walkero
/
fontconfig
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
fontconfig/src/fcfreetype.c

2960 lines
96 KiB
C
Raw Normal View History

Initial revision
2002-02-15 00:34:13 +01:00
/*
Replace RCS Id tags with the file name
2008-08-12 22:34:24 +02:00
* fontconfig/src/fcfreetype.c
Initial revision
2002-02-15 00:34:13 +01:00
*
Change files from ISO-Latin-1 to UTF-8
2004-12-07 02:14:46 +01:00
* Copyright © 2001 Keith Packard
Initial revision
2002-02-15 00:34:13 +01:00
*
* Permission to use, copy, modify, distribute, and sell this software and its
* documentation for any purpose is hereby granted without fee, provided that
* the above copyright notice appear in all copies and that both that
* copyright notice and this permission notice appear in supporting
Cleanup copyright notices to replace "Keith Packard" with "the author(s)"
2010-11-10 22:45:42 +01:00
* documentation, and that the name of the author(s) not be used in
Initial revision
2002-02-15 00:34:13 +01:00
* advertising or publicity pertaining to distribution of the software without
Cleanup copyright notices to replace "Keith Packard" with "the author(s)"
2010-11-10 22:45:42 +01:00
* specific, written prior permission. The authors make no
Initial revision
2002-02-15 00:34:13 +01:00
* representations about the suitability of this software for any purpose. It
* is provided "as is" without express or implied warranty.
*
Replace 'KEITH PACKARD' with 'THE AUTHOR(S)' in license text in all files
2009-03-12 21:00:08 +01:00
* THE AUTHOR(S) DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
Initial revision
2002-02-15 00:34:13 +01:00
* INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO
Replace 'KEITH PACKARD' with 'THE AUTHOR(S)' in license text in all files
2009-03-12 21:00:08 +01:00
* EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY SPECIAL, INDIRECT OR
Initial revision
2002-02-15 00:34:13 +01:00
* CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
* DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
* TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
* PERFORMANCE OF THIS SOFTWARE.
*/
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
/*
Change files from ISO-Latin-1 to UTF-8
2004-12-07 02:14:46 +01:00
Copyright © 2002-2003 by Juliusz Chroboczek
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
*/
Include $(top_srcdir), $(top_srcdir)/src before anything else. Shuffle order of includes for building out of srcdir on win32. reviewed by: plam
2006-04-25 07:57:41 +02:00
#include "fcint.h"
Remove freetype requirement for build-time applications. This avoids requiring the freetype development files when cross compiling
2008-01-02 17:47:14 +01:00
#include "fcftint.h"
Initial revision
2002-02-15 00:34:13 +01:00
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
Switch to FreeType 2.1.7 style includes. Bug #150. reviewed by: Keith Packard <keithp@keithp.com>
2003-11-18 08:53:04 +01:00
#include <ft2build.h>
#include FT_FREETYPE_H
#include FT_TRUETYPE_TABLES_H
#include FT_SFNT_NAMES_H
#include FT_TRUETYPE_IDS_H
#include FT_TYPE1_TABLES_H
Create FC_FONTFORMAT from FT_Get_X11_Font_Format function where available. This provides font file format information (BDF, Type 1, PCF, TrueType) for each font. Closes #109.
2004-12-05 01:26:06 +01:00
#if HAVE_FT_GET_X11_FONT_FORMAT
#include FT_XFREE86_H
#endif
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#if HAVE_FT_GET_BDF_PROPERTY
Switch to FreeType 2.1.7 style includes. Bug #150. reviewed by: Keith Packard <keithp@keithp.com>
2003-11-18 08:53:04 +01:00
#include FT_BDF_H
#include FT_MODULE_H
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
#endif
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
#include "ftglue.h"
Avoid #warning directives on non-GCC compilers. (bug 7683) Detect GCC and use #warning only on GCC systems.
2006-09-01 21:36:31 +02:00
#if HAVE_WARNING_CPP_DIRECTIVE
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#if !HAVE_FT_GET_BDF_PROPERTY
Add FcResultOutOfMemory to provide an accurate error when FcFontSetMatch fails in this way Make #warning about lacking various FreeType features indicate which version those features appeared so users know how to fix the problem (Thanks to Anton Tropashko)
2004-05-29 21:32:41 +02:00
#warning "No FT_Get_BDF_Property: Please install freetype 2.1.4 or later"
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#endif
#if !HAVE_FT_GET_PS_FONT_INFO
Add FcResultOutOfMemory to provide an accurate error when FcFontSetMatch fails in this way Make #warning about lacking various FreeType features indicate which version those features appeared so users know how to fix the problem (Thanks to Anton Tropashko)
2004-05-29 21:32:41 +02:00
#warning "No FT_Get_PS_Font_Info: Please install freetype 2.1.1 or later"
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#endif
Avoid #warning directives on non-GCC compilers. (bug 7683) Detect GCC and use #warning only on GCC systems.
2006-09-01 21:36:31 +02:00
#endif
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
Switch to RFC 3066 based lang names
2002-07-07 01:47:44 +02:00
/*
Update iso639-2 language coverage info, fix Georgian orthography to eliminate Mingrelian and Svan glyphs, use coverage for inclusion and OS/2 for Han exclusion, restructure fclang.c to use fclang.h from fc-lang dir
2002-07-08 09:31:53 +02:00
* Keep Han languages separated by eliminating languages
* that the codePageRange bits says aren't supported
Switch to RFC 3066 based lang names
2002-07-07 01:47:44 +02:00
*/
Update iso639-2 language coverage info, fix Georgian orthography to eliminate Mingrelian and Svan glyphs, use coverage for inclusion and OS/2 for Han exclusion, restructure fclang.c to use fclang.h from fc-lang dir
2002-07-08 09:31:53 +02:00
Initial revision
2002-02-15 00:34:13 +01:00
static const struct {
Eliminate relocations from FcCodePageRange structure (bug 10982). FcCodePageRange was using char pointers; replace them with char arrays.
2007-10-18 18:48:31 +02:00
char bit;
const FcChar8 lang[6];
Initial revision
2002-02-15 00:34:13 +01:00
} FcCodePageRange[] = {
Eliminate relocations from FcCodePageRange structure (bug 10982). FcCodePageRange was using char pointers; replace them with char arrays.
2007-10-18 18:48:31 +02:00
{ 17, "ja" },
{ 18, "zh-cn" },
{ 19, "ko" },
{ 20, "zh-tw" },
Initial revision
2002-02-15 00:34:13 +01:00
};
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_CODE_PAGE_RANGE (int) (sizeof FcCodePageRange / sizeof FcCodePageRange[0])
Initial revision
2002-02-15 00:34:13 +01:00
Switch to RFC 3066 based lang names
2002-07-07 01:47:44 +02:00
FcBool
Update iso639-2 language coverage info, fix Georgian orthography to eliminate Mingrelian and Svan glyphs, use coverage for inclusion and OS/2 for Han exclusion, restructure fclang.c to use fclang.h from fc-lang dir
2002-07-08 09:31:53 +02:00
FcFreeTypeIsExclusiveLang (const FcChar8 *lang)
Switch to RFC 3066 based lang names
2002-07-07 01:47:44 +02:00
{
Update iso639-2 language coverage info, fix Georgian orthography to eliminate Mingrelian and Svan glyphs, use coverage for inclusion and OS/2 for Han exclusion, restructure fclang.c to use fclang.h from fc-lang dir
2002-07-08 09:31:53 +02:00
int i;
Add strong/weak pattern value binding, add known charsets for automatic lang computation
2002-06-29 22:31:02 +02:00
Update iso639-2 language coverage info, fix Georgian orthography to eliminate Mingrelian and Svan glyphs, use coverage for inclusion and OS/2 for Han exclusion, restructure fclang.c to use fclang.h from fc-lang dir
2002-07-08 09:31:53 +02:00
for (i = 0; i < NUM_CODE_PAGE_RANGE; i++)
{
Treat zh-hk fonts differently from zh-tw fonts. This patch may cause fontconfig to treat A-X fonts differently from A-Y fonts; please mail the fontconfig list if this causes any problems. reviewed by: plam
2006-01-25 03:54:37 +01:00
if (FcLangCompare (lang, FcCodePageRange[i].lang) == FcLangEqual)
Switch to RFC 3066 based lang names
2002-07-07 01:47:44 +02:00
return FcTrue;
Update iso639-2 language coverage info, fix Georgian orthography to eliminate Mingrelian and Svan glyphs, use coverage for inclusion and OS/2 for Han exclusion, restructure fclang.c to use fclang.h from fc-lang dir
2002-07-08 09:31:53 +02:00
}
Switch to RFC 3066 based lang names
2002-07-07 01:47:44 +02:00
return FcFalse;
}
Add strong/weak pattern value binding, add known charsets for automatic lang computation
2002-06-29 22:31:02 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
typedef struct {
Add some more consts, fixing more GCC4 warnings.
2005-10-22 17:12:05 +02:00
const FT_UShort platform_id;
const FT_UShort encoding_id;
Convert fromcode to char[12] from char *.
2005-09-23 04:08:40 +02:00
const char fromcode[12];
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
} FcFtEncoding;
#define TT_ENCODING_DONT_CARE 0xffff
#define FC_ENCODING_MAC_ROMAN "MACINTOSH"
static const FcFtEncoding fcFtEncoding[] = {
Accept TT_PLATFORM_MICROSOFT, TT_MS_ID_SYMBOL_CS from name table The OT spec says: "When building a Unicode font for Windows, the platform ID should be 3 and the encoding ID should be 1. When building a symbol font for Windows, the platform ID should be 3 and the encoding ID should be 0." We were ignoring the SYMBOL_CS entry before. It's UTF-16/UCS-2 like the UNICODE_CS. Also, always use UTF-16BE instead of UCS-2BE. The conversion was doing UTF-16BE anyway.
2010-04-07 18:47:37 +02:00
{ TT_PLATFORM_APPLE_UNICODE, TT_ENCODING_DONT_CARE, "UTF-16BE" },
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{ TT_PLATFORM_MACINTOSH, TT_MAC_ID_ROMAN, "MACINTOSH" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_ID_JAPANESE, "SJIS" },
Accept TT_PLATFORM_MICROSOFT, TT_MS_ID_SYMBOL_CS from name table The OT spec says: "When building a Unicode font for Windows, the platform ID should be 3 and the encoding ID should be 1. When building a symbol font for Windows, the platform ID should be 3 and the encoding ID should be 0." We were ignoring the SYMBOL_CS entry before. It's UTF-16/UCS-2 like the UNICODE_CS. Also, always use UTF-16BE instead of UCS-2BE. The conversion was doing UTF-16BE anyway.
2010-04-07 18:47:37 +02:00
{ TT_PLATFORM_MICROSOFT, TT_MS_ID_SYMBOL_CS, "UTF-16BE" },
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{ TT_PLATFORM_MICROSOFT, TT_MS_ID_UNICODE_CS, "UTF-16BE" },
{ TT_PLATFORM_MICROSOFT, TT_MS_ID_SJIS, "SJIS-WIN" },
[fcfreetype] Fix typo in GB2312 encoding name string (#19845) This was causing failure when opening iconv converter, hence rendering GB2312-encoded fonts with no other usable encoding unusable.
2009-01-31 09:38:19 +01:00
{ TT_PLATFORM_MICROSOFT, TT_MS_ID_GB2312, "GB2312" },
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{ TT_PLATFORM_MICROSOFT, TT_MS_ID_BIG_5, "BIG-5" },
{ TT_PLATFORM_MICROSOFT, TT_MS_ID_WANSUNG, "Wansung" },
Josselin Mouette: Include 2.3 release information in changelog Add Josselin Mouette as an Uploader Set hinting_type to low priority configuration option Manish Singh: yes_bitmaps.conf -> yes-bitmaps.conf Funda Wang: Johap -> Johab
2005-03-05 21:58:39 +01:00
{ TT_PLATFORM_MICROSOFT, TT_MS_ID_JOHAB, "Johab" },
Accept TT_PLATFORM_MICROSOFT, TT_MS_ID_SYMBOL_CS from name table The OT spec says: "When building a Unicode font for Windows, the platform ID should be 3 and the encoding ID should be 1. When building a symbol font for Windows, the platform ID should be 3 and the encoding ID should be 0." We were ignoring the SYMBOL_CS entry before. It's UTF-16/UCS-2 like the UNICODE_CS. Also, always use UTF-16BE instead of UCS-2BE. The conversion was doing UTF-16BE anyway.
2010-04-07 18:47:37 +02:00
{ TT_PLATFORM_MICROSOFT, TT_MS_ID_UCS_4, "UTF-16BE" },
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{ TT_PLATFORM_ISO, TT_ISO_ID_7BIT_ASCII, "ASCII" },
Accept TT_PLATFORM_MICROSOFT, TT_MS_ID_SYMBOL_CS from name table The OT spec says: "When building a Unicode font for Windows, the platform ID should be 3 and the encoding ID should be 1. When building a symbol font for Windows, the platform ID should be 3 and the encoding ID should be 0." We were ignoring the SYMBOL_CS entry before. It's UTF-16/UCS-2 like the UNICODE_CS. Also, always use UTF-16BE instead of UCS-2BE. The conversion was doing UTF-16BE anyway.
2010-04-07 18:47:37 +02:00
{ TT_PLATFORM_ISO, TT_ISO_ID_10646, "UTF-16BE" },
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{ TT_PLATFORM_ISO, TT_ISO_ID_8859_1, "ISO-8859-1" },
};
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_FC_FT_ENCODING (int) (sizeof (fcFtEncoding) / sizeof (fcFtEncoding[0]))
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
typedef struct {
Add some more consts, fixing more GCC4 warnings.
