: Updated.
* include/freetype/freetype.h (FT_Get_Color_Glyph_Paint): New method
for retrieving the root paint object for a color glyph by specifying
a glyph ID.
(FT_Get_Paint_Layers): New method for retrieving the layers of a
`PaintColorGlyph`.
(FT_Get_ColorLine_Stops): New method for retrieving the stops of a
color.
(FT_Get_Paint): New method for resolving an `FT_OpaquePaint` into an
`FT_COLR_Paint` object.
* src/sdf/module.mk, src/sfd/rules.mk: New files.
* src/sdf/ftsdf.h (ft_sdf_raster): New forward declaration.
* include/freetype/ftmoderr.h (FT_MODERRDEF): Add error definition
for the 'sdf' module.
* include/freetype/internal/fttrace.h (FT_TRACE_DEF): Add trace
definition for the `sdf' module.
* modules.cfg (RASTER_MODULES): Add the `sdf' module to the list of
rasterizers.
* src/sdf/ftsdf.c (sdf_generate_bounding_box): New function, which
is an optimized version of `sdf_generate`.
(sdf_generate_subdivision): New function.
* src/sdf/ftsdf.c (sdf_generate): New function, currently disabled.
This is a proof-of-concept implementation: It doesn't use any
optimization, it simply checks all grid points against all contours.
* src/sdf/ftsdf.c (get_min_distance_line, get_min_distance_conic,
get_min_distance_cubic): New functions. Note that
`get_min_distance_conic` comes with two implementations (using an
analytical and an iterative method, to be controlled with the
`USE_NEWTON_FOR_CONIC` macro).
* src/sdf/ftsdf.c (FT_DEBUG_INNER, FT_ASSIGNP_INNER)
[FT_DEBUG_LEVEL_TRACE && FT_DEBUG_MEMORY]: New macros.
(SDF_MemoryUser) [FT_DEBUG_LEVEL_TRACE && FT_DEBUG_MEMORY]: New
struct for memory usage tracing.
(sdf_alloc, sdf_free) [FT_DEBUG_LEVEL_TRACE && FT_DEBUG_MEMORY]: New
functions for memory usage tracing.
(SDF_ALLOC, SDF_FREE): New macros for memory management.
(SDF_MEMORY_TRACKER_DECLARE, SDF_MEMORY_TRACKER_SETUP,
SDF_MEMORY_TRACKER_DONE): New macros to set up memory usage tracing.
(USE_NEWTON_FOR_CONIC, MAX_NEWTON_DIVISIONS, MAX_NEWTON_STEPS,
CORNER_CHECK_EPSILON, CG_DIMEN): New configuration macros for
controlling the process of finding the shortest distance.
(MUL_26D6, VEC_26D6_DOT): New auxiliary macros.
(SDF_TRaster, SDF_Edge, SDF_Contour, SDF_Shape, SDF_Signed_Distance,
SDF_Params): New structs for setting up SDF data.
(SDF_Edge_Type, SDF_Contour_Orientation): New enums for SDF data.
(zero_vector, null_edge, null_contour, null_shape, max_sdf): Useful
constants.
(sdf_edge_new, sdf_edge_done, sdf_contour_new, sdf_contour_done,
sdf_shape_new, sdf_shape_done): New constructors and destructors.
Here is a breakdown of what the files will contain.
* src/sdf/ftsdfrend.c, src/sdf/ftsdfrend.h: The 'sdf' and 'bsdf'
renderers.
* src/sdf/ftsdf.c, src/sdf/ftsdf.h: The rasterizer for the 'sdf'
renderer.
* src/sdf/ftbsdf.c, src/sdf/ftbsdf.h: The rasterizer for the 'bsdf'
renderer.
* src/sdf/ftsdfcommon.h: Commmon properties and functions for both
rasterizers.
* src/sdf/ftsdferrs.h: Common error defines.
* src/sdf/sdf.c: For building a single object of the entire module.
* src/base/ftobjs.c (FT_Render_Glyph_Internal): Do not return if the
glyph's slot format is `FT_GLYPH_FORMAT_BITMAP`. The forthcoming
'bsdf' renderer will require bitmaps for processing.
* src/base/ftobjs.c (ft_add_renderer, ft_remove_renderer): Remove
renderer's glyph format check before adding and removing them. The
'bsdf' renderer will have a format `FT_GLYPH_FORMAT_BITMAP`.
The old code used to divide the darkening amount by em_ratio twice,
leading to unnecessarily bold stems on certain fonts with higher
units per em (e.g. Inter). This patch fixes it.
The return value of af_loader_compute_darkening was also changed to
use 16.16 fixed point to get rid of a redundant truncation operation.
This should slightly improve the precision, although it's still
bottlenecked by the emboldening function, which uses 26.6 fixed point.
* src/autofit/afloader.[ch]
(af_loader_compute_darkening): Return FT_Fixed.
(af_loader_embolden_glyph_in_slot): Revise calculations.
* src/truetype/ttgload.c (tt_get_metrics, load_truetype_glyph):
Previously, the code would populate the phantom points before
calling the `get_glyph_metrics` callback. For formats like PCL XL
format 1, class 2 downloaded fonts (where metrics are removed from
the TTF header), this causes problems when the hinting program uses
the phantom points (misplaced and distorted glyphs) due to the
metrics being unset (all zeros).
(tt_get_metrics_incr_overrides): Renamed to...
(tt_get_metrics_incremental): ... this. Updated caller
* include/freetype/ftincrem.h: Update the documentation to make it
clearer that `get_glyph_metrics` is to retrieve metrics from a
non-standard source, but *not* for the purpose of imposing custom
metrics.
* src/dlg/dlgwrap.c: Duplicate some feature test macros from
`dlg.c`, which must come first before loading standard headers. For
example, `freetype.h` loads `stdio.h` if compiled in debug mode.
Based on a patch by Priyesh.
* include/freetype/internal/fttrace.h (FT_MAX_TRACE_LEVEL_LENGTH):
New macro.
* src/base/ftdebug.c, builds/windows/ftdebug.c (ft_log_handler):
Print logs after a fixed width to handle different lengths of
`FT_COMPONENT` entries.
Use `ft_strrchr` to check for final newline character.
We only support Visual C++ 2010 and newer.
* builds/windows/vc2010/script.bat: New windows batch file to copy
necessary 'dlg' files from `submodules/dlg` to `src/dlg`. This file
is used as a pre-built event in Visual C++.
* builds/windows/ftdebug.c: Synchronize with `src/base/ftdebug.c`.
* builds/windows/vc2010/freetype.vcxproj.filters: Add
`src/dlgwrap.c`.
* builds/windows/vc2010/freetype.vcxproj
(AdditionalIncludeDirectories): Add include files of dlg for 'Debug'
and 'Debug Static' configurations on both 'x64' and 'win32'
platforms.
(PreprocessorDefinitions): Add `FT_LOGGING` for 'Debug' and 'Debug
Static' configurations on both 'x64' and 'win32' platforms.
Add `DLG_STATIC' for 'Debug' configuration on 'x64' and 'win32'
platforms.
(DisableLanguageExtensions): We need to disable the `/Za` option
when building 'dlg' with FreeType as 'dlg' strictly follows the C99
standard. Visual C++ produces behaves unexpectedly when
compiling a C99 file with `/Za` option enabled.
* include/freetype/ftlogging.h (FT_Custom_Log_Handler): New function
typedef to store the custom callback logging function.
(FT_Set_Log_Handler, FT_Set_Default_Log_Handler): New functions to
set and reset custom log handler.
* include/freetype/internal/ftdebug.h (custom_output_handler): New
variable to support a custom callback logging function.
(FT_Logging_Callback): A new function typedef to print log using
custom callback logging function, which is set using
`FT_Set_Log_Handler`.
(FT_Log): Use it.
* src/base/ftdebug.c (FT_Set_Log_Handler,
FT_Set_Default_Log_Handler, FT_Logging_Callback): Add function
definitions.
* include/freetype/ftlogging.h: New header file.
* include/freetype/internal/ftdebug.h [FT_LOGGING]: Include
`ftlogging.h`.
* src/base/ftdebug.c (ft_custom_trace_level): New variable.
(ft_debug_init): Update to support change of levels of tracing
components of FreeType at run-time.
