Commit Graph

1138 Commits

Author SHA1 Message Date
Tom Kacvinsky a006a0f5ec 2001-10-21 Tom Kacvinsky <tjk@ams.org>
* src/cff/cffload.c (CFF_Done_Font): Free subfonts array only if
        we are working with a CID keyed CFF font.  Otherwise, a variable
        that was never allocated memory might freed.  This is a correction
        to the previous patch for freeing subfonts.
2001-10-21 20:30:54 +00:00
Tom Kacvinsky dfa4568432 In function CFF_Done_Font, make sure that subfonts are freed only if
we are working with a CID keyed CFF font.  This is to avoid freeing a
variable that might not have been allocated.
2001-10-21 20:26:59 +00:00
Tom Kacvinsky 726a4fa6f1 2001-10-21 Tom Kacvinsky <tjk@ams.org>
* src/cff/cffload.c (CFF_Done_Font): Free the subfonts array to
        avoid a memory leak.
2001-10-21 16:30:59 +00:00
Tom Kacvinsky 98289a2a65 In function CFF_Done_Font, make sure to free the subfonts array to
avoid a memory leak.
2001-10-21 16:29:16 +00:00
David Turner 12d40dac42 * src/type1/t1load.c: fixed memory leak bugs that appeared with certain
synthetized fonts (e.g. CourierBoldOblique, a.k.a. "cobo_____.pfb")

	* src/pshinter/pshalgo2.c, src/pshinter/pshalgo1.c,
	src/pshinter/pshglob.c: removing compiler warnings in pedantic modes
	(in multi-object compilation mode, mainly)
2001-10-21 15:41:11 +00:00
Tom Kacvinsky e6eaadf683 Added another entry for t1load.c related to synthetic fonts: make sure
that custom encodings are not loaded twice (thus avoiding a memory
leak).
2001-10-20 20:49:58 +00:00
Tom Kacvinsky 629bf286e7 In function parse_encoding, add a test to make sure that custom
encodings (i.e., neither StandardEncoding nor ExpertEncoding) are not
loaded twice when the Type 1 font is synthetic.
2001-10-20 20:46:36 +00:00
Tom Kacvinsky 3beab5bd13 Added entries for changes to t1load.c. 2001-10-20 17:53:22 +00:00
Tom Kacvinsky 48f26bc748 In functions parse_font_name and parse_subrs, added a test for when
loading synthetic fonts to make sure that the font name and
subrotuines are not loaded twice.  This is to remove a memory leak
that occurs because the original allocation for these objects is not
deallocated when the objects are read the second time.
2001-10-20 17:44:48 +00:00
Werner Lemberg 620ec14955 renaming todo -> TODO 2001-10-20 01:29:38 +00:00
David Turner 7e4b52d17b getting rid of compiler warnings 2001-10-19 09:17:49 +00:00
David Turner 3b9f1e096d another small test of the CVS loginfo setup 2001-10-19 08:15:28 +00:00
David Turner 0d7ccf1d4e adding or updating files to the old build system to compile
the library with the Postscript hinter correctly..
2001-10-19 07:34:20 +00:00
David Turner b3e1b292cd just testing the CVS loginfo setup 2001-10-18 15:02:47 +00:00
David Turner 043be66c15 just testing the CVS loginfo setup 2001-10-18 14:58:45 +00:00
David Turner b347aaa206 just testing the CVS loginfo setup 2001-10-18 14:57:28 +00:00
David Turner ec3c28a847 just testing the CVS loginfo setup 2001-10-18 14:54:49 +00:00
David Turner 8e005d502d just testing the CVS loginfo setup 2001-10-18 14:50:01 +00:00
David Turner 6043d2bc5d just testing the CVS loginfo setup 2001-10-18 14:49:12 +00:00
David Turner d5fb3512b0 just testing the CVS loginfo setup 2001-10-18 14:45:28 +00:00
David Turner 0ad006b76f just testing the CVS loginfo setup 2001-10-18 14:36:46 +00:00
David Turner 12c6830662 just testing the CVS loginfo setup 2001-10-18 14:35:52 +00:00
David Turner 8d6aae02d8 just testing the CVS loginfo setup 2001-10-18 14:33:11 +00:00
David Turner 835368543f just testing the CVS loginfo setup 2001-10-18 14:31:41 +00:00
David Turner a0b314c044 just testing the CVS loginfo setup 2001-10-18 14:27:43 +00:00
David Turner 3b25cc7b03 just testing the CVS loginfo setup 2001-10-18 14:25:38 +00:00
David Turner 6ea520aad4 just testing the CVS loginfo setup 2001-10-18 14:12:44 +00:00
David Turner 3cb2500966 just testing the CVS loginfo setup 2001-10-18 14:10:23 +00:00
David Turner 5a4c1ff381 just testing the CVS loginfo setup 2001-10-18 14:07:06 +00:00
David Turner 02f3b4889c just testing the CVS loginfo setup 2001-10-18 14:02:15 +00:00
David Turner 730d0540a4 just testing the CVS loginfo setup 2001-10-18 14:00:54 +00:00
David Turner 0ebf28c73e just testing the CVS loginfo setup 2001-10-18 13:58:05 +00:00
David Turner 29d5a0ecbd just testing the CVS loginfo setup 2001-10-18 13:52:47 +00:00
David Turner afa3854202 documentation update 2001-10-18 13:22:01 +00:00
David Turner f373e2d9bc rewrote "glnames.py" to fix a rather nasty bug that made FreeType
return incorrect glyph names for certain glyph indices..
2001-10-18 13:20:01 +00:00
David Turner c9349c9343 updating ChangeLog and Jamfile for the new Postscript hinter 2001-10-18 11:59:08 +00:00
David Turner c55099de40 changing pixel size computations for better accuracy (experimental) 2001-10-18 11:57:52 +00:00
David Turner 540e7cd6c6 Adding glyph hinter debugging program.
WARNING:
You cannot compile this program without a (yet) unavailable
third-party library named "Nirvana".. this is reserved for FreeType
development exclusively..
2001-10-18 11:50:31 +00:00
David Turner 5893c1bb2a add support for new postscript hinter 2001-10-18 11:49:26 +00:00
David Turner 573a98d048 add support for new postscript hinter 2001-10-18 11:46:25 +00:00
David Turner a83bc088ec adding new postscript hinter 2001-10-18 11:38:43 +00:00
David Turner 0e49a4b8e7 adding support to new postscript hinter 2001-10-18 11:26:21 +00:00
David Turner a1471038bd fixing a "bug" that affected the Postscript hinter: growing a memory block
though "realloc" didn't clear the new bytes of memory..
2001-10-18 11:23:51 +00:00
David Turner 39bea028f8 fixed typo 2001-10-18 10:27:56 +00:00
David Turner 1c5802a4b2 updated FT_Div64by32 2001-10-18 09:51:09 +00:00
David Turner ea5a981c7d minimal optimisation submitted by Graham Asher 2001-10-17 14:29:51 +00:00
David Turner 034a19326b documentation update for 2.0.5 2001-10-17 13:57:02 +00:00
David Turner 23bcde193e implementing a new function named "FT_Get_Postscript_Name" to
retrieve a face's "unique" Postscript name
2001-10-17 13:48:10 +00:00
David Turner 40d006af7c tagging files, checking new CVS -> mail script 2001-10-17 11:33:01 +00:00
Werner Lemberg 80b8d77024 Fixed a bug in `glnames.py' that prevented it from generating
correct glyph names tables.  This resulted in the unavailability of
certain glyphs like `Cacute', `cacute' and `lslash' in Unicode
charmaps, even if these were present in the font (causing problems
for Polish users).