2005-10-22 17:12:05 +02:00
const FT_UShort platform_id;
const FT_UShort language_id;
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
const char lang[8];
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
} FcFtLanguage;
#define TT_LANGUAGE_DONT_CARE 0xffff
static const FcFtLanguage fcFtLanguage[] = {
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
{ TT_PLATFORM_APPLE_UNICODE, TT_LANGUAGE_DONT_CARE, "" },
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ENGLISH, "en" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_FRENCH, "fr" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_GERMAN, "de" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ITALIAN, "it" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_DUTCH, "nl" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SWEDISH, "sv" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SPANISH, "es" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_DANISH, "da" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_PORTUGUESE, "pt" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_NORWEGIAN, "no" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_HEBREW, "he" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_JAPANESE, "ja" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ARABIC, "ar" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_FINNISH, "fi" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_GREEK, "el" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ICELANDIC, "is" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MALTESE, "mt" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TURKISH, "tr" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_CROATIAN, "hr" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_CHINESE_TRADITIONAL, "zh-tw" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_URDU, "ur" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_HINDI, "hi" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_THAI, "th" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_KOREAN, "ko" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_LITHUANIAN, "lt" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_POLISH, "pl" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_HUNGARIAN, "hu" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ESTONIAN, "et" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_LETTISH, "lv" },
/* { TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SAAMISK, ??? */
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_FAEROESE, "fo" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_FARSI, "fa" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_RUSSIAN, "ru" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_CHINESE_SIMPLIFIED, "zh-cn" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_FLEMISH, "nl" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_IRISH, "ga" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ALBANIAN, "sq" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ROMANIAN, "ro" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_CZECH, "cs" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SLOVAK, "sk" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SLOVENIAN, "sl" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_YIDDISH, "yi" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SERBIAN, "sr" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MACEDONIAN, "mk" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_BULGARIAN, "bg" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_UKRAINIAN, "uk" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_BYELORUSSIAN, "be" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_UZBEK, "uz" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_KAZAKH, "kk" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_AZERBAIJANI, "az" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_AZERBAIJANI_CYRILLIC_SCRIPT, "az" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_AZERBAIJANI_ARABIC_SCRIPT, "ar" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ARMENIAN, "hy" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_GEORGIAN, "ka" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MOLDAVIAN, "mo" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_KIRGHIZ, "ky" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TAJIKI, "tg" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TURKMEN, "tk" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MONGOLIAN, "mo" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MONGOLIAN_MONGOLIAN_SCRIPT,"mo" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MONGOLIAN_CYRILLIC_SCRIPT, "mo" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_PASHTO, "ps" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_KURDISH, "ku" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_KASHMIRI, "ks" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SINDHI, "sd" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TIBETAN, "bo" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_NEPALI, "ne" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SANSKRIT, "sa" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MARATHI, "mr" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_BENGALI, "bn" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ASSAMESE, "as" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_GUJARATI, "gu" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_PUNJABI, "pa" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ORIYA, "or" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MALAYALAM, "ml" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_KANNADA, "kn" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TAMIL, "ta" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TELUGU, "te" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SINHALESE, "si" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_BURMESE, "my" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_KHMER, "km" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_LAO, "lo" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_VIETNAMESE, "vi" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_INDONESIAN, "id" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TAGALOG, "tl" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MALAY_ROMAN_SCRIPT, "ms" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MALAY_ARABIC_SCRIPT, "ms" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_AMHARIC, "am" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TIGRINYA, "ti" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_GALLA, "om" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SOMALI, "so" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SWAHILI, "sw" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_RUANDA, "rw" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_RUNDI, "rn" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_CHEWA, "ny" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MALAGASY, "mg" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_ESPERANTO, "eo" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_WELSH, "cy" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_BASQUE, "eu" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_CATALAN, "ca" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_LATIN, "la" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_QUECHUA, "qu" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_GUARANI, "gn" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_AYMARA, "ay" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TATAR, "tt" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_UIGHUR, "ug" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_DZONGKHA, "dz" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_JAVANESE, "jw" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SUNDANESE, "su" },
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
#if 0 /* these seem to be errors that have been dropped */
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SCOTTISH_GAELIC },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_IRISH_GAELIC },
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
#endif
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
/* The following codes are new as of 2000-03-10 */
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_GALICIAN, "gl" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_AFRIKAANS, "af" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_BRETON, "br" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_INUKTITUT, "iu" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_SCOTTISH_GAELIC, "gd" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_MANX_GAELIC, "gv" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_IRISH_GAELIC, "ga" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_TONGAN, "to" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_GREEK_POLYTONIC, "el" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_GREELANDIC, "ik" },
{ TT_PLATFORM_MACINTOSH, TT_MAC_LANGID_AZERBAIJANI_ROMAN_SCRIPT,"az" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_SAUDI_ARABIA, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_IRAQ, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_EGYPT, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_LIBYA, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_ALGERIA, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_MOROCCO, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_TUNISIA, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_OMAN, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_YEMEN, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_SYRIA, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_JORDAN, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_LEBANON, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_KUWAIT, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_UAE, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_BAHRAIN, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_QATAR, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_BULGARIAN_BULGARIA, "bg" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CATALAN_SPAIN, "ca" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CHINESE_TAIWAN, "zh-tw" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CHINESE_PRC, "zh-cn" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CHINESE_HONG_KONG, "zh-hk" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CHINESE_SINGAPORE, "zh-sg" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CHINESE_MACAU, "zh-mo" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CZECH_CZECH_REPUBLIC, "cs" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_DANISH_DENMARK, "da" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GERMAN_GERMANY, "de" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GERMAN_SWITZERLAND, "de" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GERMAN_AUSTRIA, "de" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GERMAN_LUXEMBOURG, "de" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GERMAN_LIECHTENSTEI, "de" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GREEK_GREECE, "el" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_UNITED_STATES, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_UNITED_KINGDOM, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_AUSTRALIA, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_CANADA, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_NEW_ZEALAND, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_IRELAND, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_SOUTH_AFRICA, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_JAMAICA, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_CARIBBEAN, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_BELIZE, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_TRINIDAD, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_ZIMBABWE, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_PHILIPPINES, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_SPAIN_TRADITIONAL_SORT,"es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_MEXICO, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_SPAIN_INTERNATIONAL_SORT,"es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_GUATEMALA, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_COSTA_RICA, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_PANAMA, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_DOMINICAN_REPUBLIC,"es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_VENEZUELA, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_COLOMBIA, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_PERU, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_ARGENTINA, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_ECUADOR, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_CHILE, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_URUGUAY, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_PARAGUAY, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_BOLIVIA, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_EL_SALVADOR, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_HONDURAS, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_NICARAGUA, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_PUERTO_RICO, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FINNISH_FINLAND, "fi" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_FRANCE, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_BELGIUM, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_CANADA, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_SWITZERLAND, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_LUXEMBOURG, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_MONACO, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_HEBREW_ISRAEL, "he" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_HUNGARIAN_HUNGARY, "hu" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ICELANDIC_ICELAND, "is" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ITALIAN_ITALY, "it" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ITALIAN_SWITZERLAND, "it" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_JAPANESE_JAPAN, "ja" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KOREAN_EXTENDED_WANSUNG_KOREA,"ko" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KOREAN_JOHAB_KOREA, "ko" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_DUTCH_NETHERLANDS, "nl" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_DUTCH_BELGIUM, "nl" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_NORWEGIAN_NORWAY_BOKMAL, "no" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_NORWEGIAN_NORWAY_NYNORSK, "nn" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_POLISH_POLAND, "pl" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_PORTUGUESE_BRAZIL, "pt" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_PORTUGUESE_PORTUGAL, "pt" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_RHAETO_ROMANIC_SWITZERLAND,"rm" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ROMANIAN_ROMANIA, "ro" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MOLDAVIAN_MOLDAVIA, "mo" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_RUSSIAN_RUSSIA, "ru" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_RUSSIAN_MOLDAVIA, "ru" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CROATIAN_CROATIA, "hr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SERBIAN_SERBIA_LATIN, "sr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SERBIAN_SERBIA_CYRILLIC, "sr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SLOVAK_SLOVAKIA, "sk" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ALBANIAN_ALBANIA, "sq" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SWEDISH_SWEDEN, "sv" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SWEDISH_FINLAND, "sv" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_THAI_THAILAND, "th" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TURKISH_TURKEY, "tr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_URDU_PAKISTAN, "ur" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_INDONESIAN_INDONESIA, "id" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_UKRAINIAN_UKRAINE, "uk" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_BELARUSIAN_BELARUS, "be" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SLOVENE_SLOVENIA, "sl" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ESTONIAN_ESTONIA, "et" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_LATVIAN_LATVIA, "lv" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_LITHUANIAN_LITHUANIA, "lt" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CLASSIC_LITHUANIAN_LITHUANIA,"lt" },
#ifdef TT_MS_LANGID_MAORI_NEW_ZELAND
/* this seems to be an error that have been dropped */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MAORI_NEW_ZEALAND, "mi" },
#endif
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FARSI_IRAN, "fa" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_VIETNAMESE_VIET_NAM, "vi" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARMENIAN_ARMENIA, "hy" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_AZERI_AZERBAIJAN_LATIN, "az" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_AZERI_AZERBAIJAN_CYRILLIC, "az" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_BASQUE_SPAIN, "eu" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SORBIAN_GERMANY, "wen" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MACEDONIAN_MACEDONIA, "mk" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SUTU_SOUTH_AFRICA, "st" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TSONGA_SOUTH_AFRICA, "ts" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TSWANA_SOUTH_AFRICA, "tn" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_VENDA_SOUTH_AFRICA, "ven" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_XHOSA_SOUTH_AFRICA, "xh" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ZULU_SOUTH_AFRICA, "zu" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_AFRIKAANS_SOUTH_AFRICA, "af" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GEORGIAN_GEORGIA, "ka" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FAEROESE_FAEROE_ISLANDS, "fo" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_HINDI_INDIA, "hi" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MALTESE_MALTA, "mt" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SAAMI_LAPONIA, "se" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SCOTTISH_GAELIC_UNITED_KINGDOM,"gd" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_IRISH_GAELIC_IRELAND, "ga" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MALAY_MALAYSIA, "ms" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MALAY_BRUNEI_DARUSSALAM, "ms" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KAZAK_KAZAKSTAN, "kk" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SWAHILI_KENYA, "sw" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_UZBEK_UZBEKISTAN_LATIN, "uz" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_UZBEK_UZBEKISTAN_CYRILLIC, "uz" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TATAR_TATARSTAN, "tt" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_BENGALI_INDIA, "bn" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_PUNJABI_INDIA, "pa" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GUJARATI_INDIA, "gu" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ORIYA_INDIA, "or" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TAMIL_INDIA, "ta" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TELUGU_INDIA, "te" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KANNADA_INDIA, "kn" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MALAYALAM_INDIA, "ml" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ASSAMESE_INDIA, "as" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MARATHI_INDIA, "mr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SANSKRIT_INDIA, "sa" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KONKANI_INDIA, "kok" },
/* new as of 2001-01-01 */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ARABIC_GENERAL, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CHINESE_GENERAL, "zh" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_GENERAL, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_WEST_INDIES, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_REUNION, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_CONGO, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_SENEGAL, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_CAMEROON, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_COTE_D_IVOIRE, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_MALI, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_BOSNIAN_BOSNIA_HERZEGOVINA,"bs" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_URDU_INDIA, "ur" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TAJIK_TAJIKISTAN, "tg" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_YIDDISH_GERMANY, "yi" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KIRGHIZ_KIRGHIZSTAN, "ky" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TURKMEN_TURKMENISTAN, "tk" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MONGOLIAN_MONGOLIA, "mn" },
/* the following seems to be inconsistent;
here is the current "official" way: */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TIBETAN_BHUTAN, "bo" },
/* and here is what is used by Passport SDK */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TIBETAN_CHINA, "bo" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_DZONGHKA_BHUTAN, "dz" },
/* end of inconsistency */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_WELSH_WALES, "cy" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KHMER_CAMBODIA, "km" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_LAO_LAOS, "lo" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_BURMESE_MYANMAR, "my" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GALICIAN_SPAIN, "gl" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MANIPURI_INDIA, "mni" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SINDHI_INDIA, "sd" },
/* the following one is only encountered in Microsoft RTF specification */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KASHMIRI_PAKISTAN, "ks" },
/* the following one is not in the Passport list, looks like an omission */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KASHMIRI_INDIA, "ks" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_NEPALI_NEPAL, "ne" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_NEPALI_INDIA, "ne" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRISIAN_NETHERLANDS, "fy" },
/* new as of 2001-03-01 (from Office Xp) */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_HONG_KONG, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_INDIA, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_MALAYSIA, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_ENGLISH_SINGAPORE, "en" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SYRIAC_SYRIA, "syr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SINHALESE_SRI_LANKA, "si" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_CHEROKEE_UNITED_STATES, "chr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_INUKTITUT_CANADA, "iu" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_AMHARIC_ETHIOPIA, "am" },
#if 0
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TAMAZIGHT_MOROCCO },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TAMAZIGHT_MOROCCO_LATIN },
#endif
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_PASHTO_AFGHANISTAN, "ps" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FILIPINO_PHILIPPINES, "phi" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_DHIVEHI_MALDIVES, "div" },
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_OROMO_ETHIOPIA, "om" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TIGRIGNA_ETHIOPIA, "ti" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_TIGRIGNA_ERYTHREA, "ti" },
/* New additions from Windows Xp/Passport SDK 2001-11-10. */
/* don't ask what this one means... It is commented out currently. */
#if 0
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GREEK_GREECE2 },
#endif
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_UNITED_STATES, "es" },
/* The following two IDs blatantly violate MS specs by using a */
/* sublanguage >,. */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SPANISH_LATIN_AMERICA, "es" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_NORTH_AFRICA, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_MOROCCO, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FRENCH_HAITI, "fr" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_BENGALI_BANGLADESH, "bn" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_PUNJABI_ARABIC_PAKISTAN, "ar" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_MONGOLIAN_MONGOLIA_MONGOLIAN,"mn" },
#if 0
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_EDO_NIGERIA },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_FULFULDE_NIGERIA },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_IBIBIO_NIGERIA },
#endif
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_HAUSA_NIGERIA, "ha" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_YORUBA_NIGERIA, "yo" },
/* language codes from, to, are (still) unknown. */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_IGBO_NIGERIA, "ibo" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_KANURI_NIGERIA, "kau" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_GUARANI_PARAGUAY, "gn" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_HAWAIIAN_UNITED_STATES, "haw" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_LATIN, "la" },
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_SOMALI_SOMALIA, "so" },
#if 0
/* Note: Yi does not have a (proper) ISO 639-2 code, since it is mostly */
/* not written (but OTOH the peculiar writing system is worth */
/* studying). */
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_YI_CHINA },
#endif
{ TT_PLATFORM_MICROSOFT, TT_MS_LANGID_PAPIAMENTU_NETHERLANDS_ANTILLES,"pap" },
};
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_FC_FT_LANGUAGE (int) (sizeof (fcFtLanguage) / sizeof (fcFtLanguage[0]))
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
typedef struct {
FT_UShort language_id;
Convert fromcode to char[12] from char *.
2005-09-23 04:08:40 +02:00
char fromcode[12];
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
} FcMacRomanFake;
static const FcMacRomanFake fcMacRomanFake[] = {
{ TT_MS_LANGID_JAPANESE_JAPAN, "SJIS-WIN" },
{ TT_MS_LANGID_ENGLISH_UNITED_STATES, "ASCII" },
};
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
static FcChar8 *
FcFontCapabilities(FT_Face face);
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_FC_MAC_ROMAN_FAKE (int) (sizeof (fcMacRomanFake) / sizeof (fcMacRomanFake[0]))
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
Decode MacRoman encoding in name table without iconv iconv support was turned off by default in f30a5d76. Some fonts, like Apple Color Emoji, only have their English name in a MacRoman entry. As such, decode MacRoman ourselves.
2014-07-24 21:28:09 +02:00
/* From http://www.unicode.org/Public/MAPPINGS/VENDORS/APPLE/ROMAN.TXT */
static const FcChar16 fcMacRomanNonASCIIToUnicode[128] = {
/*0x80*/ 0x00C4, /* LATIN CAPITAL LETTER A WITH DIAERESIS */
/*0x81*/ 0x00C5, /* LATIN CAPITAL LETTER A WITH RING ABOVE */
/*0x82*/ 0x00C7, /* LATIN CAPITAL LETTER C WITH CEDILLA */
/*0x83*/ 0x00C9, /* LATIN CAPITAL LETTER E WITH ACUTE */
/*0x84*/ 0x00D1, /* LATIN CAPITAL LETTER N WITH TILDE */
/*0x85*/ 0x00D6, /* LATIN CAPITAL LETTER O WITH DIAERESIS */
/*0x86*/ 0x00DC, /* LATIN CAPITAL LETTER U WITH DIAERESIS */
/*0x87*/ 0x00E1, /* LATIN SMALL LETTER A WITH ACUTE */
/*0x88*/ 0x00E0, /* LATIN SMALL LETTER A WITH GRAVE */
/*0x89*/ 0x00E2, /* LATIN SMALL LETTER A WITH CIRCUMFLEX */
/*0x8A*/ 0x00E4, /* LATIN SMALL LETTER A WITH DIAERESIS */
/*0x8B*/ 0x00E3, /* LATIN SMALL LETTER A WITH TILDE */
/*0x8C*/ 0x00E5, /* LATIN SMALL LETTER A WITH RING ABOVE */
/*0x8D*/ 0x00E7, /* LATIN SMALL LETTER C WITH CEDILLA */
/*0x8E*/ 0x00E9, /* LATIN SMALL LETTER E WITH ACUTE */
/*0x8F*/ 0x00E8, /* LATIN SMALL LETTER E WITH GRAVE */
/*0x90*/ 0x00EA, /* LATIN SMALL LETTER E WITH CIRCUMFLEX */
/*0x91*/ 0x00EB, /* LATIN SMALL LETTER E WITH DIAERESIS */
/*0x92*/ 0x00ED, /* LATIN SMALL LETTER I WITH ACUTE */
/*0x93*/ 0x00EC, /* LATIN SMALL LETTER I WITH GRAVE */
/*0x94*/ 0x00EE, /* LATIN SMALL LETTER I WITH CIRCUMFLEX */
/*0x95*/ 0x00EF, /* LATIN SMALL LETTER I WITH DIAERESIS */
/*0x96*/ 0x00F1, /* LATIN SMALL LETTER N WITH TILDE */
/*0x97*/ 0x00F3, /* LATIN SMALL LETTER O WITH ACUTE */
/*0x98*/ 0x00F2, /* LATIN SMALL LETTER O WITH GRAVE */
/*0x99*/ 0x00F4, /* LATIN SMALL LETTER O WITH CIRCUMFLEX */
/*0x9A*/ 0x00F6, /* LATIN SMALL LETTER O WITH DIAERESIS */
/*0x9B*/ 0x00F5, /* LATIN SMALL LETTER O WITH TILDE */
/*0x9C*/ 0x00FA, /* LATIN SMALL LETTER U WITH ACUTE */
/*0x9D*/ 0x00F9, /* LATIN SMALL LETTER U WITH GRAVE */
/*0x9E*/ 0x00FB, /* LATIN SMALL LETTER U WITH CIRCUMFLEX */
/*0x9F*/ 0x00FC, /* LATIN SMALL LETTER U WITH DIAERESIS */
/*0xA0*/ 0x2020, /* DAGGER */
/*0xA1*/ 0x00B0, /* DEGREE SIGN */
/*0xA2*/ 0x00A2, /* CENT SIGN */
/*0xA3*/ 0x00A3, /* POUND SIGN */
/*0xA4*/ 0x00A7, /* SECTION SIGN */
/*0xA5*/ 0x2022, /* BULLET */
/*0xA6*/ 0x00B6, /* PILCROW SIGN */
/*0xA7*/ 0x00DF, /* LATIN SMALL LETTER SHARP S */
/*0xA8*/ 0x00AE, /* REGISTERED SIGN */
/*0xA9*/ 0x00A9, /* COPYRIGHT SIGN */
/*0xAA*/ 0x2122, /* TRADE MARK SIGN */
/*0xAB*/ 0x00B4, /* ACUTE ACCENT */
/*0xAC*/ 0x00A8, /* DIAERESIS */
/*0xAD*/ 0x2260, /* NOT EQUAL TO */
/*0xAE*/ 0x00C6, /* LATIN CAPITAL LETTER AE */
/*0xAF*/ 0x00D8, /* LATIN CAPITAL LETTER O WITH STROKE */
/*0xB0*/ 0x221E, /* INFINITY */
/*0xB1*/ 0x00B1, /* PLUS-MINUS SIGN */
/*0xB2*/ 0x2264, /* LESS-THAN OR EQUAL TO */
/*0xB3*/ 0x2265, /* GREATER-THAN OR EQUAL TO */
/*0xB4*/ 0x00A5, /* YEN SIGN */
/*0xB5*/ 0x00B5, /* MICRO SIGN */
/*0xB6*/ 0x2202, /* PARTIAL DIFFERENTIAL */
/*0xB7*/ 0x2211, /* N-ARY SUMMATION */
/*0xB8*/ 0x220F, /* N-ARY PRODUCT */
/*0xB9*/ 0x03C0, /* GREEK SMALL LETTER PI */
/*0xBA*/ 0x222B, /* INTEGRAL */
/*0xBB*/ 0x00AA, /* FEMININE ORDINAL INDICATOR */
/*0xBC*/ 0x00BA, /* MASCULINE ORDINAL INDICATOR */
/*0xBD*/ 0x03A9, /* GREEK CAPITAL LETTER OMEGA */
/*0xBE*/ 0x00E6, /* LATIN SMALL LETTER AE */
/*0xBF*/ 0x00F8, /* LATIN SMALL LETTER O WITH STROKE */
/*0xC0*/ 0x00BF, /* INVERTED QUESTION MARK */
/*0xC1*/ 0x00A1, /* INVERTED EXCLAMATION MARK */
/*0xC2*/ 0x00AC, /* NOT SIGN */
/*0xC3*/ 0x221A, /* SQUARE ROOT */
/*0xC4*/ 0x0192, /* LATIN SMALL LETTER F WITH HOOK */
/*0xC5*/ 0x2248, /* ALMOST EQUAL TO */
/*0xC6*/ 0x2206, /* INCREMENT */
/*0xC7*/ 0x00AB, /* LEFT-POINTING DOUBLE ANGLE QUOTATION MARK */
/*0xC8*/ 0x00BB, /* RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK */
/*0xC9*/ 0x2026, /* HORIZONTAL ELLIPSIS */
/*0xCA*/ 0x00A0, /* NO-BREAK SPACE */
/*0xCB*/ 0x00C0, /* LATIN CAPITAL LETTER A WITH GRAVE */
/*0xCC*/ 0x00C3, /* LATIN CAPITAL LETTER A WITH TILDE */
/*0xCD*/ 0x00D5, /* LATIN CAPITAL LETTER O WITH TILDE */
/*0xCE*/ 0x0152, /* LATIN CAPITAL LIGATURE OE */
/*0xCF*/ 0x0153, /* LATIN SMALL LIGATURE OE */
/*0xD0*/ 0x2013, /* EN DASH */
/*0xD1*/ 0x2014, /* EM DASH */
/*0xD2*/ 0x201C, /* LEFT DOUBLE QUOTATION MARK */
/*0xD3*/ 0x201D, /* RIGHT DOUBLE QUOTATION MARK */
/*0xD4*/ 0x2018, /* LEFT SINGLE QUOTATION MARK */
/*0xD5*/ 0x2019, /* RIGHT SINGLE QUOTATION MARK */
/*0xD6*/ 0x00F7, /* DIVISION SIGN */
/*0xD7*/ 0x25CA, /* LOZENGE */
/*0xD8*/ 0x00FF, /* LATIN SMALL LETTER Y WITH DIAERESIS */
/*0xD9*/ 0x0178, /* LATIN CAPITAL LETTER Y WITH DIAERESIS */
/*0xDA*/ 0x2044, /* FRACTION SLASH */
/*0xDB*/ 0x20AC, /* EURO SIGN */
/*0xDC*/ 0x2039, /* SINGLE LEFT-POINTING ANGLE QUOTATION MARK */
/*0xDD*/ 0x203A, /* SINGLE RIGHT-POINTING ANGLE QUOTATION MARK */
/*0xDE*/ 0xFB01, /* LATIN SMALL LIGATURE FI */
/*0xDF*/ 0xFB02, /* LATIN SMALL LIGATURE FL */
/*0xE0*/ 0x2021, /* DOUBLE DAGGER */
/*0xE1*/ 0x00B7, /* MIDDLE DOT */
/*0xE2*/ 0x201A, /* SINGLE LOW-9 QUOTATION MARK */
/*0xE3*/ 0x201E, /* DOUBLE LOW-9 QUOTATION MARK */
/*0xE4*/ 0x2030, /* PER MILLE SIGN */
/*0xE5*/ 0x00C2, /* LATIN CAPITAL LETTER A WITH CIRCUMFLEX */
/*0xE6*/ 0x00CA, /* LATIN CAPITAL LETTER E WITH CIRCUMFLEX */
/*0xE7*/ 0x00C1, /* LATIN CAPITAL LETTER A WITH ACUTE */
/*0xE8*/ 0x00CB, /* LATIN CAPITAL LETTER E WITH DIAERESIS */
/*0xE9*/ 0x00C8, /* LATIN CAPITAL LETTER E WITH GRAVE */
/*0xEA*/ 0x00CD, /* LATIN CAPITAL LETTER I WITH ACUTE */
/*0xEB*/ 0x00CE, /* LATIN CAPITAL LETTER I WITH CIRCUMFLEX */
/*0xEC*/ 0x00CF, /* LATIN CAPITAL LETTER I WITH DIAERESIS */
/*0xED*/ 0x00CC, /* LATIN CAPITAL LETTER I WITH GRAVE */
/*0xEE*/ 0x00D3, /* LATIN CAPITAL LETTER O WITH ACUTE */
/*0xEF*/ 0x00D4, /* LATIN CAPITAL LETTER O WITH CIRCUMFLEX */
/*0xF0*/ 0xF8FF, /* Apple logo */
/*0xF1*/ 0x00D2, /* LATIN CAPITAL LETTER O WITH GRAVE */
/*0xF2*/ 0x00DA, /* LATIN CAPITAL LETTER U WITH ACUTE */
/*0xF3*/ 0x00DB, /* LATIN CAPITAL LETTER U WITH CIRCUMFLEX */
/*0xF4*/ 0x00D9, /* LATIN CAPITAL LETTER U WITH GRAVE */
/*0xF5*/ 0x0131, /* LATIN SMALL LETTER DOTLESS I */
/*0xF6*/ 0x02C6, /* MODIFIER LETTER CIRCUMFLEX ACCENT */
/*0xF7*/ 0x02DC, /* SMALL TILDE */
/*0xF8*/ 0x00AF, /* MACRON */
/*0xF9*/ 0x02D8, /* BREVE */
/*0xFA*/ 0x02D9, /* DOT ABOVE */
/*0xFB*/ 0x02DA, /* RING ABOVE */
/*0xFC*/ 0x00B8, /* CEDILLA */
/*0xFD*/ 0x02DD, /* DOUBLE ACUTE ACCENT */
/*0xFE*/ 0x02DB, /* OGONEK */
/*0xFF*/ 0x02C7, /* CARON */
};
Fix the issues with GNU libiconv vs. libc iconv (which especially appear on Solarii). Approach suggested by Tim Mooney. reviewed by: plam
2006-04-25 17:33:07 +02:00
#if USE_ICONV
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
#include <iconv.h>
#endif
Guess that mac roman names with lots of high bits are actually SJIS. Many Japanese fonts incorrectly include names tagged as Roman encoding and English language which are actually Japanese names in the SJIS encoding. Guess that names with a large number of high bits set are SJIS encoded Japanese names rather than English names.