(FT_Trace_Set_Level): New function to change the levels of tracing
components at run-time.
(FT_Trace_Set_Default_Level): New function to reset the levels of
tracing components back to default.
* include/freetype/internal/ftdebug.h (FT_LOGGING_TAG,
FT_LOGGING_TAG_): New macros to resolve the value of `FT_COMPONENT'
into a string.
(ft_add_tag, ft_remove_tag): New functions to add and remove dlg tags.
* src/base/ftdebug.c: Add new variables to control the logging of
timestamp and name of `FT_COMPONENT` along with actual logs.
(ft_add_tag, ft_remove_tag): Add function definitions.
(ft_log_handler): Updates to print timestamp and name of
`FT_COMPONENT`.
(ft_debug_init) [FT_LOGGING]: Users can now control the logging of
timestamp and name of `FT_COMPONENT` by adding tags in the
`FT2_DEBUG` environment variable.
* include/freetype/internal/ftdebug.h: Added dlg's header files.
(FT_LOG): New macro to redirect trace logs to dlg's API's whenever
`FT_LOGGING' is defined.
(ft_logging_init, ft_logging_deinit): New functions to handle
initialization and uninitialization of logging related variables.
(ft_log_handler): New function to handle logs of FreeType.
* src/base/ftdebug.c: Add necessary logging related variables.
(ft_logging_init, ft_logging_deinit, ft_log_handler): Add function
definitions.
* src/base/ftinit.c (FT_Init_FreeType) [FT_LOGGING]: Call
`ft_logging_init`.
(FT_Done_FreeType) [FT_LOGGING]: Call `ft_logging_deinit`.
* src/base/ftobjs.c (FT_New_Library): Call `ft_debug_init` only if
`FT_LOGGING` is not defined.
* autogen.sh (copy_submodule_files): New script to copy all the
necessary source and include files from `submodules/dlg` to
`src/dlg`.
* src/dlg/dlgwrap.c: New wrapper file for `src/dlg.c`. It enables
the build of 'dlg' if the `FT_LOGGING` macro is defined.
* src/dlg/rules.mk: New sub-Makefile.
* builds/freetype.mk (DLG_DIR): New variable to include the
header files of the 'dlg' library.
(INCLUDES): Add `DLG_DIR`.
(FT_CFLAGS): Add `-std=c99' flag.
Include `src/dlg/rules.mk` file to build 'dlg' library.
(OBJ_S, OBJ_M): Add `DLG_OBJS_M` and `DLG_OBJS_S`.
* builds/toplevel.mk: For builds directly from the git repository
we need to copy files from `submodule/dlg` to `src/dlg`.
* include/freetype/config/ftoption.h, devel/ftoption.h (FT_LOGGING):
New macro to enable or disable the logging facility in FreeType.
==========================
Tag sources with `VER-2-10-4'.
* docs/VERSION.TXT: Add entry for version 2.10.4.
* docs/CHANGES: Updated.
* README, src/base/ftver.rc, builds/windows/vc2010/index.html,
builds/windows/visualc/index.html,
builds/windows/visualce/index.html,
builds/wince/vc2005-ce/index.html,
builds/wince/vc2008-ce/index.html, docs/freetype-config.1:
s/2.10.3/2.10.4/, s/2103/2104/.
* include/freetype/freetype.h (FREETYPE_PATCH): Set to 4.
* builds/unix/configure.raw (version_info): Set to 23:4:17.
* CMakeLists.txt (VERSION_PATCH): Set to 4.
==========================
Tag sources with `VER-2-10-3'.
* docs/VERSION.TXT: Add entry for version 2.10.3.
* README, src/base/ftver.rc, builds/windows/vc2010/index.html,
builds/windows/visualc/index.html,
builds/windows/visualce/index.html,
builds/wince/vc2005-ce/index.html,
builds/wince/vc2008-ce/index.html, docs/freetype-config.1:
s/2.10.2/2.10.3/, s/2102/2103/.
* include/freetype/freetype.h (FREETYPE_PATCH): Set to 3.
* builds/unix/configure.raw (version_info): Set to 23:3:17.
* CMakeLists.txt (VERSION_PATCH): Set to 3.
This corresponds to the following commits in the ttfautohint git
repository:
bb6842bd3bd437b7b4a7921b0376c860f5e73d18 Typo, formatting.
d5c91ddb1cb310257a3dfe9a8e20e1fc51335faa Add Medefaidrin script.
* src/autofit/afblue.dat: Add blue zone data for Medefaidrin.
* src/autofit/afblue.c, src/autofit/afblue.h: Regenerated.
* src/autofit/afscript.h: Add Medefaidrin standard characters.
* src/autofit/afranges.c, src/autofit/afstyles.h: Add Medefaidrin
data.
* scr/tools/scripts/make_distribution_archives.py: (_TOP_DIR,
_SCRIPT_DIR): Updated to new location.
(main): s/shutils.copyfile/shutils.copy/ to preserve file
permissions.
(main): Prefix source file paths with `git_dir` while copying files
to allow calls of the script from other places than the top-level
directory.
Towards the end of the the function there is a call to
`FT_Outline_Get_CBox` that retrieves the glyph bbox in scaled units.
That sets `horiBearing{X,Y}` and `vertBearingX` but `vertBearingY`
is left alone, and is not scaled.
Patch from Eric Muller <emuller@amazon.com>.
Besides dropout control the second horizontal sweep is supposed to
clean up straight horizontal edges that are mishandled by the first
vertical sweep when a line passes through pixel centers. This line
would present as perfectly aligned span edges in the second sweep.
* src/raster/ftraster.c (Horizontal_Sweep_Span): Replace the old
implementation with a better one focusing on aligned span edges only.
Windows seems to perform smart dropout control at 26.6 precision.
To mimick Windows independent of increased precision, we need to tweak
the macro so that some close calls break down rather than up.
* src/raster/ftraster.c (SMART): Tweak the macro.
* src/truetype/ttinterp.h (TT_Round_Func): Change the last argument.
* src/truetype/ttinterp.c (Ins_ROUND, Ins_NROUND, Ins_MDAP, Ins_MIAP,
Ins_MDRP, Ins_MIRP): Move compensation retrieval from here...
(Round_*): ... to here.
* src/truetype/ttobjs.c (tt_size_init_bytecode): Reserve zero
compensation at color index 3.
For glyphs with OVERLAP_SIMPLE or OVERLAP_COMPOUND, set
FT_OUTLINE_OVERLAP to render them with direct oversampling, which
mitigates artifacts (see 3bb512bc9f62).
* include/freetype/ftimage.h (FT_OUTLINE_OVERLAP): Redefine to rhyme
with OVERLAP_SIMPLE.
* src/base/ftgloadr.c (FT_GlyphLoader_Rewind): Reset outline flags.
* src/truetype/ttgload.c
(TT_Load_Simple_Glyph): Retain OVERLAP_SIMPLE.
(load_truetype_glyph): Retain OVERLAP_COMPOUND.
* include/freetype/internal/autohint.h
(FT_DECLARE_AUTOHINTER_INTERFACE): New macro.
* src/autofit/afmodule.h: Use it to declare
`af_autofitter_interface'.
* include/freetype/internal/ftobjs.h (FT_DECLARE_GLYPH): New macro.
* src/base/ftbase.h: Use it to declare `ft_bitmap_glyph_class' and
`ft_outline_glyph_class'.
* src/base/ftglyph.c: Include `ftbase.h'.
* src/cff/cffparse.c (cff_parser_run): Fix type of `t2_size'.
* src/pcf/pcfdrivr.c (pcf_cmap_char_next): Fix type of `result'.
* src/psaux/psauxmod.c (psaux_module_class): Use `FT_DEFINE_MODULE'.
* src/psaux/psauxmod.h: Declare `afm_parser_funcs',
`t1_cmap_classes', `cff_decoder_funcs', and `psaux_module_class'.
* src/pshinter/pshmod.c: Include `pshmod.h'.
* src/sfnt/sfwoff2.c (ROUND4, WRITE_SHORT): Fix implicit sign
conversion.
(compute_ULong_sum): Fix return type.
Fix implicit sign conversion.
(store_points): Fix type of `last_flag', `repeat_count', and `flag'.
Use casts to avoid warnings.