* src/tools/glnames.py (mac_standard_names): Fixed.
(t1_standard_strings): Some fixes and renamed to ...
(sid_standard_names): This.
(t1_expert_encoding): Fixed.
(the_adobe_glyph_list): Renamed to ...
(adobe_glyph_names): This.
(the_adobe_glyphs): Renamed to ...
(adobe_glyph_values): This.
(dump_mac_indices, dump_glyph_list, dump_unicode_values, main):
Updated.
* src/psnames/pstables.h: Regenerated.
* src/psnames/psmodule.c (PS_Unicode_Value): Fix offset.
Fix return value.
Use `sid_standard_table' and `ps_names_to_unicode' instead of
`t1_standard_glyphs' and `names_to_unicode'.
(PS_Macintosh_Name): Use `ps_glyph_names' instead of
`standard_glyph_names'.
(PS_Standard_Strings): Use `sid_standard_names' instead of
`t1_standard_glyphs'.
* doc/BUGS, doc/TODO: New documents.
* src/cache/ftlru.c (FT_Lru_Lookup_Node): Fixed a bug that prevented
correct LRU behaviour.
setjmp() and longjmp() are now used for rollback (i.e. when memory
pool overflow occurs).
Function names are now all uniformly prefixed with `gray_'.

* src/smooth/ftgrays.c: Include <setjmp.h>.
(ErrRaster_MemoryOverflow): New macro.
(TArea): New type to store area values in each cell (using `int' was
too small on 16-bit systems).  <limits.h> is included to properly
get the needed data type.
(TCell, TRaster): Use it.
(TRaster): New element `jump_buffer'.
(gray_compute_cbox): Use `RAS_ARG' as the only parameter and get
`outline' from it.
(gray_record_cell): Use longjmp().
(gray_set_cell): Use gray_record_cell() for error handling.
(gray_render_line, gray_render_conic, gray_render_cubic): Simplify.
(gray_convert_glyph_inner): New function, using setjmp().
(gray_convert_glyph): Use it.


Provide a public API to manage multiple size objects for a given
FT_Face in the new header file `ftsizes.h'.

* include/freetype/ftsizes.h: New header file,
* include/freetype/internal/ftobjs.h: Use it.
Remove declarations of FT_New_Size and FT_Done_Size (moved to
ftsizes.h).
* include/freetype/config/ftheader.h (FT_SIZES_H): New macro.
* src/base/ftobjs.c (FT_Activate_Size): New function.
* src/cache/ftcmanag.c: Include ftsizes.h.
(ftc_manager_init_size, ftc_manager_flush_size): Use
FT_Activate_Size.
2001-10-10 19:56:42 +00:00