2006-09-02 06:30:54 +02:00
/*
* A shift-JIS will have many high bits turned on
*/
static FcBool
FcLooksLikeSJIS (FcChar8 *string, int len)
{
int nhigh = 0, nlow = 0;
while (len-- > 0)
{
if (*string++ & 0x80) nhigh++;
else nlow++;
}
/*
* Heuristic -- if more than 1/3 of the bytes have the high-bit set,
* this is likely to be SJIS and not ROMAN
*/
if (nhigh * 2 > nlow)
return FcTrue;
return FcFalse;
}
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
static FcChar8 *
FcSfntNameTranscode (FT_SfntName *sname)
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
{
Convert fromcode to char[12] from char *.
2005-09-23 04:08:40 +02:00
int i;
const char *fromcode;
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
#if USE_ICONV
iconv_t cd;
#endif
FcChar8 *utf8;
for (i = 0; i < NUM_FC_FT_ENCODING; i++)
if (fcFtEncoding[i].platform_id == sname->platform_id &&
(fcFtEncoding[i].encoding_id == TT_ENCODING_DONT_CARE ||
fcFtEncoding[i].encoding_id == sname->encoding_id))
break;
if (i == NUM_FC_FT_ENCODING)
return 0;
fromcode = fcFtEncoding[i].fromcode;
/*
Guess that mac roman names with lots of high bits are actually SJIS. Many Japanese fonts incorrectly include names tagged as Roman encoding and English language which are actually Japanese names in the SJIS encoding. Guess that names with a large number of high bits set are SJIS encoded Japanese names rather than English names.
2006-09-02 06:30:54 +02:00
* Many names encoded for TT_PLATFORM_MACINTOSH are broken
* in various ways. Kludge around them.
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
*/
Guess that mac roman names with lots of high bits are actually SJIS. Many Japanese fonts incorrectly include names tagged as Roman encoding and English language which are actually Japanese names in the SJIS encoding. Guess that names with a large number of high bits set are SJIS encoded Japanese names rather than English names.
2006-09-02 06:30:54 +02:00
if (!strcmp (fromcode, FC_ENCODING_MAC_ROMAN))
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{
Guess that mac roman names with lots of high bits are actually SJIS. Many Japanese fonts incorrectly include names tagged as Roman encoding and English language which are actually Japanese names in the SJIS encoding. Guess that names with a large number of high bits set are SJIS encoded Japanese names rather than English names.
2006-09-02 06:30:54 +02:00
if (sname->language_id == TT_MAC_LANGID_ENGLISH &&
FcLooksLikeSJIS (sname->string, sname->string_len))
{
fromcode = "SJIS";
}
else if (sname->language_id >= 0x100)
{
/*
* "real" Mac language IDs are all less than 150.
* Names using one of the MS language IDs are assumed
* to use an associated encoding (Yes, this is a kludge)
*/
int f;
fromcode = NULL;
for (f = 0; f < NUM_FC_MAC_ROMAN_FAKE; f++)
if (fcMacRomanFake[f].language_id == sname->language_id)
{
fromcode = fcMacRomanFake[f].fromcode;
break;
}
if (!fromcode)
return 0;
}
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
}
if (!strcmp (fromcode, "UCS-2BE") || !strcmp (fromcode, "UTF-16BE"))
{
FcChar8 *src = sname->string;
int src_len = sname->string_len;
int len;
int wchar;
int ilen, olen;
FcChar8 *u8;
FcChar32 ucs4;
/*
* Convert Utf16 to Utf8
*/
if (!FcUtf16Len (src, FcEndianBig, src_len, &len, &wchar))
return 0;
/*
* Allocate plenty of space. Freed below
*/
utf8 = malloc (len * FC_UTF8_MAX_LEN + 1);
if (!utf8)
return 0;
u8 = utf8;
while ((ilen = FcUtf16ToUcs4 (src, FcEndianBig, &ucs4, src_len)) > 0)
{
src_len -= ilen;
src += ilen;
olen = FcUcs4ToUtf8 (ucs4, u8);
u8 += olen;
}
*u8 = '\0';
goto done;
}
if (!strcmp (fromcode, "ASCII") || !strcmp (fromcode, "ISO-8859-1"))
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
{
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
FcChar8 *src = sname->string;
int src_len = sname->string_len;
int olen;
FcChar8 *u8;
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
FcChar32 ucs4;
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
/*
* Convert Latin1 to Utf8. Freed below
*/
utf8 = malloc (src_len * 2 + 1);
if (!utf8)
return 0;
u8 = utf8;
while (src_len > 0)
{
ucs4 = *src++;
src_len--;
olen = FcUcs4ToUtf8 (ucs4, u8);
u8 += olen;
}
*u8 = '\0';
goto done;
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
}
Ouch, fix buffer
2014-07-24 21:34:20 +02:00
if (!strcmp (fromcode, FC_ENCODING_MAC_ROMAN))
Decode MacRoman encoding in name table without iconv iconv support was turned off by default in f30a5d76. Some fonts, like Apple Color Emoji, only have their English name in a MacRoman entry. As such, decode MacRoman ourselves.
2014-07-24 21:28:09 +02:00
{
FcChar8 *src = sname->string;
int src_len = sname->string_len;
int olen;
FcChar8 *u8;
FcChar32 ucs4;
/*
* Convert Latin1 to Utf8. Freed below
*/
Ouch, fix buffer
2014-07-24 21:34:20 +02:00
utf8 = malloc (src_len * 3 + 1);
Decode MacRoman encoding in name table without iconv iconv support was turned off by default in f30a5d76. Some fonts, like Apple Color Emoji, only have their English name in a MacRoman entry. As such, decode MacRoman ourselves.
2014-07-24 21:28:09 +02:00
if (!utf8)
return 0;
u8 = utf8;
while (src_len > 0)
{
ucs4 = *src++;
if (ucs4 >= 128)
ucs4 = fcMacRomanNonASCIIToUnicode[ucs4 - 128];
src_len--;
olen = FcUcs4ToUtf8 (ucs4, u8);
u8 += olen;
}
*u8 = '\0';
goto done;
}
Don't force bitmap font enable in default configuration; allows users to override this in ~/.fonts.conf Updated translations Destroy font configuration on exit to help valgrind Use own transcoding routines in preference to iconv which appears to have leaks in some translators. Call iconv_close after using iconv (oops). Prefer unicode encoding of Euro char as some fonts mis-encode Euro in other ones. Must fetch bitmap glyphs to get width values to check for monospace/dual-width fonts.
2005-04-21 21:03:53 +02:00
#if USE_ICONV
cd = iconv_open ("UTF-8", fromcode);
if (cd && cd != (iconv_t) (-1))
{
size_t in_bytes_left = sname->string_len;
size_t out_bytes_left = sname->string_len * FC_UTF8_MAX_LEN;
char *inbuf, *outbuf;
utf8 = malloc (out_bytes_left + 1);
if (!utf8)
{
iconv_close (cd);
return 0;
}
outbuf = (char *) utf8;
inbuf = (char *) sname->string;
while (in_bytes_left)
{
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
size_t did = iconv (cd,
Don't force bitmap font enable in default configuration; allows users to override this in ~/.fonts.conf Updated translations Destroy font configuration on exit to help valgrind Use own transcoding routines in preference to iconv which appears to have leaks in some translators. Call iconv_close after using iconv (oops). Prefer unicode encoding of Euro char as some fonts mis-encode Euro in other ones. Must fetch bitmap glyphs to get width values to check for monospace/dual-width fonts.
2005-04-21 21:03:53 +02:00
&inbuf, &in_bytes_left,
&outbuf, &out_bytes_left);
if (did == (size_t) (-1))
{
iconv_close (cd);
free (utf8);
return 0;
}
}
iconv_close (cd);
*outbuf = '\0';
goto done;
}
#endif
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
return 0;
done:
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
if (FcStrCmpIgnoreBlanksAndCase (utf8, (FcChar8 *) "") == 0)
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{
free (utf8);
return 0;
}
return utf8;
}
Add some more consts, fixing more GCC4 warnings.
2005-10-22 17:12:05 +02:00
static const FcChar8 *
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
FcSfntNameLanguage (FT_SfntName *sname)
{
int i;
Guess that mac roman names with lots of high bits are actually SJIS. Many Japanese fonts incorrectly include names tagged as Roman encoding and English language which are actually Japanese names in the SJIS encoding. Guess that names with a large number of high bits set are SJIS encoded Japanese names rather than English names.
2006-09-02 06:30:54 +02:00
FT_UShort platform_id = sname->platform_id;
FT_UShort language_id = sname->language_id;
/*
* Many names encoded for TT_PLATFORM_MACINTOSH are broken
* in various ways. Kludge around them.
*/
if (platform_id == TT_PLATFORM_MACINTOSH &&
sname->encoding_id == TT_MAC_ID_ROMAN &&
FcLooksLikeSJIS (sname->string, sname->string_len))
{
language_id = TT_MAC_LANGID_JAPANESE;
}
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
for (i = 0; i < NUM_FC_FT_LANGUAGE; i++)
Guess that mac roman names with lots of high bits are actually SJIS. Many Japanese fonts incorrectly include names tagged as Roman encoding and English language which are actually Japanese names in the SJIS encoding. Guess that names with a large number of high bits set are SJIS encoded Japanese names rather than English names.
2006-09-02 06:30:54 +02:00
if (fcFtLanguage[i].platform_id == platform_id &&
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
(fcFtLanguage[i].language_id == TT_LANGUAGE_DONT_CARE ||
Guess that mac roman names with lots of high bits are actually SJIS. Many Japanese fonts incorrectly include names tagged as Roman encoding and English language which are actually Japanese names in the SJIS encoding. Guess that names with a large number of high bits set are SJIS encoded Japanese names rather than English names.
2006-09-02 06:30:54 +02:00
fcFtLanguage[i].language_id == language_id))
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
{
if (fcFtLanguage[i].lang[0] == '\0')
return NULL;
else
return (FcChar8 *) fcFtLanguage[i].lang;
}
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
return 0;
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
}
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
/* Order is significant. For example, some B&H fonts are hinted by
URW++, and both strings appear in the notice. */
Add Culmus foundry to the vendor list Maps fonts produced by the Culmus project <http://culmus.sourceforge.net> to the XLFD foundry name culmus. For TrueType fonts, maps the vendor code CLM from the TrueType vendor id field. For Type1 fonts, which use heuristics to guess mappings to XLFD foundries from words in the copyright notice, add the names of the main contributors to the Culmus product to recognize the fonts under their copyright. Patch from Maxim Iorsh
2013-02-07 09:56:17 +01:00
static const char *FcNoticeFoundries[][2] =
{
{"Adobe", "adobe"},
{"Bigelow", "b&h"},
{"Bitstream", "bitstream"},
{"Gnat", "culmus"},
{"Iorsh", "culmus"},
{"HanYang System", "hanyang"},
{"Font21", "hwan"},
{"IBM", "ibm"},
{"International Typeface Corporation", "itc"},
{"Linotype", "linotype"},
{"LINOTYPE-HELL", "linotype"},
{"Microsoft", "microsoft"},
{"Monotype", "monotype"},
{"Omega", "omega"},
{"Tiro Typeworks", "tiro"},
{"URW", "urw"},
{"XFree86", "xfree86"},
{"Xorg", "xorg"},
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
};
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_NOTICE_FOUNDRIES (int) (sizeof (FcNoticeFoundries) / sizeof (FcNoticeFoundries[0]))
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
static const FcChar8 *
Add instructions for doing a release clean up .spec file; perhaps this will be useful to somebody... Update to 2.2.93 Make 'scanopen' static so GCC doesn't whine about lacking prototype Add WARN_CFLAGS to pass -W flags for GCC systems Change various char types around to match across function calls. Fixed bug in using available_sizes[i].height which is in pixels, not 64ths of a pixel.
2004-04-14 20:08:41 +02:00
FcNoticeFoundry(const FT_String *notice)
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
{
int i;
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
if (notice)
for(i = 0; i < NUM_NOTICE_FOUNDRIES; i++)
Store FcNoticeFoundries in read-only memory. Use a single character array and a separate table of integer indices.
2007-10-25 09:36:37 +02:00
{
Add Culmus foundry to the vendor list Maps fonts produced by the Culmus project <http://culmus.sourceforge.net> to the XLFD foundry name culmus. For TrueType fonts, maps the vendor code CLM from the TrueType vendor id field. For Type1 fonts, which use heuristics to guess mappings to XLFD foundries from words in the copyright notice, add the names of the main contributors to the Culmus product to recognize the fonts under their copyright. Patch from Maxim Iorsh
2013-02-07 09:56:17 +01:00
const char *n = FcNoticeFoundries[i][0];
const char *f = FcNoticeFoundries[i][1];
Store FcNoticeFoundries in read-only memory. Use a single character array and a separate table of integer indices.
2007-10-25 09:36:37 +02:00
if (strstr ((const char *) notice, n))
return (const FcChar8 *) f;
}
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
return 0;
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
}
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
static FcBool
Add instructions for doing a release clean up .spec file; perhaps this will be useful to somebody... Update to 2.2.93 Make 'scanopen' static so GCC doesn't whine about lacking prototype Add WARN_CFLAGS to pass -W flags for GCC systems Change various char types around to match across function calls. Fixed bug in using available_sizes[i].height which is in pixels, not 64ths of a pixel.