(reconstruct_glyf): Fix implicit sign conversion.
Use cast to avoid warning.
(get_x_mins): Fix implicit sign conversion.
* src/sfnt/ttcmap.c: Undef `TTCMAPCITEM'.
* src/sfnt/ttcmap.h: Define `TTCMAPCITEM' and include `ttcmapc.h' to
declare cmap classes.
* src/smooth/ftsmooth.c (ft_smooth_overlap_spans): Use cast.
* src/truetype/ttinterp.c (Ins_MIAP): Fix typo.
The following builds were still failing due to previous changes:
make multi
make multi CC="c++"
make CC="c++"
This patch fixes the issues, which were missing includes to get the
right macro definitions in multi-build mode.
Also, `FT_UNUSED' is actually used by third-party code, so move it
back to `public-macros.h' to avoid breaking it.
* include/freetype/config/public-macros.h (FT_EXPORT): Remove
special definition for C++.
(FT_UNUSED): Define here instead of...
* include/freetype/config/compiler-macros.h: ... here.
(FT_FUNCTION_DECLARATION): Remove special definition for C++.
(FT_LOCAL_ARRAY_DEF): Fix definition.
* src/cache/ftccback.h, src/lzw/ftzopen.h, src/gxvalid/gxvmort.h,
src/gxvalid/gxvmorx.h: Add `FT_BEGIN_HEADER' and `FT_END_HEADER'.
The following builds were failing due to previous changes:
make multi
make multi CC="c++"
* include/freetype/config/ftconfig.h: Remove `FT_END_HEADER'.
* include/freetype/config/ftheader.h (FT_BEGIN_HEADER,
FT_END_HEADER): Protect against redefinition.
* src/cache/ftccache.h, src/cache/ftcmru.h, src/pcf/pcfutil.h,
src/psaux/pserror.h, src/psaux/psft.h, src/psaux/psstack.h,
src/sfnt/woff2tags.h: Include `compiler-macros.h'.
* src/sfnt/woff2tags.c: Include `woff2tags.h'.
* src/psaux/t1decode.c (t1_decoder_parse_metrics): Copy
corresponding code from old engine's `t1_decoder_parse_charstrings'
function to handle `op_callsubr' and `op_return'.
* builds/unix/ftconfig.h.in, builds/vms/ftconfig.h,
include/freetype/config/ftconfig.h: Split off macro definitions
required by the FreeType API headers to...
* include/freetype/config/public-macros.h: ...this new file.
* builds/unix/ftconfig.h.in, builds/vms/ftconfig.h,
include/freetype/config/ftconfig.h: Split off macro definitions used
by the library but not to be exposed to clients to...
* include/freetype/config/compiler-macros.h: ...this new file.
* include/freetype/internal/*.h, src/raster/ftraster.h: Include
`compiler-macros.h' where needed.
This implements oversampling to metigate artifacts in pixels partially
covered by overlapping contours. It turns out that the 4x4
oversampling is sufficient but, at least, quadruples the rendering
time. The outline has to set FT_OUTLINE_OVERLAP to use this method.
* include/freetype/ftimage.h (FT_OUTLINE_OVERLAP): New flag.
* src/smooth/ftsmooth.c (ft_smooth_render): Check it to...
(ft_smooth_raster_overlap): ... inflate outline and set up direct
rendering for oversampling with...
(ft_smooth_overlap_spans): ... new span function that integrates them.
Instead of rendering 3 bitmaps side by side and reshuffling, we use
direct rendering to deliver the bitmaps on each third byte.
* src/smooth/ftsmooth.c (ft_smooth_raster_lcd)
[!FT_CONFIG_OPTION_SUBPIXEL_RENDERING]: Set up direct mode with...
(ft_smooth_lcd_spans): ... new span function.
This makes `ft_smooth_render' a lot smaller and easier to follow. It
also cleanly separates Harmony and ClearType-style LCD rendering
algorithms. Now I only wish to move LCD filtering and geometry from
FT_Library to FT_Renderer.
* src/smooth/ftsmooth.c (ft_smooth_render): Move LCD code from here...
(ft_smooth_raster_lcd, ft_smooth_raster_lcdv): ... to here.
[FT_CONFIG_OPTION_SUBPIXEL_RENDERING]: Reorganize #ifdef's.
* src/cff/cffobjs.c (cff_face_init): If a call to `cff_strcpy' fails
by returning NULL in `cff_face_init', `remove_style' is still
called. This means that the NULL pointer is dereferenced, causing a
crash.
* src/cff/cffobjs.c (cff_size_init): If a call to `funcs->create'
fails to allocate one of the `internal->subfont' variables, make
sure to free `internal->topfont' and any successfully allocated
subfonts.
* src/base/ftobjs.c (FT_New_Size): Avoid trying to free
`size->internal`, unless `size' has been allocated. This
mistake appeared in the fix for 58611.
* src/cff/cffobjs.c (cff_slot_done), src/cid/cidobjs.c
(cid_slot_done): If `ft_glyphslot_init' fails to allocate
`internal', then the class' `done_slot' callback (called by
`ft_glyphslot_done') must not dereference the pointer to `internal'.
* src/cff/cffload.c (cff_index_get_pointers): When new_bytes fails
to allocate, make sure to free the table. Do the same for both
allocations if there is a later error.
We no longer have to take care of the 8.3 file name limit; this
allows us (a) to introduce longer, meaningful file names, and (b) to
avoid macro names in `#include' lines altogether since some
compilers (most notably Visual C++) doesn't support this properly.
*/*: Replace
#include FOO_H
with
#include <freetype/foo.h>
or something similar. Also update the documentation.
These have not been used in a very, very long time, so better remove
them. A corresponding patch will be submitted to the
`freetype2-demos' repository.
* src/Jamfile, src/*/Jamfile, Jamrules: Delete.
The LCD modules were never truly independent. They mostly served as
a way to disable patented LCD rendering, which is no longer necessary.
The `smooth' module now handles LCD modes as well.
* src/smooth/ftsmooth.c (ft_smooth_lcd_renderer_class.
ft_smooth_lcdv_renderer_class): Deleted.
(ft_render_smooth): Reworked from `ft_render_smooth_generic'.
* src/smooth/ftsmooth.h: Remove dedicated LCD classes.
* src/smooth/module.mk: Remove dedicated LCD modules.
* include/freetype/config/ftmodule.h: Ditto.
* builds/amiga/include/config/ftmodule.h: Ditto.
* include/freetype/ftmodapi.h: Do not mention LCD modules.
Tag sources with `VER-2-10-2'.
* docs/VERSION.TXT: Add entry for version 2.10.2.
* README, Jamfile (RefDoc), src/base/ftver.rc,
builds/windows/vc2010/index.html, builds/windows/visualc/index.html,
builds/windows/visualce/index.html,
builds/wince/vc2005-ce/index.html,
builds/wince/vc2008-ce/index.html, docs/freetype-config.1:
s/2.10.1/2.10.2/, s/2101/2102/.
* include/freetype/freetype.h (FREETYPE_PATCH): Set to 2.
* builds/unix/configure.raw (version_info): Set to 23:2:17.
* CMakeLists.txt (VERSION_PATCH): Set to 2.
* docs/CHANGES: Updated.
In order to support `gzip' encoded header the call to
`inflateInit2' was modified in commit 6a92b1fadde26477a9179.
However, this code breaks with the outdated internal version
of zlib. This is a temporary fix to conditionally support
`gzip' encoded header whenever a system installation of zlib
is being used.
Problem report in
https://lists.nongnu.org/archive/html/freetype-devel/2020-02/msg00023.html
* src/gzip/ftgzip.c (FT_Gzip_Uncompress): Change the the call to
`inflateInit2' depending on whether the system installation is
being used or the internal copy.
By resetting the blend as implemented with this commit fonts with
invalid `gvar' tables may keep calling into `ft_var_load_gvar' from
`tt_set_mm_blend' and failing, but the font was invalid anyway and
we want to keep seeing the failure in `tt_set_mm_blend'.
* src/truetype/ttgxvar.c (ft_var_load_gvar): Calculate length of
offset array once.
Allocate arrays after `FT_FRAME_ENTER' (extra check before
allocating and avoid needing to free array later if error entering
frame).