2004-04-14 20:08:41 +02:00
FcVendorMatch(const FT_Char vendor[4], const FT_Char *vendor_string)
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
{
/* vendor is not necessarily NUL-terminated. */
int i, len;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
len = strlen((char *) vendor_string);
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
if (memcmp(vendor, vendor_string, len) != 0)
return FcFalse;
for (i = len; i < 4; i++)
if (vendor[i] != ' ' && vendor[i] != '\0')
return FcFalse;
return FcTrue;
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
}
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
/* This table is partly taken from ttmkfdir by Joerg Pommnitz. */
/* It should not contain useless entries (such as UNKN) nor duplicate
entries for padding both with spaces and NULs. */
static const struct {
Store FcVendorFoundries in read-only memory. Create fixed size strings for vendor and foundry.
2007-10-25 09:38:44 +02:00
const FT_Char vendor[5];
const FcChar8 foundry[13];
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
} FcVendorFoundries[] = {
Store FcVendorFoundries in read-only memory. Create fixed size strings for vendor and foundry.
2007-10-25 09:38:44 +02:00
{ "ADBE", "adobe"},
{ "AGFA", "agfa"},
{ "ALTS", "altsys"},
{ "APPL", "apple"},
{ "ARPH", "arphic"},
{ "ATEC", "alltype"},
{ "B&H", "b&h"},
{ "BITS", "bitstream"},
{ "CANO", "cannon"},
Add Culmus foundry to the vendor list Maps fonts produced by the Culmus project <http://culmus.sourceforge.net> to the XLFD foundry name culmus. For TrueType fonts, maps the vendor code CLM from the TrueType vendor id field. For Type1 fonts, which use heuristics to guess mappings to XLFD foundries from words in the copyright notice, add the names of the main contributors to the Culmus product to recognize the fonts under their copyright. Patch from Maxim Iorsh
2013-02-07 09:56:17 +01:00
{ "CLM", "culmus"},
Store FcVendorFoundries in read-only memory. Create fixed size strings for vendor and foundry.
2007-10-25 09:38:44 +02:00
{ "DYNA", "dynalab"},
{ "EPSN", "epson"},
{ "FJ", "fujitsu"},
{ "IBM", "ibm"},
{ "ITC", "itc"},
{ "IMPR", "impress"},
{ "LARA", "larabiefonts"},
{ "LEAF", "interleaf"},
{ "LETR", "letraset"},
{ "LINO", "linotype"},
{ "MACR", "macromedia"},
{ "MONO", "monotype"},
{ "MS", "microsoft"},
{ "MT", "monotype"},
{ "NEC", "nec"},
{ "PARA", "paratype"},
{ "QMSI", "qms"},
{ "RICO", "ricoh"},
{ "URW", "urw"},
{ "Y&Y", "y&y"}
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
};
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_VENDOR_FOUNDRIES (int) (sizeof (FcVendorFoundries) / sizeof (FcVendorFoundries[0]))
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
static const FcChar8 *
Add instructions for doing a release clean up .spec file; perhaps this will be useful to somebody... Update to 2.2.93 Make 'scanopen' static so GCC doesn't whine about lacking prototype Add WARN_CFLAGS to pass -W flags for GCC systems Change various char types around to match across function calls. Fixed bug in using available_sizes[i].height which is in pixels, not 64ths of a pixel.
2004-04-14 20:08:41 +02:00
FcVendorFoundry(const FT_Char vendor[4])
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
{
int i;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
if (vendor)
for(i = 0; i < NUM_VENDOR_FOUNDRIES; i++)
if (FcVendorMatch (vendor, FcVendorFoundries[i].vendor))
return FcVendorFoundries[i].foundry;
return 0;
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
}
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
typedef struct _FcStringConst {
Add instructions for doing a release clean up .spec file; perhaps this will be useful to somebody... Update to 2.2.93 Make 'scanopen' static so GCC doesn't whine about lacking prototype Add WARN_CFLAGS to pass -W flags for GCC systems Change various char types around to match across function calls. Fixed bug in using available_sizes[i].height which is in pixels, not 64ths of a pixel.
2004-04-14 20:08:41 +02:00
const FcChar8 *name;
int value;
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
} FcStringConst;
static int
FcStringIsConst (const FcChar8 *string,
const FcStringConst *c,
int nc)
{
int i;
for (i = 0; i < nc; i++)
if (FcStrCmpIgnoreBlanksAndCase (string, c[i].name) == 0)
return c[i].value;
return -1;
}
static int
FcStringContainsConst (const FcChar8 *string,
const FcStringConst *c,
int nc)
{
int i;
for (i = 0; i < nc; i++)
Match 'ultra' on word boundaries to detect ultra bold fonts. (bug 2511) Added FcStrContainsWord to detect strings on word boundaries.
2006-09-02 05:25:21 +02:00
{
if (c[i].name[0] == '<')
{
if (FcStrContainsWord (string, c[i].name + 1))
return c[i].value;
}
else
{
if (FcStrContainsIgnoreBlanksAndCase (string, c[i].name))
return c[i].value;
}
}
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
return -1;
}
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
typedef FcChar8 *FC8;
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
static const FcStringConst weightConsts[] = {
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
{ (FC8) "thin", FC_WEIGHT_THIN },
{ (FC8) "extralight", FC_WEIGHT_EXTRALIGHT },
{ (FC8) "ultralight", FC_WEIGHT_ULTRALIGHT },
Add FC_WEIGHT_DEMILIGHT Part of https://bugs.freedesktop.org/show_bug.cgi?id=81453 Also hooks up FC_WEIGHT_BOOK to fcfreetype.c.
2014-07-25 22:24:26 +02:00
{ (FC8) "demilight", FC_WEIGHT_DEMILIGHT },
{ (FC8) "semilight", FC_WEIGHT_SEMILIGHT },
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
{ (FC8) "light", FC_WEIGHT_LIGHT },
{ (FC8) "book", FC_WEIGHT_BOOK },
{ (FC8) "regular", FC_WEIGHT_REGULAR },
{ (FC8) "normal", FC_WEIGHT_NORMAL },
{ (FC8) "medium", FC_WEIGHT_MEDIUM },
{ (FC8) "demibold", FC_WEIGHT_DEMIBOLD },
{ (FC8) "demi", FC_WEIGHT_DEMIBOLD },
{ (FC8) "semibold", FC_WEIGHT_SEMIBOLD },
{ (FC8) "extrabold", FC_WEIGHT_EXTRABOLD },
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
{ (FC8) "superbold", FC_WEIGHT_EXTRABOLD },
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
{ (FC8) "ultrabold", FC_WEIGHT_ULTRABOLD },
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
{ (FC8) "bold", FC_WEIGHT_BOLD },
Handle UltraBlack weight. UltraBlack is weight 950 on the CSS scale; handle this by name and value encoding it as fontconfig weight 215.
2007-10-25 10:03:40 +02:00
{ (FC8) "ultrablack", FC_WEIGHT_ULTRABLACK },
{ (FC8) "superblack", FC_WEIGHT_EXTRABLACK },
{ (FC8) "extrablack", FC_WEIGHT_EXTRABLACK },
Match 'ultra' on word boundaries to detect ultra bold fonts. (bug 2511) Added FcStrContainsWord to detect strings on word boundaries.
2006-09-02 05:25:21 +02:00
{ (FC8) "<ultra", FC_WEIGHT_ULTRABOLD }, /* only if a word */
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
{ (FC8) "black", FC_WEIGHT_BLACK },
{ (FC8) "heavy", FC_WEIGHT_HEAVY },
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
};
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_WEIGHT_CONSTS (int) (sizeof (weightConsts) / sizeof (weightConsts[0]))
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
#define FcIsWeight(s) FcStringIsConst(s,weightConsts,NUM_WEIGHT_CONSTS)
#define FcContainsWeight(s) FcStringContainsConst (s,weightConsts,NUM_WEIGHT_CONSTS)
static const FcStringConst widthConsts[] = {
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
{ (FC8) "ultracondensed", FC_WIDTH_ULTRACONDENSED },
{ (FC8) "extracondensed", FC_WIDTH_EXTRACONDENSED },
{ (FC8) "semicondensed", FC_WIDTH_SEMICONDENSED },
{ (FC8) "condensed", FC_WIDTH_CONDENSED }, /* must be after *condensed */
{ (FC8) "normal", FC_WIDTH_NORMAL },
{ (FC8) "semiexpanded", FC_WIDTH_SEMIEXPANDED },
{ (FC8) "extraexpanded", FC_WIDTH_EXTRAEXPANDED },
{ (FC8) "ultraexpanded", FC_WIDTH_ULTRAEXPANDED },
{ (FC8) "expanded", FC_WIDTH_EXPANDED }, /* must be after *expanded */
Add extended, caps, dunhill style mappings. extended -> expanded. caps, dunhill -> decorative
2008-05-04 10:27:42 +02:00
{ (FC8) "extended", FC_WIDTH_EXPANDED },
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
};
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_WIDTH_CONSTS (int) (sizeof (widthConsts) / sizeof (widthConsts[0]))
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
#define FcIsWidth(s) FcStringIsConst(s,widthConsts,NUM_WIDTH_CONSTS)
#define FcContainsWidth(s) FcStringContainsConst (s,widthConsts,NUM_WIDTH_CONSTS)
static const FcStringConst slantConsts[] = {
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
{ (FC8) "italic", FC_SLANT_ITALIC },
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
{ (FC8) "kursiv", FC_SLANT_ITALIC },
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
{ (FC8) "oblique", FC_SLANT_OBLIQUE },
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
};
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_SLANT_CONSTS (int) (sizeof (slantConsts) / sizeof (slantConsts[0]))
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
#define FcContainsSlant(s) FcStringContainsConst (s,slantConsts,NUM_SLANT_CONSTS)
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
static const FcStringConst decorativeConsts[] = {
{ (FC8) "shadow", FcTrue },
Add extended, caps, dunhill style mappings. extended -> expanded. caps, dunhill -> decorative
2008-05-04 10:27:42 +02:00
{ (FC8) "caps", FcTrue },
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
{ (FC8) "antiqua", FcTrue },
{ (FC8) "romansc", FcTrue },
{ (FC8) "embosed", FcTrue },
Add extended, caps, dunhill style mappings. extended -> expanded. caps, dunhill -> decorative
2008-05-04 10:27:42 +02:00
{ (FC8) "dunhill", FcTrue },
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
};
#define NUM_DECORATIVE_CONSTS (int) (sizeof (decorativeConsts) / sizeof (decorativeConsts[0]))
#define FcContainsDecorative(s) FcStringContainsConst (s,decorativeConsts,NUM_DECORATIVE_CONSTS)
Add Low Saxon orthography (Kenneth Rohde Christiansen <kenneth@gnu.org>) Oops. Left 'newest.set' unset, which would miscompute the newest file Add FcGetPixelSize to extract correct pixel size from bdf/pcf font properties (which report the wrong value in current FreeType) Don't attempt to check for empty glyphs in non-scalable fonts; they have no outlines...
2004-02-07 08:13:48 +01:00
static double
FcGetPixelSize (FT_Face face, int i)
{
Ok, so I messed up the test for y_ppem. Let's see if I got it right this time.
2004-02-11 19:53:05 +01:00
#if HAVE_FT_GET_BDF_PROPERTY
Add Low Saxon orthography (Kenneth Rohde Christiansen <kenneth@gnu.org>) Oops. Left 'newest.set' unset, which would miscompute the newest file Add FcGetPixelSize to extract correct pixel size from bdf/pcf font properties (which report the wrong value in current FreeType) Don't attempt to check for empty glyphs in non-scalable fonts; they have no outlines...
2004-02-07 08:13:48 +01:00
if (face->num_fixed_sizes == 1)
{
BDF_PropertyRec prop;
int rc;
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
rc = FT_Get_BDF_Property (face, "PIXEL_SIZE", &prop);
Add Low Saxon orthography (Kenneth Rohde Christiansen <kenneth@gnu.org>) Oops. Left 'newest.set' unset, which would miscompute the newest file Add FcGetPixelSize to extract correct pixel size from bdf/pcf font properties (which report the wrong value in current FreeType) Don't attempt to check for empty glyphs in non-scalable fonts; they have no outlines...
2004-02-07 08:13:48 +01:00
if (rc == 0 && prop.type == BDF_PROPERTY_TYPE_INTEGER)
return (double) prop.u.integer;
}
#endif
Ok, so I messed up the test for y_ppem. Let's see if I got it right this time.
2004-02-11 19:53:05 +01:00
#if HAVE_FT_BITMAP_SIZE_Y_PPEM
Add Low Saxon orthography (Kenneth Rohde Christiansen <kenneth@gnu.org>) Oops. Left 'newest.set' unset, which would miscompute the newest file Add FcGetPixelSize to extract correct pixel size from bdf/pcf font properties (which report the wrong value in current FreeType) Don't attempt to check for empty glyphs in non-scalable fonts; they have no outlines...
2004-02-07 08:13:48 +01:00
return (double) face->available_sizes[i].y_ppem / 64.0;
Pre-2.1.5 versions of FreeType didn't include y_ppem in the FT_Bitmap_Size record. Add a configure.in test for this and change the code accordingly (using height instead).
2004-02-10 19:38:58 +01:00
#else
Add instructions for doing a release clean up .spec file; perhaps this will be useful to somebody... Update to 2.2.93 Make 'scanopen' static so GCC doesn't whine about lacking prototype Add WARN_CFLAGS to pass -W flags for GCC systems Change various char types around to match across function calls. Fixed bug in using available_sizes[i].height which is in pixels, not 64ths of a pixel.
2004-04-14 20:08:41 +02:00
return (double) face->available_sizes[i].height;
Pre-2.1.5 versions of FreeType didn't include y_ppem in the FT_Bitmap_Size record. Add a configure.in test for this and change the code accordingly (using height instead).
2004-02-10 19:38:58 +01:00
#endif
Add Low Saxon orthography (Kenneth Rohde Christiansen <kenneth@gnu.org>) Oops. Left 'newest.set' unset, which would miscompute the newest file Add FcGetPixelSize to extract correct pixel size from bdf/pcf font properties (which report the wrong value in current FreeType) Don't attempt to check for empty glyphs in non-scalable fonts; they have no outlines...
2004-02-07 08:13:48 +01:00
}
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
static FcBool
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
FcStringInPatternElement (FcPattern *pat, const char *elt, FcChar8 *string)
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{
int e;
FcChar8 *old;
for (e = 0; FcPatternGetString (pat, elt, e, &old) == FcResultMatch; e++)
if (!FcStrCmpIgnoreBlanksAndCase (old, string))
{
return FcTrue;
}
return FcFalse;
}
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
static const FT_UShort platform_order[] = {
TT_PLATFORM_MICROSOFT,
TT_PLATFORM_APPLE_UNICODE,
TT_PLATFORM_MACINTOSH,
};
#define NUM_PLATFORM_ORDER (sizeof (platform_order) / sizeof (platform_order[0]))
static const FT_UShort nameid_order[] = {
Use/prefer WWS family/style (name table id 21/22)
2009-06-02 03:14:56 +02:00
#ifdef TT_NAME_ID_WWS_FAMILY
TT_NAME_ID_WWS_FAMILY,
#endif
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
TT_NAME_ID_PREFERRED_FAMILY,
TT_NAME_ID_FONT_FAMILY,
TT_NAME_ID_MAC_FULL_NAME,
TT_NAME_ID_FULL_NAME,
Use/prefer WWS family/style (name table id 21/22)
2009-06-02 03:14:56 +02:00
#ifdef TT_NAME_ID_WWS_SUBFAMILY
TT_NAME_ID_WWS_SUBFAMILY,
#endif
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
TT_NAME_ID_PREFERRED_SUBFAMILY,
TT_NAME_ID_FONT_SUBFAMILY,
TT_NAME_ID_TRADEMARK,
TT_NAME_ID_MANUFACTURER,
};
#define NUM_NAMEID_ORDER (sizeof (nameid_order) / sizeof (nameid_order[0]))
Initial revision
2002-02-15 00:34:13 +01:00
FcPattern *
Add FcFreeTypeQueryFace external API. Bug #7311. Expose ability to build an FcPattern directly from an FT_Face object.
2006-12-02 22:22:27 +01:00
FcFreeTypeQueryFace (const FT_Face face,
const FcChar8 *file,
int id,
FcBlanks *blanks)
Initial revision
2002-02-15 00:34:13 +01:00
{
fontconfig library: build fixes and compiler warning fixes
2002-02-15 07:01:28 +01:00
FcPattern *pat;
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
int slant = -1;
int weight = -1;
add font widths and extend weight from OS/2 table
2003-03-12 23:16:43 +01:00
int width = -1;
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
FcBool decorative = FcFalse;
fontconfig library: build fixes and compiler warning fixes
2002-02-15 07:01:28 +01:00
int i;
FcCharSet *cs;
Reimplement FC_LANG as FcTypeLang, freeze patterns, other cleanup
2002-08-22 09:36:45 +02:00
FcLangSet *ls;
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
#if 0
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
FcChar8 *family = 0;
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
#endif
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
FcChar8 *complex_;
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
const FcChar8 *foundry = 0;
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
int spacing;
fontconfig library: build fixes and compiler warning fixes
2002-02-15 07:01:28 +01:00
TT_OS2 *os2;
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#if HAVE_FT_GET_PS_FONT_INFO
Bug #46, #47 fontconfig should retrieve type 1 font information from FontInfo dictionary Patch provided by g2@magestudios.net (Gerard Escalante)
2003-04-12 00:17:11 +02:00
PS_FontInfoRec psfontinfo;
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#endif
#if HAVE_FT_GET_BDF_PROPERTY
Fix build error with BDF prop local. Free langset after query
2003-05-17 04:17:19 +02:00
BDF_PropertyRec prop;
#endif
Add fontversion field
2002-09-26 02:17:28 +02:00
TT_Header *head;
Update iso639-2 language coverage info, fix Georgian orthography to eliminate Mingrelian and Svan glyphs, use coverage for inclusion and OS/2 for Han exclusion, restructure fclang.c to use fclang.h from fc-lang dir
2002-07-08 09:31:53 +02:00
const FcChar8 *exclusiveLang = 0;
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
FT_SfntName sname;
FT_UInt snamei, snamec;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
int nfamily = 0;
int nfamily_lang = 0;
int nstyle = 0;
int nstyle_lang = 0;
int nfullname = 0;
int nfullname_lang = 0;
Fix sign-compare warnings
2012-12-30 04:47:49 +01:00
unsigned int p, n;
int platform, nameid;
Initial revision
2002-02-15 00:34:13 +01:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
FcChar8 *style = 0;
int st;
Bug 38737 - Wishlist: support FC_POSTSCRIPT_NAME Add the PostScript name into the cache and the matcher. Scoring the better font against the PostScript name by the forward-matching.
2013-01-18 03:30:10 +01:00
char psname[256];
const char *tmp;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Bug 71287 - size specific design selection support in OS/2 table version 5 This feature requires the FreeType 2.5.1 or later at the build time. Besides <range> element allows <double> elements with this changes. This may breaks the cache but not bumping in this change sets at this moment. please be aware if you want to try it and run fc-cache before/after to avoid the weird thing against it.
2013-11-20 10:44:59 +01:00
FcRange *r = NULL;
double lower_size = 0.0L, upper_size = DBL_MAX;
Bug 50733 - Add font-file hash? Add "hash" object which contains SHA256 hash value (so far) computed from the font file.
2013-02-04 08:03:29 +01:00
Call FcInitDebug from FcFreeTypeQueryFace
2014-07-24 21:01:57 +02:00
FcInitDebug (); /* We might be called with no initizalization whatsoever. */
Initial revision
2002-02-15 00:34:13 +01:00
pat = FcPatternCreate ();
if (!pat)
goto bail0;
if (!FcPatternAddBool (pat, FC_OUTLINE,
(face->face_flags & FT_FACE_FLAG_SCALABLE) != 0))
goto bail1;
if (!FcPatternAddBool (pat, FC_SCALABLE,
(face->face_flags & FT_FACE_FLAG_SCALABLE) != 0))
goto bail1;
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
/*
* Get the OS/2 table
*/
os2 = (TT_OS2 *) FT_Get_Sfnt_Table (face, ft_sfnt_os2);
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
/*
* Look first in the OS/2 table for the foundry, if
* not found here, the various notices will be searched for
* that information, either from the sfnt name tables or
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
* the Postscript FontInfo dictionary. Finally, the
* BDF properties will queried.
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
*/
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
if (os2 && os2->version >= 0x0001 && os2->version != 0xffff)
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
foundry = FcVendorFoundry(os2->achVendID);
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
if (FcDebug () & FC_DBG_SCANV)
printf ("\n");
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
/*
* Grub through the name table looking for family
* and style names. FreeType makes quite a hash
* of them
*/
snamec = FT_Get_Sfnt_Name_Count (face);
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
for (p = 0; p <= NUM_PLATFORM_ORDER; p++)
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
{
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
if (p < NUM_PLATFORM_ORDER)
platform = platform_order[p];
else
platform = 0xffff;
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
/*
* Order nameids so preferred names appear first
* in the resulting list
*/
for (n = 0; n < NUM_NAMEID_ORDER; n++)
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
nameid = nameid_order[n];
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
for (snamei = 0; snamei < snamec; snamei++)
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{
Bug 50525 - superfluous whitespace in the style
2012-06-08 12:41:59 +02:00
FcChar8 *utf8, *pp;
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
const FcChar8 *lang;
const char *elt = 0, *eltlang = 0;
int *np = 0, *nlangp = 0;
Bug 50525 - superfluous whitespace in the style
2012-06-08 12:41:59 +02:00
size_t len;
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
if (FT_Get_Sfnt_Name (face, snamei, &sname) != 0)
continue;
if (sname.name_id != nameid)
continue;
/*
* Sort platforms in preference order, accepting
* all other platforms last
*/
if (p < NUM_PLATFORM_ORDER)
{
if (sname.platform_id != platform)
continue;
}
else
{
Fix sign-compare warnings
2012-12-30 04:47:49 +01:00
unsigned int sp;
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
for (sp = 0; sp < NUM_PLATFORM_ORDER; sp++)
if (sname.platform_id == platform_order[sp])
break;
if (sp != NUM_PLATFORM_ORDER)
continue;
}
utf8 = FcSfntNameTranscode (&sname);
lang = FcSfntNameLanguage (&sname);
if (!utf8)
continue;
switch (sname.name_id) {
Use/prefer WWS family/style (name table id 21/22)
2009-06-02 03:14:56 +02:00
#ifdef TT_NAME_ID_WWS_FAMILY
case TT_NAME_ID_WWS_FAMILY:
#endif
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
case TT_NAME_ID_PREFERRED_FAMILY:
case TT_NAME_ID_FONT_FAMILY:
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
#if 0
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
case TT_NAME_ID_UNIQUE_ID:
#endif
if (FcDebug () & FC_DBG_SCANV)
printf ("found family (n %2d p %d e %d l 0x%04x) %s\n",
sname.name_id, sname.platform_id,
sname.encoding_id, sname.language_id,
utf8);
elt = FC_FAMILY;
eltlang = FC_FAMILYLANG;
np = &nfamily;
nlangp = &nfamily_lang;
break;
case TT_NAME_ID_MAC_FULL_NAME:
case TT_NAME_ID_FULL_NAME:
if (FcDebug () & FC_DBG_SCANV)
printf ("found full (n %2d p %d e %d l 0x%04x) %s\n",
sname.name_id, sname.platform_id,
sname.encoding_id, sname.language_id,
utf8);
elt = FC_FULLNAME;
eltlang = FC_FULLNAMELANG;
np = &nfullname;
nlangp = &nfullname_lang;
break;
Use/prefer WWS family/style (name table id 21/22)
2009-06-02 03:14:56 +02:00
#ifdef TT_NAME_ID_WWS_SUBFAMILY
case TT_NAME_ID_WWS_SUBFAMILY:
#endif
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
case TT_NAME_ID_PREFERRED_SUBFAMILY:
case TT_NAME_ID_FONT_SUBFAMILY:
Bug 50525 - superfluous whitespace in the style
2012-06-08 12:41:59 +02:00
if (utf8)
{
pp = utf8;
while (*pp == ' ')
pp++;
len = strlen ((const char *) pp);
memmove (utf8, pp, len + 1);
pp = utf8 + len - 1;
while (*pp == ' ')
pp--;
*(pp + 1) = 0;
}
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
if (FcDebug () & FC_DBG_SCANV)
printf ("found style (n %2d p %d e %d l 0x%04x) %s\n",
sname.name_id, sname.platform_id,
sname.encoding_id, sname.language_id,
utf8);
elt = FC_STYLE;
eltlang = FC_STYLELANG;
np = &nstyle;
nlangp = &nstyle_lang;
break;
case TT_NAME_ID_TRADEMARK:
case TT_NAME_ID_MANUFACTURER:
/* If the foundry wasn't found in the OS/2 table, look here */
if(!foundry)
foundry = FcNoticeFoundry((FT_String *) utf8);
break;
}
if (elt)
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
if (FcStringInPatternElement (pat, elt, utf8))
{
free (utf8);
continue;
}
/* add new element */
if (!FcPatternAddString (pat, elt, utf8))
{
free (utf8);
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
goto bail1;
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
}
free (utf8);
if (lang)
{
Use lang=und instead of lang=xx for "undetermined" That's the correct BCP 47 code.
2014-07-24 21:37:51 +02:00
/* pad lang list with 'und' to line up with elt */
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
while (*nlangp < *np)
{
Use lang=und instead of lang=xx for "undetermined" That's the correct BCP 47 code.
2014-07-24 21:37:51 +02:00
if (!FcPatternAddString (pat, eltlang, (FcChar8 *) "und"))
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
goto bail1;
++*nlangp;
}
if (!FcPatternAddString (pat, eltlang, lang))
goto bail1;
++*nlangp;
}
++*np;
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
}
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
else
free (utf8);
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
}
}
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
}
Use explicit platform/nameid order when scanning ttf files. Instead of accepting whatever order names appear in the font file, use an explicit ordering for both platform and nameid. Platforms are high precedence than nameids. The platform order is: microsoft, apple unicode, macintosh, (other) The family nameid order is: preferred family, font family The fullname nameid order is: mac full name, full name The style nameid order is preferred subfamily, font subfamily This will change the names visible to users in various application UIs, but should not change how existing font names are matched as all names remain present in the resulting database. The hope is that family names will, in general, be less ambiguous. Testing here shows that commercial fonts have longer names now while DejaVu has a shorter family name, and moves more of the font description to the style name.
2006-12-02 20:47:07 +01:00
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
if (!nfamily && face->family_name &&
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
FcStrCmpIgnoreBlanksAndCase ((FcChar8 *) face->family_name, (FcChar8 *) "") != 0)
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{
if (FcDebug () & FC_DBG_SCANV)
Check for non-empty face->family_name and face->style_name before using those for the font. Empty names match everything. Bug #171.
2004-12-05 05:11:11 +01:00
printf ("using FreeType family \"%s\"\n", face->family_name);
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
if (!FcPatternAddString (pat, FC_FAMILY, (FcChar8 *) face->family_name))
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
goto bail1;
++nfamily;
}
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Check for non-empty face->family_name and face->style_name before using those for the font. Empty names match everything. Bug #171.
2004-12-05 05:11:11 +01:00
if (!nstyle && face->style_name &&
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
FcStrCmpIgnoreBlanksAndCase ((FcChar8 *) face->style_name, (FcChar8 *) "") != 0)
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
{
if (FcDebug () & FC_DBG_SCANV)
Check for non-empty face->family_name and face->style_name before using those for the font. Empty names match everything. Bug #171.
2004-12-05 05:11:11 +01:00
printf ("using FreeType style \"%s\"\n", face->style_name);
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
if (!FcPatternAddString (pat, FC_STYLE, (FcChar8 *) face->style_name))
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
goto bail1;
++nstyle;
}
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Allow passing NULL for file to FcFreeTypeQueryFace()
2014-05-16 23:08:52 +02:00
if (!nfamily && file && *file)
Initial revision
2002-02-15 00:34:13 +01:00
{
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
FcChar8 *start, *end;
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
FcChar8 *family;
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
start = (FcChar8 *) strrchr ((char *) file, '/');
if (start)
start++;
Initial revision
2002-02-15 00:34:13 +01:00
else
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
start = (FcChar8 *) file;
end = (FcChar8 *) strrchr ((char *) start, '.');
if (!end)
end = start + strlen ((char *) start);
More complete memory tracking. Install always overwrites header files
2002-09-01 00:17:32 +02:00
/* freed below */
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
family = malloc (end - start + 1);
strncpy ((char *) family, (char *) start, end - start);
family[end - start] = '\0';
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
if (FcDebug () & FC_DBG_SCANV)
Check for non-empty face->family_name and face->style_name before using those for the font. Empty names match everything. Bug #171.
2004-12-05 05:11:11 +01:00
printf ("using filename for family %s\n", family);
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
if (!FcPatternAddString (pat, FC_FAMILY, family))
{
free (family);
goto bail1;
}
free (family);
++nfamily;
Initial revision
2002-02-15 00:34:13 +01:00
}
Add some Utf16 support, extract font family and style names from name table for sfnt fonts
2002-07-13 07:43:25 +02:00
Bug 38737 - Wishlist: support FC_POSTSCRIPT_NAME Add the PostScript name into the cache and the matcher. Scoring the better font against the PostScript name by the forward-matching.
2013-01-18 03:30:10 +01:00
/* Add the PostScript name into the cache */
tmp = FT_Get_Postscript_Name (face);
if (!tmp)
{
FcChar8 *family, *familylang = NULL;
size_t len;
int n = 0;
/* Workaround when FT_Get_Postscript_Name didn't give any name.
* try to find out the English family name and convert.
*/
while (FcPatternObjectGetString (pat, FC_FAMILYLANG_OBJECT, n, &familylang) == FcResultMatch)
{
if (FcStrCmp (familylang, (const FcChar8 *)"en") == 0)
break;
n++;
familylang = NULL;
}
if (!familylang)
n = 0;
if (FcPatternObjectGetString (pat, FC_FAMILY_OBJECT, n, &family) != FcResultMatch)
goto bail1;
len = strlen ((const char *)family);
/* the literal name in PostScript Language is limited to 127 characters though,
* It is the architectural limit. so assuming 255 characters may works enough.
*/
for (i = 0; i < len && i < 255; i++)
{
/* those characters are not allowed to be the literal name in PostScript */
static const char exclusive_chars[] = "\x04()/<>[]{}\t\f\r\n ";
if (strchr(exclusive_chars, family[i]) != NULL)
psname[i] = '-';
else
psname[i] = family[i];
}
psname[i] = 0;
}
else
{
strcpy (psname, tmp);
}
if (!FcPatternAddString (pat, FC_POSTSCRIPT_NAME, (const FcChar8 *)psname))
goto bail1;
Allow passing NULL for file to FcFreeTypeQueryFace()
2014-05-16 23:08:52 +02:00
if (file && *file && !FcPatternAddString (pat, FC_FILE, file))
Initial revision
2002-02-15 00:34:13 +01:00
goto bail1;
Reinstate basename patch, but keep a hash table linking FcPatterns to their fully-qualified font names for clients' benefit. Clients only pay for the font names once they request the FC_FILE property from an FcPattern, but the font name is malloc'd at that point (i.e. not mmapped: that's impossible, since it may vary between machines.) Clients do have to pay for a copy of the path name per cache file. Note that FcPatternGetString now does some rewriting if you ask for an FC_FILE, appending the pathname as appropriate.
2005-11-01 06:26:27 +01:00
Initial revision
2002-02-15 00:34:13 +01:00
if (!FcPatternAddInteger (pat, FC_INDEX, id))
goto bail1;
Stop setting FC_SPACING from font hints. Theyre always wrong
2003-02-27 08:04:59 +01:00
#if 0
/*
* don't even try this -- CJK 'monospace' fonts are really
* dual width, and most other fonts don't bother to set
* the attribute. Sigh.
*/
Initial revision
2002-02-15 00:34:13 +01:00
if ((face->face_flags & FT_FACE_FLAG_FIXED_WIDTH) != 0)
if (!FcPatternAddInteger (pat, FC_SPACING, FC_MONO))
goto bail1;
#endif
Add fontversion field
2002-09-26 02:17:28 +02:00
/*
* Find the font revision (if available)
*/
head = (TT_Header *) FT_Get_Sfnt_Table (face, ft_sfnt_head);
if (head)
{
if (!FcPatternAddInteger (pat, FC_FONTVERSION, head->Font_Revision))
goto bail1;
}
else
{
if (!FcPatternAddInteger (pat, FC_FONTVERSION, 0))
goto bail1;
}
Add strong/weak pattern value binding, add known charsets for automatic lang computation
2002-06-29 22:31:02 +02:00
if (os2 && os2->version >= 0x0001 && os2->version != 0xffff)
{
for (i = 0; i < NUM_CODE_PAGE_RANGE; i++)
{
FT_ULong bits;
int bit;
if (FcCodePageRange[i].bit < 32)
{
bits = os2->ulCodePageRange1;
bit = FcCodePageRange[i].bit;
}
else
{
bits = os2->ulCodePageRange2;
bit = FcCodePageRange[i].bit - 32;
}
if (bits & (1 << bit))
{
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
/*
Update iso639-2 language coverage info, fix Georgian orthography to eliminate Mingrelian and Svan glyphs, use coverage for inclusion and OS/2 for Han exclusion, restructure fclang.c to use fclang.h from fc-lang dir
2002-07-08 09:31:53 +02:00
* If the font advertises support for multiple
* "exclusive" languages, then include support
* for any language found to have coverage
*/
if (exclusiveLang)
{
exclusiveLang = 0;
break;
}
exclusiveLang = FcCodePageRange[i].lang;
Add strong/weak pattern value binding, add known charsets for automatic lang computation
2002-06-29 22:31:02 +02:00
}
}
}
add font widths and extend weight from OS/2 table
2003-03-12 23:16:43 +01:00
if (os2 && os2->version != 0xffff)
{
Linearly interpolate weight values Rest of Part of https://bugs.freedesktop.org/show_bug.cgi?id=81453 Adds new API: FcWeightFromOpenType() FcWeightToOpenType()
2014-07-25 23:59:26 +02:00
weight = FcWeightFromOpenType (os2->usWeightClass);
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
if ((FcDebug() & FC_DBG_SCANV) && weight != -1)
printf ("\tos2 weight class %d maps to weight %d\n",
os2->usWeightClass, weight);
add font widths and extend weight from OS/2 table
2003-03-12 23:16:43 +01:00
switch (os2->usWidthClass) {
case 1: width = FC_WIDTH_ULTRACONDENSED; break;
case 2: width = FC_WIDTH_EXTRACONDENSED; break;
case 3: width = FC_WIDTH_CONDENSED; break;
case 4: width = FC_WIDTH_SEMICONDENSED; break;
case 5: width = FC_WIDTH_NORMAL; break;
case 6: width = FC_WIDTH_SEMIEXPANDED; break;
case 7: width = FC_WIDTH_EXPANDED; break;
case 8: width = FC_WIDTH_EXTRAEXPANDED; break;
case 9: width = FC_WIDTH_ULTRAEXPANDED; break;
}
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
if ((FcDebug() & FC_DBG_SCANV) && width != -1)
printf ("\tos2 width class %d maps to width %d\n",
os2->usWidthClass, width);
add font widths and extend weight from OS/2 table
2003-03-12 23:16:43 +01:00
}
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
if (os2 && (complex_ = FcFontCapabilities(face)))
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
{
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
if (!FcPatternAddString (pat, FC_CAPABILITY, complex_))
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
{
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
free (complex_);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
goto bail1;
}
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
free (complex_);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
}
Bug #46, #47 fontconfig should retrieve type 1 font information from FontInfo dictionary Patch provided by g2@magestudios.net (Gerard Escalante)
2003-04-12 00:17:11 +02:00
Bug 71287 - size specific design selection support in OS/2 table version 5 This feature requires the FreeType 2.5.1 or later at the build time. Besides <range> element allows <double> elements with this changes. This may breaks the cache but not bumping in this change sets at this moment. please be aware if you want to try it and run fc-cache before/after to avoid the weird thing against it.