Always call `FT_FRAME_EXIT'.
Consistently set counts immediately after array initialized.
Reset the blend (particularly `blend->glyphoffsets') on failure.
* src/truetype/ttgxvar.c (ft_var_load_gvar): Delay settings of any
`blend->xxxcount' values until the corresponding data has been
checked.
Also do some sanitizing to avoid a too early exit.
(TT_Vary_Apply_Glyph_Deltas): Improve tracing message.
* src/base/ftpsprop.c (ps_property_set) [hinting-engine]: Avoid an
incorrect return value that caused a warning. The function did the
right thing, though.
* src/base/ftoutln.c (FT_Outline_Transform): Bail on empty points.
* src/cff/cffload.c (cff_subfont_load): Use `FT_OFFSET'.
* src/psaux/psft.c (cf2_decoder_parse_substrings): Early out if
`charstring_base' or `charstring_len' are null.
* src/sfnt/ttload.c (tt_face_load_name): Use `FT_OFFSET'.
* src/base/ftgloadr.c (FT_GlyphLoader_Adjust_Points,
FT_GlyphLoader_Adjust_Subglyphs): Use `FT_OFFSET'.
(FT_GlyphLoader_CreateExtra): Add short cut if some values are zero.
Also reduce number of SFNT table lookups.
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=18065
* include/freetype/internal/wofftypes.h (WOFF2_InfoRec): Add fields
`glyf_table', `loca_table', and `head_table'.
* src/sfnt/sfwoff2.c (reconstruct_glyf): Update signature.
Use table pointers in `info' parameter.
(get_x_mins): Check `maxp_table'
Use table pointers in `info' parameter.
(reconstruct_font): Use and set table pointers in `info' parameter.
Fix check for `glyf' and `loca' tables.
Update call to `reconstruct_glyf'.
(woff2_open_font): Updated.
* include/freetype/internal/ftmemory.h (FT_MEM_FREE): Use
`FT_DEBUG_INNER' to set source code file name and line.
* src/base/ftdbgmem.c (ft_mem_table_remove): Better formatting of
tracing message.
Also fix memory deallocation in case of error.
`head' problem reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=17820
* src/sfnt/sfwoff2.c (reconstruct_glyf): Don't use `stream_close'.
Abort if `head_table' is NULL.
Don't free `transformed_buf' in case of error.
(woff2_open_font): Don't set `uncompressed_buf' to NULL.
* src/sfnt/sfwoff2.c (woff2_open_font): Use `FT_UInt32' for
`file_offset'. This fixes builds on platforms where `FT_LONG64' is
not defined while still being sufficient to store a file offset.
If table tag is not 0x3f, we expect a value between 0 and 62. If
this is not the case, exit with errors.
* src/sfnt/sfwoff2/c: Check whether table tag makes sense.
* src/sfnt/woff2tags.c: Return 0 if tag is out of bounds.
`reconstruct_hmtx' requires `info->x_mins' and `info->num_glyphs' to
reconstruct the hmtx table. In case glyf is not transformed, we
call `get_x_mins' which does the necessary work.
* src/sfnt/sfwoff2.c (get_x_mins): New function.
(reconstruct_font): Call get_x_mins.
Set correct value of `face->num_faces' for WOFF2 fonts. This is
being handled separately because we only load the tables for the
requested font face in `woff2_open_font' and create a single-face
sfnt stream.
The full discussion is at:
https://lists.gnu.org/archive/html/freetype-devel/2019-08/msg00000.html
* src/sfnt/sfobjs.c (sfnt_open_font): Add parameter
`woff2_num_faces'.
(sfnt_init_face): Introduce variable `woff2_num_faces', and change
`face->root.num_faces' if `woff2_num_faces' is set.
* src/sfnt/sfwoff2.c (woff2_open_font): Validate requested face
index and handle negative face indices.
* src/sfnt/sfwoff2.h (woff2_open_font): Add parameter `num_faces' to
declaration.
We do this by using `totalSfntSize' as an initial reference, and
extending the buffer when required. This reduces rendering time
considerably.
* include/freetype/internal/wofftypes.h (WOFF2_HeaderRec): Add
`totalSfntSize', rename `total_sfnt_size' to `actual_sfnt_size'.
* src/sfnt/sfwoff2.c (write_buf): Add parameter `dst_size' to keep
track of and update total size of stream.
(WRITE_SFNT_BUF, WRITE_SFNT_BUF_AT): Modify macros accordingly.
(pad4, store_loca, reconstruct_glyf, reconstruct_hmtx,
reconstruct_font): Update parameters to accept `sfnt_size'.
(woff2_open_font): Add variable `sfnt_size'. Use WOFF2 header field
`totalSfntSize' as initial reference (if value makes sense) and
allocate `totalSfntSize' bytes for the sfnt stream. `write_buf'
handles reallocation if and when required. Also resize the stream
to `actual_sfnt_size' after reconstruction.
Add necessary functions to reconstruct loca and hmtx tables (the two
remaining tables that can have a transform). `woff2_open_font' is
now capable of loading a woff2 font face. This code may still need
more refining and better memory management.
* include/freetype/internal/wofftypes.h (WOFF2_HeaderRec): Add total
(final) size of sfnt stream.
(WOFF2_InfoRec): Add header checksum value.
* src/sfnt/sfobjs.c (sfnt_open_font): Change `face_instance_index'
parameter to its pointer so its value can be modified by
`woff2_open_font'.
* src/sfnt/sfwoff2.c: (WRITE_SFNT_BUF_AT): New macro to write into
sfnt buffer at given position.
(write_buf): Add parameter `extend_buf' which allows caller to
specify whether buffer should be reallocated before copying data.
(WRITE_SFNT_BUF): Updated.
(pad4, store_loca, reconstruct_htmx): New functions.
(reconstruct_glyf): Calculate loca values and store them.
(reconstruct_font): Call `reconstruct_hmtx', write table record
entries, and calculate table checksums. Also calculate font
checksum and update `checksumAdjustment' entry in head table.
(woff2_open_font): Open stream for sfnt buffer, swap out input
stream and return.
* src/sfnt/sfwoff2.h (woff2_open_font): Modify parameter to accept
pointer to `face_index'.
Reconstruct `glyf' table if it is transformed in the uncompressed
table stream. Also add necessary structures, macros and functions.
* include/freetype/internal/wofftypes.h (WOFF2_InfoRec,
WOFF2_SubstreamRec, WOFF2_PointRec): New structures.
(WOFF2_TableRec): s/OrigLength/dst_length/.
* src/sfnt/sfwoff2.c (READ_255USHORT, READ_BASE128): Use
`FT_SET_ERROR' to set implicit `error' variable.
(WRITE_SHORT): New macro.
(N_CONTOUR_STREAM, N_POINTS_STREAM, FLAG_STREAM, GLYPH_STREAM,
COMPOSITE_STREAM, BBOX_STREAM, INSTRUCTION_STREAM): New macros to
refer to substreams of the transformed `glyf' tables.
(Read255UShort, ReadBase128): Return errors set by `FT_READ_XXX'
macros.
(with_sign, safe_int_addition): New functions to add sign to values
based on a flag and perform safe addition respectively.
(triplet_decode): Decode variable-length (flag, xCoordinate,
yCoordinate) triplet for a simple glyph. See
https://www.w3.org/TR/WOFF2/#triplet_decoding
(store_points, compute_bbox, composteGlyph_size, reconstruct_glyf):
New functions.
(reconstruct_font): Call `reconstruct_glyf'.
* src/sfnt/sfwoff2.h: Add required constants.
* src/sfnt/woff2tags.h: Move out constants to `sfwoff2.h'.
Copy un-transformed tables to the sfnt stream.
* src/sfnt/sfwoff2.c: (WRITE_SFNT_BUF): New macro.
(write_buf): New function. Extend memory of `dst' buffer and copy
bytes from `src'.
(compute_ULong_sum): New function. Calculate checksum of table.
(reconstruct_font): Change `FT_Byte* sfnt' to `FT_Byte**
sfnt_bytes'. This has been done because we reallocate memory to
`sfnt' multiple times, which may change the pointer value of `sfnt'.
This new pointer must be propogated back to the caller. Same reason
for using a double pointer in `write_buf'.