2013-11-20 10:44:59 +01:00
#if defined (HAVE_TT_OS2_USUPPEROPTICALPOINTSIZE) && defined (HAVE_TT_OS2_USLOWEROPTICALPOINTSIZE)
if (os2 && os2->version >= 0x0005 && os2->version != 0xffff)
{
/* usLowerPointSize and usUpperPointSize is actually twips */
lower_size = os2->usLowerOpticalPointSize / 20.0L;
upper_size = os2->usUpperOpticalPointSize / 20.0L;
}
#endif
if (os2)
{
r = FcRangeCreateDouble (lower_size, upper_size);
if (!FcPatternAddRange (pat, FC_SIZE, r))
{
FcRangeDestroy (r);
goto bail1;
}
FcRangeDestroy (r);
}
else
{
for (i = 0; i < face->num_fixed_sizes; i++)
{
double d = FcGetPixelSize (face, i);
r = FcRangeCreateDouble (d, d);
if (!FcPatternAddRange (pat, FC_SIZE, r))
{
FcRangeDestroy (r);
goto bail1;
}
FcRangeDestroy (r);
}
}
Bug #46, #47 fontconfig should retrieve type 1 font information from FontInfo dictionary Patch provided by g2@magestudios.net (Gerard Escalante)
2003-04-12 00:17:11 +02:00
/*
* Type 1: Check for FontInfo dictionary information
* Code from g2@magestudios.net (Gerard Escalante)
*/
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#if HAVE_FT_GET_PS_FONT_INFO
Bug #46, #47 fontconfig should retrieve type 1 font information from FontInfo dictionary Patch provided by g2@magestudios.net (Gerard Escalante)
2003-04-12 00:17:11 +02:00
if (FT_Get_PS_Font_Info(face, &psfontinfo) == 0)
{
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
if (weight == -1 && psfontinfo.weight)
Clean up ps font weight matching and check for NULL
2003-04-15 19:01:39 +02:00
{
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
weight = FcIsWeight ((FcChar8 *) psfontinfo.weight);
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
if (FcDebug() & FC_DBG_SCANV)
printf ("\tType1 weight %s maps to %d\n",
psfontinfo.weight, weight);
Clean up ps font weight matching and check for NULL
2003-04-15 19:01:39 +02:00
}
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Debug output for unknown ps weight names. ignore italic_angle for PS fonts as FreeType already checks that
2003-04-24 19:31:03 +02:00
#if 0
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
/*
Debug output for unknown ps weight names. ignore italic_angle for PS fonts as FreeType already checks that
2003-04-24 19:31:03 +02:00
* Don't bother with italic_angle; FreeType already extracts that
* information for us and sticks it into style_flags
*/
if (psfontinfo.italic_angle)
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
slant = FC_SLANT_ITALIC;
Debug output for unknown ps weight names. ignore italic_angle for PS fonts as FreeType already checks that
2003-04-24 19:31:03 +02:00
else
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
slant = FC_SLANT_ROMAN;
Debug output for unknown ps weight names. ignore italic_angle for PS fonts as FreeType already checks that
2003-04-24 19:31:03 +02:00
#endif
Implemented foundry generation for Type 1 and TrueType
2003-04-12 01:45:59 +02:00
if(!foundry)
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
foundry = FcNoticeFoundry(psfontinfo.notice);
Bug #46, #47 fontconfig should retrieve type 1 font information from FontInfo dictionary Patch provided by g2@magestudios.net (Gerard Escalante)
2003-04-12 00:17:11 +02:00
}
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#endif /* HAVE_FT_GET_PS_FONT_INFO */
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#if HAVE_FT_GET_BDF_PROPERTY
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
/*
* Finally, look for a FOUNDRY BDF property if no other
* mechanism has managed to locate a foundry
*/
if (!foundry)
{
int rc;
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
rc = FT_Get_BDF_Property(face, "FOUNDRY", &prop);
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
if(rc == 0 && prop.type == BDF_PROPERTY_TYPE_ATOM)
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
foundry = (FcChar8 *) prop.u.atom;
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
}
if (width == -1)
{
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
if (FT_Get_BDF_Property(face, "RELATIVE_SETWIDTH", &prop) == 0 &&
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
(prop.type == BDF_PROPERTY_TYPE_INTEGER ||
prop.type == BDF_PROPERTY_TYPE_CARDINAL))
{
FT_Int32 value;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
if (prop.type == BDF_PROPERTY_TYPE_INTEGER)
value = prop.u.integer;
else
value = (FT_Int32) prop.u.cardinal;
switch ((value + 5) / 10) {
case 1: width = FC_WIDTH_ULTRACONDENSED; break;
case 2: width = FC_WIDTH_EXTRACONDENSED; break;
case 3: width = FC_WIDTH_CONDENSED; break;
case 4: width = FC_WIDTH_SEMICONDENSED; break;
case 5: width = FC_WIDTH_NORMAL; break;
case 6: width = FC_WIDTH_SEMIEXPANDED; break;
case 7: width = FC_WIDTH_EXPANDED; break;
case 8: width = FC_WIDTH_EXTRAEXPANDED; break;
case 9: width = FC_WIDTH_ULTRAEXPANDED; break;
}
}
Use FcIsWidth to share code Set FT_LOAD_IGNORE_GLOBAL_ADVANCE_WIDTH when scanning fonts to avoid misclassifying some Han fonts as monospaced.
2003-05-12 22:48:59 +02:00
if (width == -1 &&
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
FT_Get_BDF_Property (face, "SETWIDTH_NAME", &prop) == 0 &&
Fix a crash issue when empty strings are set to the BDF properties
2014-01-31 03:10:02 +01:00
prop.type == BDF_PROPERTY_TYPE_ATOM && prop.u.atom != NULL)
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
{
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
width = FcIsWidth ((FcChar8 *) prop.u.atom);
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
if (FcDebug () & FC_DBG_SCANV)
Use FcIsWidth to share code Set FT_LOAD_IGNORE_GLOBAL_ADVANCE_WIDTH when scanning fonts to avoid misclassifying some Han fonts as monospaced.
2003-05-12 22:48:59 +02:00
printf ("\tsetwidth %s maps to %d\n", prop.u.atom, width);
Search through the BDF properties for width and foundry information
2003-04-16 20:08:47 +02:00
}
}
#endif
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
/*
* Look for weight, width and slant names in the style value
*/
Move existing fonts.conf to fonts.conf.bak Add detection of iconv Document new selectfont elements Switch to UTF-8 in comment Add fullname, and family/style/fullname language entries Respect selectfont/*/glob Add support for selectfont Add multi-lingual family/style/fullname support Expose FcListPatternMatchAny (which selectfont/*/pattern uses) Add new FcPatternRemove/FcPatternAppend. FcObjectStaticName stores computed pattern element names which are required to be static.
2004-12-04 20:41:10 +01:00
for (st = 0; FcPatternGetString (pat, FC_STYLE, st, &style) == FcResultMatch; st++)
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
{
if (weight == -1)
{
weight = FcContainsWeight (style);
if (FcDebug() & FC_DBG_SCANV)
printf ("\tStyle %s maps to weight %d\n", style, weight);
}
if (width == -1)
{
width = FcContainsWidth (style);
if (FcDebug() & FC_DBG_SCANV)
printf ("\tStyle %s maps to width %d\n", style, width);
}
if (slant == -1)
{
slant = FcContainsSlant (style);
if (FcDebug() & FC_DBG_SCANV)
printf ("\tStyle %s maps to slant %d\n", style, slant);
}
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
if (decorative == FcFalse)
{
decorative = FcContainsDecorative (style) > 0;
if (FcDebug() & FC_DBG_SCANV)
printf ("\tStyle %s maps to decorative %d\n", style, decorative);
}
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
}
/*
* Pull default values from the FreeType flags if more
* specific values not found above
*/
if (slant == -1)
{
slant = FC_SLANT_ROMAN;
if (face->style_flags & FT_STYLE_FLAG_ITALIC)
slant = FC_SLANT_ITALIC;
}
if (weight == -1)
{
weight = FC_WEIGHT_MEDIUM;
if (face->style_flags & FT_STYLE_FLAG_BOLD)
weight = FC_WEIGHT_BOLD;
}
Force FC_FOUNDRY and FC_WIDTH to always be set so that matches looking for explicit values prefer exact matches
2004-03-07 00:44:11 +01:00
if (width == -1)
width = FC_WIDTH_NORMAL;
if (foundry == 0)
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
foundry = (FcChar8 *) "unknown";
Force FC_FOUNDRY and FC_WIDTH to always be set so that matches looking for explicit values prefer exact matches
2004-03-07 00:44:11 +01:00
add font widths and extend weight from OS/2 table
2003-03-12 23:16:43 +01:00
if (!FcPatternAddInteger (pat, FC_SLANT, slant))
goto bail1;
if (!FcPatternAddInteger (pat, FC_WEIGHT, weight))
goto bail1;
Force FC_FOUNDRY and FC_WIDTH to always be set so that matches looking for explicit values prefer exact matches
2004-03-07 00:44:11 +01:00
if (!FcPatternAddInteger (pat, FC_WIDTH, width))
goto bail1;
add font widths and extend weight from OS/2 table
2003-03-12 23:16:43 +01:00
Force FC_FOUNDRY and FC_WIDTH to always be set so that matches looking for explicit values prefer exact matches
2004-03-07 00:44:11 +01:00
if (!FcPatternAddString (pat, FC_FOUNDRY, foundry))
goto bail1;
Move foundry detection data into fcfreetype.c (which is getting rather large at this point)
2003-04-16 18:19:38 +02:00
Add FcMatchScan to resolve Delicious font matching issues (bug #6769) The Delicious family includes one named Delicious Heavy, a bold variant which is unfortunately marked as having normal weight. Because the family name is 'Delicious', fontconfig accidentally selects this font instead of the normal weight variant. The fix here rewrites the scanned data by running the scanned pattern through a new substitution sequence tagged with <match target=scan>; a sample for the Delicious family is included to demonstrate how it works (and fix Delicious at the same time). Also added was a new match predicate -- the 'decorative' predicate which is automatically detected in fonts by searching style names for key decorative phrases like SmallCaps, Shadow, Embosed and Antiqua. Suggestions for additional decorative key words are welcome. This should have little effect on font matching except when two fonts share the same characteristics except for this value.
2006-09-03 02:52:12 +02:00
if (!FcPatternAddBool (pat, FC_DECORATIVE, decorative))
goto bail1;
Rewrite hashing to use FT_Stream directly This is more robust but introduces a small change in behavior: For .pcf.gz fonts, the new code calculates the hash of the uncompressed font data whereas the original code was calculating the hash of the compressed data. No big deal IMO.
2014-05-16 23:02:58 +02:00
Add strong/weak pattern value binding, add known charsets for automatic lang computation
2002-06-29 22:31:02 +02:00
/*
* Compute the unicode coverage for the font
*/
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
cs = FcFreeTypeCharSetAndSpacing (face, blanks, &spacing);
Initial revision
2002-02-15 00:34:13 +01:00
if (!cs)
goto bail1;
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#if HAVE_FT_GET_BDF_PROPERTY
Extract spacing from XLFD atom
2003-05-14 22:23:24 +02:00
/* For PCF fonts, override the computed spacing with the one from
the property */
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
if(FT_Get_BDF_Property(face, "SPACING", &prop) == 0 &&
Fix a crash issue when empty strings are set to the BDF properties
2014-01-31 03:10:02 +01:00
prop.type == BDF_PROPERTY_TYPE_ATOM && prop.u.atom != NULL) {
Supplied by: mfabian@suse.de (Mike FABIAN) Bug #260 fc-cache generates wrong spacing values for bitmap fonts Was using (strcmp (a,b)) instead of (!strcmp(a,b)).
2004-03-02 17:48:51 +01:00
if(!strcmp(prop.u.atom, "c") || !strcmp(prop.u.atom, "C"))
Extract spacing from XLFD atom
2003-05-14 22:23:24 +02:00
spacing = FC_CHARCELL;
Supplied by: mfabian@suse.de (Mike FABIAN) Bug #260 fc-cache generates wrong spacing values for bitmap fonts Was using (strcmp (a,b)) instead of (!strcmp(a,b)).
2004-03-02 17:48:51 +01:00
else if(!strcmp(prop.u.atom, "m") || !strcmp(prop.u.atom, "M"))
Extract spacing from XLFD atom
2003-05-14 22:23:24 +02:00
spacing = FC_MONO;
Supplied by: mfabian@suse.de (Mike FABIAN) Bug #260 fc-cache generates wrong spacing values for bitmap fonts Was using (strcmp (a,b)) instead of (!strcmp(a,b)).
2004-03-02 17:48:51 +01:00
else if(!strcmp(prop.u.atom, "p") || !strcmp(prop.u.atom, "P"))
Extract spacing from XLFD atom
2003-05-14 22:23:24 +02:00
spacing = FC_PROPORTIONAL;
}
#endif
Initial revision
2002-02-15 00:34:13 +01:00
/*
* Skip over PCF fonts that have no encoded characters; they're
* usually just Unicode fonts transcoded to some legacy encoding
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
* FT forces us to approximate whether a font is a PCF font
Restore code to skip over PCF fonts that have no encoded characters. (We guess that a font is PCF by presence of the PIXEL_SIZE BDF property.)
2005-12-20 21:35:47 +01:00
* or not by whether it has any BDF properties. Try PIXEL_SIZE;
* I don't know how to get a list of BDF properties on the font. -PL
Initial revision
2002-02-15 00:34:13 +01:00
*/
Update iso639-2 language coverage info, fix Georgian orthography to eliminate Mingrelian and Svan glyphs, use coverage for inclusion and OS/2 for Han exclusion, restructure fclang.c to use fclang.h from fc-lang dir
2002-07-08 09:31:53 +02:00
if (FcCharSetCount (cs) == 0)
Initial revision
2002-02-15 00:34:13 +01:00
{
Restore code to skip over PCF fonts that have no encoded characters. (We guess that a font is PCF by presence of the PIXEL_SIZE BDF property.)
2005-12-20 21:35:47 +01:00
#if HAVE_FT_GET_BDF_PROPERTY
if(FT_Get_BDF_Property(face, "PIXEL_SIZE", &prop) == 0)
Initial revision
2002-02-15 00:34:13 +01:00
goto bail2;
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
#endif
Restore code to skip over PCF fonts that have no encoded characters. (We guess that a font is PCF by presence of the PIXEL_SIZE BDF property.)
2005-12-20 21:35:47 +01:00
}
Initial revision
2002-02-15 00:34:13 +01:00
if (!FcPatternAddCharSet (pat, FC_CHARSET, cs))
goto bail2;
Add strong/weak pattern value binding, add known charsets for automatic lang computation
2002-06-29 22:31:02 +02:00
Reimplement FC_LANG as FcTypeLang, freeze patterns, other cleanup
2002-08-22 09:36:45 +02:00
ls = FcFreeTypeLangSet (cs, exclusiveLang);
if (!ls)
goto bail2;
if (!FcPatternAddLangSet (pat, FC_LANG, ls))
Fix build error with BDF prop local. Free langset after query
2003-05-17 04:17:19 +02:00
{
FcLangSetDestroy (ls);
Reimplement FC_LANG as FcTypeLang, freeze patterns, other cleanup
2002-08-22 09:36:45 +02:00
goto bail2;
Fix build error with BDF prop local. Free langset after query
2003-05-17 04:17:19 +02:00
}
FcLangSetDestroy (ls);
Switch to RFC 3066 based lang names
2002-07-07 01:47:44 +02:00
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
if (spacing != FC_PROPORTIONAL)
if (!FcPatternAddInteger (pat, FC_SPACING, spacing))
goto bail2;
Initial revision
2002-02-15 00:34:13 +01:00
if (!(face->face_flags & FT_FACE_FLAG_SCALABLE))
{
for (i = 0; i < face->num_fixed_sizes; i++)
if (!FcPatternAddDouble (pat, FC_PIXEL_SIZE,
Add Low Saxon orthography (Kenneth Rohde Christiansen <kenneth@gnu.org>) Oops. Left 'newest.set' unset, which would miscompute the newest file Add FcGetPixelSize to extract correct pixel size from bdf/pcf font properties (which report the wrong value in current FreeType) Don't attempt to check for empty glyphs in non-scalable fonts; they have no outlines...
2004-02-07 08:13:48 +01:00
FcGetPixelSize (face, i)))
Fix memory leak in error condition code.
2006-02-13 23:19:30 +01:00
goto bail2;
Initial revision
2002-02-15 00:34:13 +01:00
if (!FcPatternAddBool (pat, FC_ANTIALIAS, FcFalse))
Fix memory leak in error condition code.
2006-02-13 23:19:30 +01:00
goto bail2;
Initial revision
2002-02-15 00:34:13 +01:00
}
Create FC_FONTFORMAT from FT_Get_X11_Font_Format function where available. This provides font file format information (BDF, Type 1, PCF, TrueType) for each font. Closes #109.
2004-12-05 01:26:06 +01:00
#if HAVE_FT_GET_X11_FONT_FORMAT
/*
* Use the (not well documented or supported) X-specific function
* from FreeType to figure out the font format
*/
{
const char *font_format = FT_Get_X11_Font_Format (face);
if (font_format)
Minor
2013-01-16 14:05:07 +01:00
if (!FcPatternAddString (pat, FC_FONTFORMAT, (FcChar8 *) font_format))
goto bail2;
Create FC_FONTFORMAT from FT_Get_X11_Font_Format function where available. This provides font file format information (BDF, Type 1, PCF, TrueType) for each font. Closes #109.
2004-12-05 01:26:06 +01:00
}
#endif
Initial revision
2002-02-15 00:34:13 +01:00
Grub through style to find weight/slant/width values when other techniques fail
2003-05-02 03:11:53 +02:00
/*
* Drop our reference to the charset
*/
FcCharSetDestroy (cs);
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Initial revision
2002-02-15 00:34:13 +01:00
return pat;
bail2:
FcCharSetDestroy (cs);
bail1:
FcPatternDestroy (pat);
bail0:
Add FcFreeTypeQueryFace external API. Bug #7311. Expose ability to build an FcPattern directly from an FT_Face object.
2006-12-02 22:22:27 +01:00
return NULL;
}
FcPattern *
FcFreeTypeQuery(const FcChar8 *file,
int id,
FcBlanks *blanks,
int *count)
{
FT_Face face;
FT_Library ftLibrary;
FcPattern *pat = NULL;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Add FcFreeTypeQueryFace external API. Bug #7311. Expose ability to build an FcPattern directly from an FT_Face object.
2006-12-02 22:22:27 +01:00
if (FT_Init_FreeType (&ftLibrary))
return NULL;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Add FcFreeTypeQueryFace external API. Bug #7311. Expose ability to build an FcPattern directly from an FT_Face object.
2006-12-02 22:22:27 +01:00
if (FT_New_Face (ftLibrary, (char *) file, id, &face))
goto bail;
*count = face->num_faces;
pat = FcFreeTypeQueryFace (face, file, id, blanks);
Initial revision
2002-02-15 00:34:13 +01:00
FT_Done_Face (face);
bail:
FT_Done_FreeType (ftLibrary);
Add FcFreeTypeQueryFace external API. Bug #7311. Expose ability to build an FcPattern directly from an FT_Face object.
2006-12-02 22:22:27 +01:00
return pat;
Initial revision
2002-02-15 00:34:13 +01:00
}
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
/*
* For our purposes, this approximation is sufficient
*/
Lean on autoconf to find useful FreeType functions (bug 95)
2003-06-26 01:21:03 +02:00
#if !HAVE_FT_GET_NEXT_CHAR
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
#define FT_Get_Next_Char(face, ucs4, gi) ((ucs4) >= 0xffffff ? \
(*(gi) = 0), 0 : \
(*(gi) = 1), (ucs4) + 1)
Add FcResultOutOfMemory to provide an accurate error when FcFontSetMatch fails in this way Make #warning about lacking various FreeType features indicate which version those features appeared so users know how to fix the problem (Thanks to Anton Tropashko)
2004-05-29 21:32:41 +02:00
#warning "No FT_Get_Next_Char: Please install freetype version 2.1.0 or newer"
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
#endif
typedef struct _FcCharEnt {
FcChar16 bmp;
unsigned char encode;
} FcCharEnt;
struct _FcCharMap {
const FcCharEnt *ent;
int nent;
};
typedef struct _FcFontDecode {
FT_Encoding encoding;
const FcCharMap *map;
FcChar32 max;
} FcFontDecode;
static const FcCharEnt AdobeSymbolEnt[] = {
{ 0x0020, 0x20 }, /* SPACE # space */
{ 0x0021, 0x21 }, /* EXCLAMATION MARK # exclam */
{ 0x0023, 0x23 }, /* NUMBER SIGN # numbersign */
{ 0x0025, 0x25 }, /* PERCENT SIGN # percent */
{ 0x0026, 0x26 }, /* AMPERSAND # ampersand */
{ 0x0028, 0x28 }, /* LEFT PARENTHESIS # parenleft */
{ 0x0029, 0x29 }, /* RIGHT PARENTHESIS # parenright */
{ 0x002B, 0x2B }, /* PLUS SIGN # plus */
{ 0x002C, 0x2C }, /* COMMA # comma */
{ 0x002E, 0x2E }, /* FULL STOP # period */
{ 0x002F, 0x2F }, /* SOLIDUS # slash */
{ 0x0030, 0x30 }, /* DIGIT ZERO # zero */
{ 0x0031, 0x31 }, /* DIGIT ONE # one */
{ 0x0032, 0x32 }, /* DIGIT TWO # two */
{ 0x0033, 0x33 }, /* DIGIT THREE # three */
{ 0x0034, 0x34 }, /* DIGIT FOUR # four */
{ 0x0035, 0x35 }, /* DIGIT FIVE # five */
{ 0x0036, 0x36 }, /* DIGIT SIX # six */
{ 0x0037, 0x37 }, /* DIGIT SEVEN # seven */
{ 0x0038, 0x38 }, /* DIGIT EIGHT # eight */
{ 0x0039, 0x39 }, /* DIGIT NINE # nine */
{ 0x003A, 0x3A }, /* COLON # colon */
{ 0x003B, 0x3B }, /* SEMICOLON # semicolon */
{ 0x003C, 0x3C }, /* LESS-THAN SIGN # less */
{ 0x003D, 0x3D }, /* EQUALS SIGN # equal */
{ 0x003E, 0x3E }, /* GREATER-THAN SIGN # greater */
{ 0x003F, 0x3F }, /* QUESTION MARK # question */
{ 0x005B, 0x5B }, /* LEFT SQUARE BRACKET # bracketleft */
{ 0x005D, 0x5D }, /* RIGHT SQUARE BRACKET # bracketright */
{ 0x005F, 0x5F }, /* LOW LINE # underscore */
{ 0x007B, 0x7B }, /* LEFT CURLY BRACKET # braceleft */
{ 0x007C, 0x7C }, /* VERTICAL LINE # bar */
{ 0x007D, 0x7D }, /* RIGHT CURLY BRACKET # braceright */
{ 0x00A0, 0x20 }, /* NO-BREAK SPACE # space */
{ 0x00AC, 0xD8 }, /* NOT SIGN # logicalnot */
{ 0x00B0, 0xB0 }, /* DEGREE SIGN # degree */
{ 0x00B1, 0xB1 }, /* PLUS-MINUS SIGN # plusminus */
{ 0x00B5, 0x6D }, /* MICRO SIGN # mu */
{ 0x00D7, 0xB4 }, /* MULTIPLICATION SIGN # multiply */
{ 0x00F7, 0xB8 }, /* DIVISION SIGN # divide */
{ 0x0192, 0xA6 }, /* LATIN SMALL LETTER F WITH HOOK # florin */
{ 0x0391, 0x41 }, /* GREEK CAPITAL LETTER ALPHA # Alpha */
{ 0x0392, 0x42 }, /* GREEK CAPITAL LETTER BETA # Beta */
{ 0x0393, 0x47 }, /* GREEK CAPITAL LETTER GAMMA # Gamma */
{ 0x0394, 0x44 }, /* GREEK CAPITAL LETTER DELTA # Delta */
{ 0x0395, 0x45 }, /* GREEK CAPITAL LETTER EPSILON # Epsilon */
{ 0x0396, 0x5A }, /* GREEK CAPITAL LETTER ZETA # Zeta */
{ 0x0397, 0x48 }, /* GREEK CAPITAL LETTER ETA # Eta */
{ 0x0398, 0x51 }, /* GREEK CAPITAL LETTER THETA # Theta */
{ 0x0399, 0x49 }, /* GREEK CAPITAL LETTER IOTA # Iota */
{ 0x039A, 0x4B }, /* GREEK CAPITAL LETTER KAPPA # Kappa */
{ 0x039B, 0x4C }, /* GREEK CAPITAL LETTER LAMDA # Lambda */
{ 0x039C, 0x4D }, /* GREEK CAPITAL LETTER MU # Mu */
{ 0x039D, 0x4E }, /* GREEK CAPITAL LETTER NU # Nu */
{ 0x039E, 0x58 }, /* GREEK CAPITAL LETTER XI # Xi */
{ 0x039F, 0x4F }, /* GREEK CAPITAL LETTER OMICRON # Omicron */
{ 0x03A0, 0x50 }, /* GREEK CAPITAL LETTER PI # Pi */
{ 0x03A1, 0x52 }, /* GREEK CAPITAL LETTER RHO # Rho */
{ 0x03A3, 0x53 }, /* GREEK CAPITAL LETTER SIGMA # Sigma */
{ 0x03A4, 0x54 }, /* GREEK CAPITAL LETTER TAU # Tau */
{ 0x03A5, 0x55 }, /* GREEK CAPITAL LETTER UPSILON # Upsilon */
{ 0x03A6, 0x46 }, /* GREEK CAPITAL LETTER PHI # Phi */
{ 0x03A7, 0x43 }, /* GREEK CAPITAL LETTER CHI # Chi */
{ 0x03A8, 0x59 }, /* GREEK CAPITAL LETTER PSI # Psi */
{ 0x03A9, 0x57 }, /* GREEK CAPITAL LETTER OMEGA # Omega */
{ 0x03B1, 0x61 }, /* GREEK SMALL LETTER ALPHA # alpha */
{ 0x03B2, 0x62 }, /* GREEK SMALL LETTER BETA # beta */
{ 0x03B3, 0x67 }, /* GREEK SMALL LETTER GAMMA # gamma */
{ 0x03B4, 0x64 }, /* GREEK SMALL LETTER DELTA # delta */
{ 0x03B5, 0x65 }, /* GREEK SMALL LETTER EPSILON # epsilon */
{ 0x03B6, 0x7A }, /* GREEK SMALL LETTER ZETA # zeta */
{ 0x03B7, 0x68 }, /* GREEK SMALL LETTER ETA # eta */
{ 0x03B8, 0x71 }, /* GREEK SMALL LETTER THETA # theta */
{ 0x03B9, 0x69 }, /* GREEK SMALL LETTER IOTA # iota */
{ 0x03BA, 0x6B }, /* GREEK SMALL LETTER KAPPA # kappa */
{ 0x03BB, 0x6C }, /* GREEK SMALL LETTER LAMDA # lambda */
{ 0x03BC, 0x6D }, /* GREEK SMALL LETTER MU # mu */
{ 0x03BD, 0x6E }, /* GREEK SMALL LETTER NU # nu */
{ 0x03BE, 0x78 }, /* GREEK SMALL LETTER XI # xi */
{ 0x03BF, 0x6F }, /* GREEK SMALL LETTER OMICRON # omicron */
{ 0x03C0, 0x70 }, /* GREEK SMALL LETTER PI # pi */
{ 0x03C1, 0x72 }, /* GREEK SMALL LETTER RHO # rho */
{ 0x03C2, 0x56 }, /* GREEK SMALL LETTER FINAL SIGMA # sigma1 */
{ 0x03C3, 0x73 }, /* GREEK SMALL LETTER SIGMA # sigma */
{ 0x03C4, 0x74 }, /* GREEK SMALL LETTER TAU # tau */
{ 0x03C5, 0x75 }, /* GREEK SMALL LETTER UPSILON # upsilon */
{ 0x03C6, 0x66 }, /* GREEK SMALL LETTER PHI # phi */
{ 0x03C7, 0x63 }, /* GREEK SMALL LETTER CHI # chi */
{ 0x03C8, 0x79 }, /* GREEK SMALL LETTER PSI # psi */
{ 0x03C9, 0x77 }, /* GREEK SMALL LETTER OMEGA # omega */
{ 0x03D1, 0x4A }, /* GREEK THETA SYMBOL # theta1 */
{ 0x03D2, 0xA1 }, /* GREEK UPSILON WITH HOOK SYMBOL # Upsilon1 */
{ 0x03D5, 0x6A }, /* GREEK PHI SYMBOL # phi1 */
{ 0x03D6, 0x76 }, /* GREEK PI SYMBOL # omega1 */
{ 0x2022, 0xB7 }, /* BULLET # bullet */
{ 0x2026, 0xBC }, /* HORIZONTAL ELLIPSIS # ellipsis */
{ 0x2032, 0xA2 }, /* PRIME # minute */
{ 0x2033, 0xB2 }, /* DOUBLE PRIME # second */
{ 0x2044, 0xA4 }, /* FRACTION SLASH # fraction */
{ 0x20AC, 0xA0 }, /* EURO SIGN # Euro */
{ 0x2111, 0xC1 }, /* BLACK-LETTER CAPITAL I # Ifraktur */
{ 0x2118, 0xC3 }, /* SCRIPT CAPITAL P # weierstrass */
{ 0x211C, 0xC2 }, /* BLACK-LETTER CAPITAL R # Rfraktur */
{ 0x2126, 0x57 }, /* OHM SIGN # Omega */
{ 0x2135, 0xC0 }, /* ALEF SYMBOL # aleph */
{ 0x2190, 0xAC }, /* LEFTWARDS ARROW # arrowleft */
{ 0x2191, 0xAD }, /* UPWARDS ARROW # arrowup */
{ 0x2192, 0xAE }, /* RIGHTWARDS ARROW # arrowright */
{ 0x2193, 0xAF }, /* DOWNWARDS ARROW # arrowdown */
{ 0x2194, 0xAB }, /* LEFT RIGHT ARROW # arrowboth */
{ 0x21B5, 0xBF }, /* DOWNWARDS ARROW WITH CORNER LEFTWARDS # carriagereturn */
{ 0x21D0, 0xDC }, /* LEFTWARDS DOUBLE ARROW # arrowdblleft */
{ 0x21D1, 0xDD }, /* UPWARDS DOUBLE ARROW # arrowdblup */
{ 0x21D2, 0xDE }, /* RIGHTWARDS DOUBLE ARROW # arrowdblright */
{ 0x21D3, 0xDF }, /* DOWNWARDS DOUBLE ARROW # arrowdbldown */
{ 0x21D4, 0xDB }, /* LEFT RIGHT DOUBLE ARROW # arrowdblboth */
{ 0x2200, 0x22 }, /* FOR ALL # universal */
{ 0x2202, 0xB6 }, /* PARTIAL DIFFERENTIAL # partialdiff */
{ 0x2203, 0x24 }, /* THERE EXISTS # existential */
{ 0x2205, 0xC6 }, /* EMPTY SET # emptyset */
{ 0x2206, 0x44 }, /* INCREMENT # Delta */
{ 0x2207, 0xD1 }, /* NABLA # gradient */
{ 0x2208, 0xCE }, /* ELEMENT OF # element */
{ 0x2209, 0xCF }, /* NOT AN ELEMENT OF # notelement */
{ 0x220B, 0x27 }, /* CONTAINS AS MEMBER # suchthat */
{ 0x220F, 0xD5 }, /* N-ARY PRODUCT # product */
{ 0x2211, 0xE5 }, /* N-ARY SUMMATION # summation */
{ 0x2212, 0x2D }, /* MINUS SIGN # minus */
{ 0x2215, 0xA4 }, /* DIVISION SLASH # fraction */
{ 0x2217, 0x2A }, /* ASTERISK OPERATOR # asteriskmath */
{ 0x221A, 0xD6 }, /* SQUARE ROOT # radical */
{ 0x221D, 0xB5 }, /* PROPORTIONAL TO # proportional */
{ 0x221E, 0xA5 }, /* INFINITY # infinity */
{ 0x2220, 0xD0 }, /* ANGLE # angle */
{ 0x2227, 0xD9 }, /* LOGICAL AND # logicaland */
{ 0x2228, 0xDA }, /* LOGICAL OR # logicalor */
{ 0x2229, 0xC7 }, /* INTERSECTION # intersection */
{ 0x222A, 0xC8 }, /* UNION # union */
{ 0x222B, 0xF2 }, /* INTEGRAL # integral */
{ 0x2234, 0x5C }, /* THEREFORE # therefore */
{ 0x223C, 0x7E }, /* TILDE OPERATOR # similar */
{ 0x2245, 0x40 }, /* APPROXIMATELY EQUAL TO # congruent */
{ 0x2248, 0xBB }, /* ALMOST EQUAL TO # approxequal */
{ 0x2260, 0xB9 }, /* NOT EQUAL TO # notequal */
{ 0x2261, 0xBA }, /* IDENTICAL TO # equivalence */
{ 0x2264, 0xA3 }, /* LESS-THAN OR EQUAL TO # lessequal */
{ 0x2265, 0xB3 }, /* GREATER-THAN OR EQUAL TO # greaterequal */
{ 0x2282, 0xCC }, /* SUBSET OF # propersubset */
{ 0x2283, 0xC9 }, /* SUPERSET OF # propersuperset */
{ 0x2284, 0xCB }, /* NOT A SUBSET OF # notsubset */
{ 0x2286, 0xCD }, /* SUBSET OF OR EQUAL TO # reflexsubset */
{ 0x2287, 0xCA }, /* SUPERSET OF OR EQUAL TO # reflexsuperset */
{ 0x2295, 0xC5 }, /* CIRCLED PLUS # circleplus */
{ 0x2297, 0xC4 }, /* CIRCLED TIMES # circlemultiply */
{ 0x22A5, 0x5E }, /* UP TACK # perpendicular */
{ 0x22C5, 0xD7 }, /* DOT OPERATOR # dotmath */
{ 0x2320, 0xF3 }, /* TOP HALF INTEGRAL # integraltp */
{ 0x2321, 0xF5 }, /* BOTTOM HALF INTEGRAL # integralbt */
{ 0x2329, 0xE1 }, /* LEFT-POINTING ANGLE BRACKET # angleleft */
{ 0x232A, 0xF1 }, /* RIGHT-POINTING ANGLE BRACKET # angleright */
{ 0x25CA, 0xE0 }, /* LOZENGE # lozenge */
{ 0x2660, 0xAA }, /* BLACK SPADE SUIT # spade */
{ 0x2663, 0xA7 }, /* BLACK CLUB SUIT # club */
{ 0x2665, 0xA9 }, /* BLACK HEART SUIT # heart */
{ 0x2666, 0xA8 }, /* BLACK DIAMOND SUIT # diamond */
{ 0xF6D9, 0xD3 }, /* COPYRIGHT SIGN SERIF # copyrightserif (CUS) */
{ 0xF6DA, 0xD2 }, /* REGISTERED SIGN SERIF # registerserif (CUS) */
{ 0xF6DB, 0xD4 }, /* TRADE MARK SIGN SERIF # trademarkserif (CUS) */
{ 0xF8E5, 0x60 }, /* RADICAL EXTENDER # radicalex (CUS) */
{ 0xF8E6, 0xBD }, /* VERTICAL ARROW EXTENDER # arrowvertex (CUS) */
{ 0xF8E7, 0xBE }, /* HORIZONTAL ARROW EXTENDER # arrowhorizex (CUS) */
{ 0xF8E8, 0xE2 }, /* REGISTERED SIGN SANS SERIF # registersans (CUS) */
{ 0xF8E9, 0xE3 }, /* COPYRIGHT SIGN SANS SERIF # copyrightsans (CUS) */
{ 0xF8EA, 0xE4 }, /* TRADE MARK SIGN SANS SERIF # trademarksans (CUS) */
{ 0xF8EB, 0xE6 }, /* LEFT PAREN TOP # parenlefttp (CUS) */
{ 0xF8EC, 0xE7 }, /* LEFT PAREN EXTENDER # parenleftex (CUS) */
{ 0xF8ED, 0xE8 }, /* LEFT PAREN BOTTOM # parenleftbt (CUS) */
{ 0xF8EE, 0xE9 }, /* LEFT SQUARE BRACKET TOP # bracketlefttp (CUS) */
{ 0xF8EF, 0xEA }, /* LEFT SQUARE BRACKET EXTENDER # bracketleftex (CUS) */
{ 0xF8F0, 0xEB }, /* LEFT SQUARE BRACKET BOTTOM # bracketleftbt (CUS) */
{ 0xF8F1, 0xEC }, /* LEFT CURLY BRACKET TOP # bracelefttp (CUS) */
{ 0xF8F2, 0xED }, /* LEFT CURLY BRACKET MID # braceleftmid (CUS) */
{ 0xF8F3, 0xEE }, /* LEFT CURLY BRACKET BOTTOM # braceleftbt (CUS) */
{ 0xF8F4, 0xEF }, /* CURLY BRACKET EXTENDER # braceex (CUS) */
{ 0xF8F5, 0xF4 }, /* INTEGRAL EXTENDER # integralex (CUS) */
{ 0xF8F6, 0xF6 }, /* RIGHT PAREN TOP # parenrighttp (CUS) */
{ 0xF8F7, 0xF7 }, /* RIGHT PAREN EXTENDER # parenrightex (CUS) */
{ 0xF8F8, 0xF8 }, /* RIGHT PAREN BOTTOM # parenrightbt (CUS) */
{ 0xF8F9, 0xF9 }, /* RIGHT SQUARE BRACKET TOP # bracketrighttp (CUS) */
{ 0xF8FA, 0xFA }, /* RIGHT SQUARE BRACKET EXTENDER # bracketrightex (CUS) */
{ 0xF8FB, 0xFB }, /* RIGHT SQUARE BRACKET BOTTOM # bracketrightbt (CUS) */
{ 0xF8FC, 0xFC }, /* RIGHT CURLY BRACKET TOP # bracerighttp (CUS) */
{ 0xF8FD, 0xFD }, /* RIGHT CURLY BRACKET MID # bracerightmid (CUS) */
{ 0xF8FE, 0xFE }, /* RIGHT CURLY BRACKET BOTTOM # bracerightbt (CUS) */
};
static const FcCharMap AdobeSymbol = {
AdobeSymbolEnt,
sizeof (AdobeSymbolEnt) / sizeof (AdobeSymbolEnt[0]),
};
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
static const FcFontDecode fcFontDecoders[] = {
{ ft_encoding_unicode, 0, (1 << 21) - 1 },
{ ft_encoding_symbol, &AdobeSymbol, (1 << 16) - 1 },
};
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_DECODE (int) (sizeof (fcFontDecoders) / sizeof (fcFontDecoders[0]))
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
Don't force bitmap font enable in default configuration; allows users to override this in ~/.fonts.conf Updated translations Destroy font configuration on exit to help valgrind Use own transcoding routines in preference to iconv which appears to have leaks in some translators. Call iconv_close after using iconv (oops). Prefer unicode encoding of Euro char as some fonts mis-encode Euro in other ones. Must fetch bitmap glyphs to get width values to check for monospace/dual-width fonts.