* src/sfnt/woff2tags.h (WOFF2_DEFAULT_MAX_SIZE): New macro used for
overflow checking.
Uncompressed buffer is now an `FT_Stream'.
Perform basic checks and start iterating over tables.
* src/sfnt/sfwoff2.c (stream_close, find_table, read_num_hmetrics):
New functions.
(reconstruct_font): Modify parameters and iterate over tables.
(woff2_open_font): Updated.
We `handle' TTCs by modifying the `indices' array to point to only
those tables that are part of the requested `face_index'.
Set and use `num_tables' in `WOFF2_TtcFont'.
* src/sfnt/sfwoff2.c (reconstruct_font): New function.
(woff2_open_font): Start reconstruction of font.
Change `KnownTags' to a function (`woff2_known_tags'). This avoids
introducing a global constant array. This function returns the
specified index without *any* checks. The caller must ensure that
`index' is within array limits.
* src/sfnt/sfwoff2.c (woff2_open_font): Change `KnownTags[...]'
notation to `woff2_known_tags( ... )'.
* src/sfnt/woff2tags.c: Perform changes.
* src/sfnt/woff2tags.h: Update definitions.
WOFF2 compressed stream is now uncompressed if Brotli is available.
This data is stored in a separate buffer (uncompressed_buf) because
it does not contain direct table data. Certain tables have
transformations applied to them, and they must be reconstructed
before we can write those tables to the SFNT stream.
`face_index' is now being passed as a parameter to
`woff2_open_font'.
* src/sfnt/sfobjs.c (sfnt_open_font): Add parameter
`face_instance_index'.
* src/sfnt/sfwoff2.c (woff2_uncompress): New function.
(woff2_open_font): Call `woff2_uncompress'.
(compute_first_table_offset): Fix return type.
* src/sfnt/sfwoff2.h (woff2_open_font): Modify declaration.
Add constants required for WOFF2, and known table tags as defined in
the specification. See
https://www.w3.org/TR/WOFF2/#table_dir_format
for details.
* src/sfnt/woff2tags.c, src/sfnt/woff2tags.h: New files.
Check for WOFF2 tag, call `woff2_open_font', and implement it to read
header according to specification.
* include/freetype/internal/fttrace.h: Add `sfwoff2.c'.
* src/sfnt/rules.mk (SFNT_DRV_SRC): Add `sfwoff2.c'.
* src/sfnt/sfnt.c: Include `sfwoff2.c'.
* src/sfnt/sfobjs.c (sfnt_open_font): Check for `wOF2' tag and call
`woff2_open_font'.
* src/sfnt/sfwoff2.c, src/sfnt/sfwoff2.h: New files.
As originally intended, a Type 1 SEAC charstring would be used for
an accented glyph (like `acaron' or `uumlaut'), where the advance
width of the SEAC glyph is the same as that of the `base' glyph
(like `a' or `u'). In this case it is not uncommon for the SEAC
glyph to not use an (H)SBW opcode of its own but to rely on the
value from the base glyph.
However, out-of-spec fonts also use SEAC glyphs for ligatures (like
`oe' or `fi'), and in those cases the overall advance width is
greater than that of the `base' glyph. For this reason we have to
allow that the SEAC glyph can have an (H)SBW value of its own, and
if it has, retain this value, rather than the one from the base
glyph.
* src/psaux/psintrp.c (cf2_interpT2CharString) <cf2_escSEAC>:
Implement it.
==========================
Tag sources with `VER-2-10-1'.
* docs/VERSION.TXT: Add entry for version 2.10.1.
* README, Jamfile (RefDoc), src/base/ftver.rc,
builds/windows/vc2010/freetype.vcxproj,
builds/windows/vc2010/index.html,
builds/windows/visualc/freetype.dsp,
builds/windows/visualc/freetype.vcproj,
builds/windows/visualc/index.html,
builds/windows/visualce/freetype.dsp,
builds/windows/visualce/freetype.vcproj,
builds/windows/visualce/index.html,
builds/wince/vc2005-ce/freetype.vcproj,
builds/wince/vc2005-ce/index.html,
builds/wince/vc2008-ce/freetype.vcproj,
builds/wince/vc2008-ce/index.html: s/2.10.0/2.10.1/, s/2100/2101/.
* include/freetype/freetype.h (FREETYPE_PATCH): Set to 1.
* builds/unix/configure.raw (version_info): Set to 23:1:17.
* CMakeLists.txt (VERSION_PATCH): Set to 1.
* include/freetype/fterrors.h (FT_Error_String): Fix C++ compilation.
* src/autofit/afglobal.c (af_face_global_get_metrics): Start again
(with dummy hinter module) if no blue zones are present.
* src/autofit/aflatin.c (af_latin_metrics_init_blues): Change
signature to return error code.
If no blue zones are found, update `glyph_styles' array to hold
AF_STYLE_NONE_DFLT instead of the current style.
(af_latin_metrics_init): Return internal error code if no blue zones
are found.
The buffer size FT_MAX_GRAY_SPANS is set to 10 spans, which should be
enough to cover the entire scanline for simple glyphs in most cases:
each slightly slanted edge needs up to two spans, plus a filling span
in-between. This is not new, we used to do it before cb4388783cecc.
* src/smooth/ftgrays.c (gray_TWorker): Add `spans' and `num_spans'.
(gray_hline, gray_sweep): Implement the span buffering.
(gray_raster_render): Use negative `num_spans' to avoid the direct
mode.
Up to now, only the unscaled CVT values were varied; in other words,
the `CVAR' data was never used for bytecode hinting.
* src/truetype/ttgxvar.c (tt_cvt_ready_iterator): New auxiliary
function.
(tt_face_vary_cvt): Use it to trigger rescaling of CVT values.
If `CVAR' data is applied to variation fonts, fractional values are
possible.
* include/freetype/internal/tttypes.h (TT_FaceRec): Change type of
`cvt' from `FT_Short' to `FT_Int32'.
* src/truetype/ttgxvar.c (FT_fdot6ToFixed): New macro.
(tt_face_vary_cvt): Use it to update code to 26.6 format.
* src/truetype/ttobjs.c (tt_size_run_prep): Update code to 26.6
format.
* src/truetype/ttpload.c (tt_face_load_cvt): Stora data in 26.6
format.
Very embarassing :-)
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=14701https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=14705https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=14710
* src/truetype/ttgload.c (IS_DEFAULT_INSTANCE): Move up and add
argument; update all callers.
(TT_Process_Simple_Glyph): Use it. The `unrounded' array is active
for variation fonts only, thus also enclose related code with
`#ifdef TT_CONFIG_OPTION_GX_VAR_SUPPORT ... #endif' where
necessary.
Revert commit a113e5d from 2019-05-09, and don't use `extra_points2'
but allocate a temporary array.
Speed up the scaling of the `unrounded' array.
* src/truetype/ttgxvar.c (FT_fixedToInt, FT_FixedToFdot6): Fix type
conversions and rounding. The unsigned type must have more or equal
bits to the signed type.
This patch make FreeType use font units in 26.6 format internally
instead of integers.
* src/truetype/ttgxvar.c (FT_fixedToFdot6): New macro.
(TT_Vary_Apply_Glyph_Deltas): Add argument to output unrounded font
coordinates.
* src/truetype/ttgxvar.h: Updated.
* src/truetype/ttgload.c (TT_Process_Simple_Glyph): Use
`extra_points2' array to temporarily hold unrounded point
coordinates; use them to compute scaled coordinates and linear
advance width and height.
(load_truetype_code): Adjust similarly.
Reported by Sender Ghost <lightside@gmx.com>.
* src/autofit/afcjk.c (af_cjk_hints_compute_edges): Catch a corner
case where `edge->first' could be NULL.
* src/pfr/pfrobjs.c (pfr_slot_load): Remove unnecessary test of
`size'.
* src/raster/ftraster.c (Draw_Sweep): Catch a corner case where
`draw_right' might be NULL.
* src/sfnt/ttmtx.c (tt_face_get_metrics): Fix limit test for
`aadvance'.
Ensure `abearing' always hold a meaningful result.
* src/truetype/ttgload.c (load_truetype_glyph): Ensure `subglyph' is
not NULL before accessing it.
* src/truetype/ttgxvar.c (TT_Set_Named_Instance): Remove unnecessary
test of `namedstyle'.
* src/type42/t42parse.c (t42_parser_done): Ensure
`parser->root.funcs.done' is not NULL before accessing it.
The previous implementation is correct but it is too complex.
The revised algorithm is based on the fact that each split moves
the control points closer to the trisection points on the chord.
The corresponding distances are good surrogates for the curve
deviation from the straight line.
This cubic flattening algorithm is somewhat similar to the conic
algorithm based the distance from the control point to the middle of
the chord. The cubic distances, however, decrease less predictably
but are easy enough to calculate on each step.
* src/smooth/ftgrays.c (gray_render_cubic): Replace the split
condition.
This commit fixes the changes from 2018-07-21, which broke charmap
iteration. We now add the default character as a new glyph with
index 0, thus increasing the number of glyphs by one (as before).
* src/pcf/pcfread.c (pcf_get_metrics): Adjust to new artificial
glyph with index 0.
Limit number of elements to 65534.
(pcf_get_bitmaps): Ditto.
Unify two loops into one; this avoids allocation of an intermediate
array.
(pcf_get_encodings): Don't flip indices but copy glyph metrics of
default character to index 0.
Also handle invalid default character.
* docs/CHANGES: Updated.
642bc7590c701c8cd35a9f60fa899cfa518b17ff introduced dynamically
allocated memory when parsing CFF files with the "old" engine. Bounds
checks have never been updated, however, leading to pointless
comparisons of pointers in some cases. This commit presents a
solution for bounds checks in the CFF module with an extended logic
for the "old" engine while staying as concise as possible for the
"new" one.
* src/cff/cffparse.h: Introduce the struct `CFF_T2_StringRec' and
the additional field `t2_strings' within `CFF_ParserRec'.
* src/cff/cffparse.c (cff_parser_within_limits): Move all boundary
checks into this new function and update the rest of `cffparse.c' to
use it.
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=12137
* src/autofit/afblue.dat: Use U+200D (ZERO-WIDTH JOINER) characters
to get medial forms for some Mongolian characters.
* src/autofit/afblue.c, src/autofit/afblue.h: Regenerated.
As a de-facto standard, layouts using this script are constructed
horizontally line by line, then the lines are rotated clockwise for
vertical display.
* src/autofit/afblue.dat: Add blue zone data for Mongolian.
* src/autofit/afblue.c, src/autofit/afblue.h: Regenerated.
* src/autofit/afscript.h: Add Mongolian standard characters.
* src/autofit/afranges.c, src/autofit/afstyles.h: Add Mongolian
data.
==========================
Tag sources with `VER-2-10-0'.
* docs/VERSION.TXT: Add entry for version 2.10.0.
* docs/CHANGES: Updated.
* README, Jamfile (RefDoc), src/base/ftver.rc,
builds/windows/vc2010/freetype.vcxproj,
builds/windows/vc2010/index.html,
builds/windows/visualc/freetype.dsp,
builds/windows/visualc/freetype.vcproj,
builds/windows/visualc/index.html,
builds/windows/visualce/freetype.dsp,
builds/windows/visualce/freetype.vcproj,
builds/windows/visualce/index.html,
builds/wince/vc2005-ce/freetype.vcproj,
builds/wince/vc2005-ce/index.html,
builds/wince/vc2008-ce/freetype.vcproj,
builds/wince/vc2008-ce/index.html: s/2.9.1/2.10.0/, s/291/2100/.
* include/freetype/freetype.h (FREETYPE_MINOR): Set to 10.
(FREETYPE_PATCH): Set to 0.
* builds/unix/configure.raw (version_info): Set to 23:0:17.
* CMakeLists.txt (VERSION_MINOR): Set to 10.
(VERSION_PATCH): Set to 0.
* builds/toplevel.mk (version, winversion): Since the minor version
number has two digits now, never omit the patch number. We would
get ambiguous zip file names otherwise.
(dist): Remove remnants of `docmaker' tool.
(do-dist): Remove unused intermediate files.
* src/cff/cffparse.c (destrict_c2s_item): Guard function with
CFF_CONFIG_OPTION_OLD_ENGINE macro.
This change should allow Freetype to work on WASM/Emscripten without
needing `-s EMULATE_FUNCTION_POINTER_CASTS=1'.
* src/autofit/afdummy.c (af_dummy_hints_apply): Fix signature.
* src/cid/cidload.c (cid_parse_font_matrix, parse_fd_array,
parse_expansion_factor, parse_font_name): Return `void', not
`FT_Error'.
* include/freetype/internal/ftobjs.h (FT_CMap_CharVarIsDefaultFunc):
Fix signature.
* src/psaux/psintrp.c (cf2_interpT2CharString): The call to
`cf2_arrstack_setCount' may fail because the allocator ran out of
memory. When this happens the stack is still written to before the
error condition is checked. This means that FreeType writes outside
of allocated memory. This commit moves the error check prior to the
stack assignment, hence the function now properly returns with an
error condition.
Instead of setting typo or win metrics as the new FT_Face metrics
indiscriminately, apply only typo deltas to the currently active FT_Face
metrics. This prevents line height differences when e.g. the default
outlines were used as the regular face and instances for everything else.
* src/truetype/ttgxvar.c (tt_apply_mvar): Implement.
If the OS/2 table exists and fsSelection bit 7 (USE_TYPO_METRICS) is set,
use the sTypo* set of values to compute the FT_Face's ascender, descender
and height. Otherwise, fall back to old behavior.
* src/sfnt/sfobjs.c (sfnt_load_face): Implement.
All of the Type 1 path building is done with code common to the
revised CFF engine, with the exception of closepath, which was still
calling ps_builder_close_contour(), thus previously cached segments
were not always written to the path, and glyph corruption, or even
invalid outlines were possible.
* src/psauc/psinterp.c (cf2_interpT2CharString) <cf2_cmdCLOSEPATH>:
Switch to calling `cf2_glyphpath_closeOpenPath'.
For multiple master fonts, common usage (in Postscript) is to modify
the WeightVector of an existing font instance, this addition
supports that use.
* include/freetype/ftmm.h, src/base/ftmm.c (FT_Set_MM_WeightVector,
FT_Get_MM_WeightVector): New API functions.
* include/freetype/internalservices/svmm.h
(FT_Set_MM_WeightVector_Func, FT_Get_MM_WeightVector_Func): New
function types.
(MultiMasters): Add `set_mm_weightvector' and `get_mm_weightvector'
members.
(FT_DEFINE_SERVICE_MULTIMASTERSREC): Updated.
* src/cffcffdrivr.c (cff_set_mm_weightvector,
cff_get_mm_weightvector): New functions.
(cff_service_multi_masters): Register them.
* src/truetype/ttdriver.c (tt_service_gx_multi_masters): Updated.
This driver doesn't use the new interface.
* src/type1/t1load.c (T1_Set_MM_WeightVector,
T1_Get_MM_WeightVector): New functions.
* src/type1/t1driver.c (t1_service_multi_masters): Register them.
* src/type1/t1load.h: Updated.
It's been more than 2^31 seconds since 1904.
* include/freetype/tttables.h (TT_Header): Change field types.
* src/sfnt/ttload.c (tt_face_load_generic_header): Updated.
This is a follow-up to the previous commit.
* src/cff/cffobjs.c (cff_face_init), src/sfnt/sfobjs.c
(sfnt_load_face), src/type1/t1objs.c (T1_Face_Init),
src/type42/t42objs.c (T42_Face_Init): Implement it.
* include/freetype/ftsystem.h: Include FT_TYPES_H.
(*FT_Alloc_Func, *FT_Realloc_Func): Use size_t for the size arguments.
* src/raster/ftmisc.h: Ditto.
* builds/amiga/src/base/ftsystem.c, builds/unix/ftsystem.c,
* builds/vms/ftsystem.c, src/base/ftsystem.c (ft_alloc, ft_realloc):
Use size_t for the size arguments.
* src/base/ftdbgmem.c (ft_mem_debug_alloc, ft_mem_debug_realloc): Use
FT_Offset, aka size_t, for the size arguments.
Before this commit we had code like
(FT_Bool)( globals->glyph_styles[gindex] & 0x8000)
Since `FT_Bool' is defined to be an `unsigned char', the code
evaluated to something like
(unsigned char)( 0x8532 & 0x8000)
which in turn expanded to
(unsigned char)( 0x8000)
and finally yielded 0x00 – i.e., false – not as expected.
Problem reported and analyzed by Tony Smith <tony.smith@macro4.com>.
* include/freetype/fttypes.h (FT_BOOL): Add a comparison against
zero so that we always have a Boolean expression.
*/*: Replace castings to `FT_Bool' with calls to `FT_BOOL' where
possible.
This makes FT_Get_Char_Index and FT_Get_Next_Char 4-5 times faster.
* src/bdf/bdfdrivr.c (bdf_cmap_char_{index,next}): Help binary search
with continuous prediction.
For bitmap fonts, `FT_Load_Glyph' should either return an error or
not set the format to `FT_GLYPH_FORMAT_OUTLINE'. However, in this
case `FT_Load_Glyph' calls into the auto-hinter which calls back
into `FT_Load_Glyph' with `FT_LOAD_NO_SCALE' in the flags, which
marks the glyph as `FT_GLYPH_FORMAT_OUTLINE' with an empty path
(even though it doesn't have any path). It appears that the
auto-hinter should not be called when the face doesn't have
outlines. The current test for using the auto-hinter in
`FT_Load_Glyph' checks if the driver supports scalable outlines, but
not if the face supports scalable outlines.
* src/base/ftobjs.c (FT_Load_Glyph): Directly check whether we have
scalable outlines.
PCF comes with charmap lookup table, aka PCF encodings. Using it
directly makes FT_Get_Char_Index and FT_Get_Next_Char 4-5 times
faster than the original BDF-like binary searches.
* src/pcf/pcf.h (PCF_EncodingRec): Removed.
(PCF_FaceRec): Remove `nencodings' and `encodings'.
* src/pcf/pcfdrivr.c (pcf_cmap_char_{index,next}): Replaced.
* src/pcf/pcfread.c (pcf_get_encodings): Store data differently.
* src/pcf/pcf.h (PCF_Face): Updated to include...
(PCF_EncRec): ... this new structure to store charmap geometry.
* src/pcf/pcfread.c (pcf_get_encodings): Store charmap geometry.
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=10317
* src/truetype/ttgxvar.c (ft_var_load_gvar): Properly exit memory
frame if we have invalid glyph variation data offsets.
(tt_face_vary_cvt): Protect against missing `tuplecoords' array.
Fix typo.
* src/sfnt/sfdriver (IS_WIN, IS_APPLE): Omit language ID checks.
(get_win_string, get_apple_string): Return NULL when the PostScript
font name characters is not according to specification.
(get_win_string): Make trace output work if the high byte if
non-zero.
(sfnt_get_var_ps_name, sfnt_get_ps_name): Previously we preferred
Win PS name (when there is also an Apple PS name); change this into
a fallback to Apple PS name in case the Win PS name is invalid.
These public API functions(!) were always undocumented and have
escaped all clean-up efforts until now.
* include/freetype/ftoutln.h (FT_Outline_New_Internal,
FT_Outline_Done_Internal): Removed.
* src/base/ftoutln.h (FT_Outline_New_Internal,
FT_Outline_Done_Internal): Merge into...
(FT_Outline_New, FT_Outline_Done): ... these functions.
* docs/README: Updated.
* include/freetype/internal/ftobjs.h (ft_glyphslot_preset_bitmap):
Change return type.
* src/base/ftobjs.c (ft_glyphslot_preset_bitmap): Return the bitmap
size assessment.
* src/raster/ftrend1.c (ft_raster1_render): Use it to refuse the
rendering of enourmous or far-fetched outlines.
* src/smooth/ftsmooth.c (ft_smooth_render_generic): Ditto.
* src/autofit/afhints.h (AF_PointRec) [FT_DEBUG_AUTOFIT]: New arrays
`before' and `after'.
* src/autofit/afhints.c (af_get_strong_edge_index): New auxiliary
function.
(af_glyph_hints_dump_points): Trace `before' and `after' edges.
(af_glyph_hints_align_strong_points) [FT_DEBUG_AUTOFIT]: Set
`before' and `after' information.
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=10080
* src/psaux/cffdecode.c (cff_decoder_parse_charstrings) <cff_op_roll>
[CFF_CONFIG_OPTION_OLD_ENGINE]: Use modulo for loop count, as
documented in the specification.
Triggered by
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=10054
* src/truetype/ttgload.c (load_truetype_glyph): Don't use variable
`opened_frame' to trace whether a frame must be closed at the end of
function: This fails because `TT_Vary_Apply_Glyph_Deltas' (which
gets called for space glyphs) uses a frame by itself. Instead,
close the frame after loading the header, then use another frame for
the remaining part of the glyph later on.
Also avoid calling `tt_get_metrics' twice under some circumstances.
* include/freetype/internal/ftcalc.h (ADD_INT64, SUB_INT64,
MUL_INT64, DIV_INT64) [FT_LONG64]: New macros.
* src/base/ftcalc.c (ft_corner_orientation) [FT_LONG64]: Use
`SUB_INT64' and `MUL_INT64'.
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=10028
* src/truetype/ttgload.c (ON_CURVE_POINT, X_SHORT_VECTOR,
Y_SHORT_VECTOR, REPEAT_FLAG, X_POSITIVE, SAME_X, Y_POSITIVE, SAME_Y,
OVERLAP_SIMPLE): New macros.
(TT_Load_Simple_Glyph): Use new macros to make code more readable.
Remove useless adjustment of `outline->tags' elements.
There is no harm except some umbiguity in broken fonts with duplicate
encodings.
* src/bdf/bdflib.c (_bdf_parse_glyphs): Remove duplicate tracking.
(_bdf_parse_t): Remove large `have' bitfield.
* src/bdf/bdf.h (bdf_font_t): Remove `nmod', `umod', and `modified',
which were set but never used.
* src/bdf/bdflib.c (_bdf_parse_{glyphs,properties}, bdf_load_font):
Updated accordingly.
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=9864
* src/psaux/cffdecode.c (cff_decoder_parse_charstrings)
<cff_op_random> [CFF_CONFIG_OPTION_OLD_ENGINE]: Use top dict's
`random' field directly if parsing dictionaries.
It sometimes makes sense to suppress tracing informations, for
example, if it outputs identical messages again and again.
* include/freetype/internal/ftdebug.h: Make `ft_trace_levels' a
pointer.
(FT_Trace_Disable, FT_Trace_Enable): New declarations.
* src/base/ftdebug.c (ft_trace_levels): Rename to...
(ft_trace_levels_enabled): ... this.
(ft_trace_levels_disabled): New array.
(ft_trace_levels): New pointer.
(FT_Trace_Disable, FT_Trace_Enable): Implement.
(ft_debug_init): Updated.
* src/base/ftobjs.c (pixel_modes): Move this array to top level
from ...
(FT_Load_Glyph): ... here.
(FT_Render_Glyph_Internal): Use `width' x `height' in trace message.
Use `pixel_modes'.
Treat all size related properties as unsigned values.
* src/pcf/pcf.h (PCF_ParsePropertyRec): Use unsigned `name' and
`value'.
* src/pcf/pcfread.c (pcf_get_propeerties, pcf_load_font): Updated
parsing code and handling of AVERAGE_WIDTH, POINT_SIZE, PIXEL_SIZE,
RESOLUTION_X and RESOLUTION_Y.
Unofficial specifications hesitate to use unsigned 32-bit integers.
Negative values caused a lot of trouble in the past and it is safer
and easier to treat some properties as unsigned.
* src/pcf/pcf.h (PCF_AccelRec): Use unsigned values for `fontAscent',
`fontDescent', and `maxOverlap'.
* src/pcf/pcfread.c (pcf_load_font, pcf_get_accel): Updated.
* src/pcf/pcfdrivr.c (PCF_Glyph_Load, PCF_Size_Select,
PCF_Size_Request): Updated.
* src/cid/cidload.c: Include FT_INTERNAL_POSTSCRIPT_AUX_H
(cid_load_keyword, cid_parse_font_matrix, parse_fd_array,
parse_expansion_factor, cid_parse_dict): Add tracing calls.
(parse_font_name): New function to trace `/FontName' keywords in
/FDArray dict.
(cid_field_records): Register `parse_font_name'.
* src/truetype/ttgxvar.c (tt_set_mm_blend): Correctly set
`face->doblend' if the current call to the function yields the same
blend coordinates as the previous call.
* src/base/ftutil.c (ft_mem_alloc, ft_mem_realloc): Only call
`FT_MEM_ZERO' if we actually have a buffer.
(ft_mem_dup): Only call `ft_memcpy' if we actually have a buffer.
We no longer use predefined _DLL, which can be defined for static
builds too with /MD. We use DLL_EXPORT and DLL_IMPORT instead,
following libtool convention.
* CMakeLists.txt [WIN32], builds/windows/vc2010/freetype.vcxproj:
Define DLL_EXPORT manually.
* include/freetype/config/ftconfig.h, builds/unix/ftconfig.in,
builds/vms/ftconfig.h, builds/windows/vc2010/index.html,
src/base/ftver.rc: /_DLL/d, s/FT2_DLLIMPORT/DLL_IMPORT/.
For Multiple Masters fonts We don't support intermediate designs;
this implies that
number_of_designs == 2 ^^ number_of_axes
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=9557
* src/type1/t1load.c (T1_Open_Face): Ensure above constraint.
(T1_Get_MM_Var): Remove now redundant test.
This code is taken from the type1 module.
Reported as
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=9510
* src/cid/cidload.c (parse_fd_array): Set some private dict default
values.
(cid_face_open): Do the sanitizing.
Fix some tracing messages.
This change makes the driver use the `defaultChar' property of PCF
files.
* src/pcf/pcf.h (PCF_FaceRec): Change type of `defaultChar' to
unsigned.
* src/pcf/pcfread.c (pcf_get_encodings): Read `defaultChar' as
unsigned.
Validate `defaultChar'.
If `defaultChar' doesn't point to glyph index zero, swap glyphs with
index zero and index `defaultChar' and adjust the encodings
accordingly.
* src/pcf/pcfdrivr.c (pcf_cmap_char_index, pcf_cmap_char_next,
PCF_Glyph_Load): Undo change from 2002-06-16 which always enforced
the first character in the font to be the default character.
This is a valid encoding tag for BDF, PCF, and Windows FNT, and
there is no reason to disallow it for these formats.
* src/base/ftobjs.c (FT_Select_Charmap): Implement it.
Taken from patch #9667, written by Steve Langasek
<vorlon@debian.org>.
This fixes a build failure (most probably a bug in gcc) on ppc64el
when building with -O3.
It turned out that we incorrectly round CVT and glyph point deltas
before accumulation, leading to severe positioning errors if there
are many delta values to sum up.
Problem reported by Akiem Helmling <akiem@underware.nl> and analyzed
by Behdad.
* src/truetype/ttgxvar.c (ft_var_readpackeddelta): Return deltas in
16.16 format.
(tt_face_var_cvt): Collect deltas in `cvt_deltas', which is a 16.16
format array, and add the accumulated values to the CVT at the end
of the function.
(TT_Vary_Apply_Glyph_Deltas): Store data in `points_org' and
`points_out' in 16.16 format.
Collect deltas in `point_deltas_x' and `point_deltas_y', which are
16.16 format arrays, and add the accumulated values to the glyph
coordinates at the end of the function.
* src/base/ftcalc.c (FT_Matrix_Check): New base function to properly
reject degenerate font matrices.
* include/freetype/internal/ftcalc.h: Updated.
* src/cff/cffparse.c (cff_parse_font_matrix), src/cid/cidload.c
(cid_parse_font_matrix), src/type1/t1load.c (t1_parse_font_matrix),
src/type42/t42parse.c (t42_parse_font_matrix): Use
`FT_Matrix_Check'.
Reported by Alexei.
* src/sfnt/ttcolr.c (tt_face_get_colr_layer): Return immediately if
`colr' is NULL.
* src/sfnt/ttcpal.c (tt_face_palette_set): Return immediately, if
`cpal' is NULL.
This function facilitates access to full capabilities of FreeType
rendering engine for custom glyphs. This can be quite useful for
consistent rendering of mathematical and chemical formulas, e.g.
https://bugs.chromium.org/p/chromium/issues/detail?id=757078
* include/freetype/ftglyph.h, src/base/ftglyph.c (FT_New_Glyph): New
function.
bdflib.c:1562 could be reached with `font->glyphs_used == 0'. That
caused an underflow of the unsigned value which results in undefined
behaviour.
* src/bdf/bdflib.c (src/bdf/bdflib.c): Bail out earlier than before
if the `ENCODING' keyword cannot be found.
Still missing: Support for negative bitmap pitch and subpixel offset
of source bitmap.
* include/freetype/ftbitmap.h, src/base/ftbitmap.c
(FT_Bitmap_Blend): New function.
This solution doesn't store any data in an `FT_GlyphSlot' object.
* include/freetype/freetype.h (FT_LayerIterator): New structure.
(FT_Get_Color_Glyph_Layer): New function.
* include/freetype/internal/sfnt.h (TT_Get_Colr_Layer_Func): New
function type.
(SFNT_Interface, FT_DEFINE_SFNT_INTERFACE): Add it.
* src/base/ftobjs.c (FT_Get_Color_Glyph_Layer): Implement it.
* src/sfnt/ttcolr.c (tt_face_get_colr_layer): New function.
* src/sfnt/ttcolr.h: Updated.
* src/sfnt/sfdriver.c (sfnt_interface): Updated.
* src/sfnt/sfdriver.c: Include `ttcpal.h'.
* src/sfnt/sfnt.c: Include `ttcpal.c'.
* src/sfnt/ttcolr.c, src/sfnt/ttcolr.h: Move CPAL stuff to ...
* src/sfnt/ttcpal.c, src/sfnt/ttcpal.c: ... these new files.
* src/sfnt/Jamfile (_sources), src/sfnt/rules.mk (SFNT_DRV_SRC):
Updated.
* include/freetype/internal/fttrace.h: Add support for `colr' and
`cpal'.
Sort entries.
* src/sfnt/ttcolr.c (tt_face_find_color): Removed.
(tt_face_colr_blend_layer): Use `face->palette' instead of calling
`tt_face_find_color'.
Use and set text foreground color.
* src/sfnt/ttcolr.c (tt_face_palette_set): New function.
(tt_face_load_colr): Allocate `face->palette' and call
`tt_face_palette_set'.
Adjust return error code in case of error.
* src/sfnt/ttcolr.h: Updated.
* include/freetype/internal/sfnt.h (TT_Set_Palette_Func): New
function type.
(SFNT_Interface, FT_DEFINE_SFNT_INTERFACE): Add it.
* src/sfnt/sfdriver.c (sfnt_interface), src/sfnt/sfobjs.c
(sfnt_done_face): Updated.
* src/raster/ftraster.c (black_TWorker, SCALED, Set_High_Precision):
Clean up after 5-level gray removal (8dc8635874).
(Vertical_Sweep_Span): Be brief.
* include/freetype/internal.tttypes.h: Include FT_COLOR_H.
(TT_FaceRec): Add `palette' field.
* src/sfnt/ttcolr.c: Include FT_COLOR_H.
(Cpal): Remove all data covered by the new `palette' field in
`TT_FaceRec'.
(tt_face_load_colr): Updated.
Read `CPAL' version 1 data.
(tt_face_load_colr_layers, tt_face_find_color): Updated.
* src/sfnt/sfobjs.c (sfnt_done_face): Free glyph color palette data.
This introduces `FT_Library_SetLcdGeometry' for setting up arbitrary
LCD subpixel geometry including non-striped patterns.
* src/base/ftlcdfil.c (FT_Library_SetLcdGeometry): New function.
* include/freetype/ftlcdfil.h: Document it.
* include/freetype/freetype.h: Minor.
* include/freetype/ftchapters.h: Minor.
* src/raster/ftraster.c [STANDALONE] (FT_Outline_Get_CBox): Add
function.
[!STANDALONE]: Include FT_OUTLINE_H.
(ft_black_render): Compute CBox and reject glyphs larger than
0xFFFF x 0xFFFF.
* src/smooth/ftgrays.c (gray_raster_render): Reject glyphs larger
than 0xFFFF x 0xFFFF.