2005-04-21 21:03:53 +02:00
static const FcChar32 prefer_unicode[] = {
0x20ac, /* EURO SIGN */
};
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
#define NUM_PREFER_UNICODE (int) (sizeof (prefer_unicode) / sizeof (prefer_unicode[0]))
Don't force bitmap font enable in default configuration; allows users to override this in ~/.fonts.conf Updated translations Destroy font configuration on exit to help valgrind Use own transcoding routines in preference to iconv which appears to have leaks in some translators. Call iconv_close after using iconv (oops). Prefer unicode encoding of Euro char as some fonts mis-encode Euro in other ones. Must fetch bitmap glyphs to get width values to check for monospace/dual-width fonts.
2005-04-21 21:03:53 +02:00
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
FcChar32
FcFreeTypeUcs4ToPrivate (FcChar32 ucs4, const FcCharMap *map)
{
int low, high, mid;
FcChar16 bmp;
low = 0;
high = map->nent - 1;
if (ucs4 < map->ent[low].bmp || map->ent[high].bmp < ucs4)
return ~0;
while (low <= high)
{
mid = (high + low) >> 1;
bmp = map->ent[mid].bmp;
if (ucs4 == bmp)
return (FT_ULong) map->ent[mid].encode;
if (ucs4 < bmp)
high = mid - 1;
else
low = mid + 1;
}
return ~0;
}
FcChar32
FcFreeTypePrivateToUcs4 (FcChar32 private, const FcCharMap *map)
{
int i;
for (i = 0; i < map->nent; i++)
if (map->ent[i].encode == private)
return (FcChar32) map->ent[i].bmp;
return ~0;
}
const FcCharMap *
FcFreeTypeGetPrivateMap (FT_Encoding encoding)
{
int i;
for (i = 0; i < NUM_DECODE; i++)
if (fcFontDecoders[i].encoding == encoding)
return fcFontDecoders[i].map;
return 0;
}
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
#include "../fc-glyphname/fcglyphname.h"
static FcChar32
FcHashGlyphName (const FcChar8 *name)
{
FcChar32 h = 0;
FcChar8 c;
while ((c = *name++))
{
h = ((h << 1) | (h >> 31)) ^ c;
}
return h;
}
Add autoconf checks for FT_Has_PS_Glyph_Names
2003-06-26 10:19:11 +02:00
#if HAVE_FT_HAS_PS_GLYPH_NAMES
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
/*
* Use Type1 glyph names for fonts which have reliable names
* and which export an Adobe Custom mapping
*/
static FcBool
FcFreeTypeUseNames (FT_Face face)
{
FT_Int map;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
if (!FT_Has_PS_Glyph_Names (face))
return FcFalse;
for (map = 0; map < face->num_charmaps; map++)
Add filename-based accept/reject to ammend available fonts. change FT_ENCODING_ADOBE_CUSTOM to ft_encoding_adobe_custom for older FreeType releases.
2003-05-07 18:13:24 +02:00
if (face->charmaps[map]->encoding == ft_encoding_adobe_custom)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
return FcTrue;
return FcFalse;
}
Add some more consts, fixing more GCC4 warnings.
2005-10-22 17:12:05 +02:00
static const FcChar8 *
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
FcUcs4ToGlyphName (FcChar32 ucs4)
{
int i = (int) (ucs4 % FC_GLYPHNAME_HASH);
int r = 0;
Eliminate relocations for glyph name table. Glyph names (now used only for dingbats) were using many relocations, causing startup latency plus per-process memory usage. Replace pointers with table indices, shrinking table size and elimninating relocations.
2007-10-18 18:43:22 +02:00
FcGlyphId gn;
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
[fc-glyphname] Rename internal arrays to prefix with _fc_ Although they were static, I was still surprised that gdb was seeing our variable "glyphs". Not helpful.
2009-11-18 15:26:24 +01:00
while ((gn = _fc_ucs_to_name[i]) != -1)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
[fc-glyphname] Rename internal arrays to prefix with _fc_ Although they were static, I was still surprised that gdb was seeing our variable "glyphs". Not helpful.
2009-11-18 15:26:24 +01:00
if (_fc_glyph_names[gn].ucs == ucs4)
return _fc_glyph_names[gn].name;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
if (!r)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
r = (int) (ucs4 % FC_GLYPHNAME_REHASH);
if (!r)
r = 1;
}
i += r;
if (i >= FC_GLYPHNAME_HASH)
i -= FC_GLYPHNAME_HASH;
}
return 0;
}
static FcChar32
FcGlyphNameToUcs4 (FcChar8 *name)
{
FcChar32 h = FcHashGlyphName (name);
int i = (int) (h % FC_GLYPHNAME_HASH);
int r = 0;
Eliminate relocations for glyph name table. Glyph names (now used only for dingbats) were using many relocations, causing startup latency plus per-process memory usage. Replace pointers with table indices, shrinking table size and elimninating relocations.
2007-10-18 18:43:22 +02:00
FcGlyphId gn;
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
[fc-glyphname] Rename internal arrays to prefix with _fc_ Although they were static, I was still surprised that gdb was seeing our variable "glyphs". Not helpful.
2009-11-18 15:26:24 +01:00
while ((gn = _fc_name_to_ucs[i]) != -1)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
[fc-glyphname] Rename internal arrays to prefix with _fc_ Although they were static, I was still surprised that gdb was seeing our variable "glyphs". Not helpful.
2009-11-18 15:26:24 +01:00
if (!strcmp ((char *) name, (char *) _fc_glyph_names[gn].name))
return _fc_glyph_names[gn].ucs;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
if (!r)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
r = (int) (h % FC_GLYPHNAME_REHASH);
if (!r)
r = 1;
}
i += r;
if (i >= FC_GLYPHNAME_HASH)
i -= FC_GLYPHNAME_HASH;
}
return 0xffff;
}
Work around FreeType bug when glyph name buffer is too small. Recent versions of FreeType do not correctly deal with glyph name buffers that are too small; work around this by declaring a buffer that can hold any PS name (127 bytes).
2007-03-12 18:32:23 +01:00
/*
* Work around a bug in some FreeType versions which fail
* to correctly bounds check glyph name buffers and overwrite
* the stack. As Postscript names have a limit of 127 characters,
* this should be sufficient.
*/
#if FC_GLYPHNAME_MAXLEN < 127
# define FC_GLYPHNAME_BUFLEN 127
#else
# define FC_GLYPHNAME_BUFLEN FC_GLYPHNAME_MAXLEN
#endif
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
/*
* Search through a font for a glyph by name. This is
* currently a linear search as there doesn't appear to be
* any defined order within the font
*/
static FT_UInt
Add some more consts, fixing more GCC4 warnings.
2005-10-22 17:12:05 +02:00
FcFreeTypeGlyphNameIndex (FT_Face face, const FcChar8 *name)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
FT_UInt gindex;
Work around FreeType bug when glyph name buffer is too small. Recent versions of FreeType do not correctly deal with glyph name buffers that are too small; work around this by declaring a buffer that can hold any PS name (127 bytes).
2007-03-12 18:32:23 +01:00
FcChar8 name_buf[FC_GLYPHNAME_BUFLEN + 2];
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
for (gindex = 0; gindex < (FT_UInt) face->num_glyphs; gindex++)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
Work around FreeType bug when glyph name buffer is too small. Recent versions of FreeType do not correctly deal with glyph name buffers that are too small; work around this by declaring a buffer that can hold any PS name (127 bytes).
2007-03-12 18:32:23 +01:00
if (FT_Get_Glyph_Name (face, gindex, name_buf, FC_GLYPHNAME_BUFLEN+1) == 0)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
if (!strcmp ((char *) name, (char *) name_buf))
return gindex;
}
return 0;
}
Add autoconf checks for FT_Has_PS_Glyph_Names
2003-06-26 10:19:11 +02:00
#endif
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
/*
* Map a UCS4 glyph to a glyph index. Use all available encoding
* tables to try and find one that works. This information is expected
* to be cached by higher levels, so performance isn't critical
*/
FT_UInt
FcFreeTypeCharIndex (FT_Face face, FcChar32 ucs4)
{
int initial, offset, decode;
FT_UInt glyphindex;
FcChar32 charcode;
Don't force bitmap font enable in default configuration; allows users to override this in ~/.fonts.conf Updated translations Destroy font configuration on exit to help valgrind Use own transcoding routines in preference to iconv which appears to have leaks in some translators. Call iconv_close after using iconv (oops). Prefer unicode encoding of Euro char as some fonts mis-encode Euro in other ones. Must fetch bitmap glyphs to get width values to check for monospace/dual-width fonts.
2005-04-21 21:03:53 +02:00
int p;
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
initial = 0;
Pass the buck; make fontconfig not crash on pango badness.
2006-02-21 16:56:41 +01:00
if (!face)
return 0;
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
/*
* Find the current encoding
*/
if (face->charmap)
{
for (; initial < NUM_DECODE; initial++)
if (fcFontDecoders[initial].encoding == face->charmap->encoding)
break;
if (initial == NUM_DECODE)
initial = 0;
}
Don't force bitmap font enable in default configuration; allows users to override this in ~/.fonts.conf Updated translations Destroy font configuration on exit to help valgrind Use own transcoding routines in preference to iconv which appears to have leaks in some translators. Call iconv_close after using iconv (oops). Prefer unicode encoding of Euro char as some fonts mis-encode Euro in other ones. Must fetch bitmap glyphs to get width values to check for monospace/dual-width fonts.
2005-04-21 21:03:53 +02:00
for (p = 0; p < NUM_PREFER_UNICODE; p++)
if (ucs4 == prefer_unicode[p])
{
initial = 0;
break;
}
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
/*
* Check each encoding for the glyph, starting with the current one
*/
for (offset = 0; offset < NUM_DECODE; offset++)
{
decode = (initial + offset) % NUM_DECODE;
if (!face->charmap || face->charmap->encoding != fcFontDecoders[decode].encoding)
if (FT_Select_Charmap (face, fcFontDecoders[decode].encoding) != 0)
continue;
if (fcFontDecoders[decode].map)
{
charcode = FcFreeTypeUcs4ToPrivate (ucs4, fcFontDecoders[decode].map);
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
if (charcode == ~0U)
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
continue;
}
else
charcode = ucs4;
glyphindex = FT_Get_Char_Index (face, (FT_ULong) charcode);
if (glyphindex)
return glyphindex;
}
Add autoconf checks for FT_Has_PS_Glyph_Names
2003-06-26 10:19:11 +02:00
#if HAVE_FT_HAS_PS_GLYPH_NAMES
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
/*
* Check postscript name table if present
*/
if (FcFreeTypeUseNames (face))
{
Add some more consts, fixing more GCC4 warnings.
2005-10-22 17:12:05 +02:00
const FcChar8 *name = FcUcs4ToGlyphName (ucs4);
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
if (name)
{
glyphindex = FcFreeTypeGlyphNameIndex (face, name);
if (glyphindex)
return glyphindex;
}
}
Add autoconf checks for FT_Has_PS_Glyph_Names
2003-06-26 10:19:11 +02:00
#endif
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
return 0;
}
static FcBool
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
FcFreeTypeCheckGlyph (FT_Face face, FcChar32 ucs4,
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
FT_UInt glyph, FcBlanks *blanks,
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
FT_Pos *advance,
FcBool using_strike)
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
{
Use FcIsWidth to share code Set FT_LOAD_IGNORE_GLOBAL_ADVANCE_WIDTH when scanning fonts to avoid misclassifying some Han fonts as monospaced.
2003-05-12 22:48:59 +02:00
FT_Int load_flags = FT_LOAD_IGNORE_GLOBAL_ADVANCE_WIDTH | FT_LOAD_NO_SCALE | FT_LOAD_NO_HINTING;
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
FT_GlyphSlot slot;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
if (using_strike)
load_flags &= ~FT_LOAD_NO_SCALE;
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
/*
* When using scalable fonts, only report those glyphs
* which can be scaled; otherwise those fonts will
* only be available at some sizes, and never when
* transformed. Avoid this by simply reporting bitmap-only
* glyphs as missing
*/
if (face->face_flags & FT_FACE_FLAG_SCALABLE)
load_flags |= FT_LOAD_NO_BITMAP;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
if (FT_Load_Glyph (face, glyph, load_flags))
return FcFalse;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
slot = face->glyph;
if (!glyph)
return FcFalse;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
*advance = slot->metrics.horiAdvance;
Fix compiler warnings
2012-12-30 04:11:09 +01:00
switch ((int) slot->format) {
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
case ft_glyph_format_bitmap:
/*
* Bitmaps are assumed to be reasonable; if
* this proves to be a rash assumption, this
* code can be easily modified
*/
return FcTrue;
case ft_glyph_format_outline:
/*
* Glyphs with contours are always OK
*/
if (slot->outline.n_contours != 0)
return FcTrue;
/*
* Glyphs with no contours are only OK if
* they're members of the Blanks set specified
* in the configuration. If blanks isn't set,
* then allow any glyph to be blank
*/
if (!blanks || FcBlanksIsMember (blanks, ucs4))
return FcTrue;
/* fall through ... */
default:
break;
}
return FcFalse;
}
Replace MIN/MAX/ABS macros which happen to have come from FreeType with fontconfig-specific ones (FC_*)
2004-05-06 04:28:37 +02:00
#define APPROXIMATELY_EQUAL(x,y) (FC_ABS ((x) - (y)) <= FC_MAX (FC_ABS (x), FC_ABS (y)) / 33)
Add new spacing value FC_DUAL (dual-width, as some CJK fonts). (bug #111) When checking for monospace and dual-width fonts, allow roughly a 3% variance in the advances.
2003-09-06 21:40:41 +02:00
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
static FcCharSet *
FcFreeTypeCharSetAndSpacingForSize (FT_Face face, FcBlanks *blanks, int *spacing, FT_Int strike_index)
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
{
Eliminate unused vars reported by Intel's compiler. reviewed by: plam
2006-02-21 16:40:18 +01:00
FcChar32 page, off, ucs4;
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
#ifdef CHECK
FcChar32 font_max = 0;
#endif
FcCharSet *fcs;
FcCharLeaf *leaf;
const FcCharMap *map;
int o;
int i;
FT_UInt glyph;
Add new spacing value FC_DUAL (dual-width, as some CJK fonts). (bug #111) When checking for monospace and dual-width fonts, allow roughly a 3% variance in the advances.
2003-09-06 21:40:41 +02:00
FT_Pos advance, advance_one = 0, advance_two = 0;
FcBool has_advance = FcFalse, fixed_advance = FcTrue, dual_advance = FcFalse;
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
FcBool using_strike = FcFalse;
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
fcs = FcCharSetCreate ();
if (!fcs)
goto bail0;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Fix compile with old FreeType that doesn't have FT_Select_Size() (bug #17498)
2008-12-28 10:48:54 +01:00
#if HAVE_FT_SELECT_SIZE
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
if (strike_index >= 0) {
if (FT_Select_Size (face, strike_index) != FT_Err_Ok)
goto bail1;
using_strike = FcTrue;
}
Fix compile with old FreeType that doesn't have FT_Select_Size() (bug #17498)
2008-12-28 10:48:54 +01:00
#endif
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
#ifdef CHECK
printf ("Family %s style %s\n", face->family_name, face->style_name);
#endif
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
for (o = 0; o < NUM_DECODE; o++)
{
if (FT_Select_Charmap (face, fcFontDecoders[o].encoding) != 0)
continue;
map = fcFontDecoders[o].map;
if (map)
{
/*
* Non-Unicode tables are easy; there's a list of all possible
* characters
*/
for (i = 0; i < map->nent; i++)
{
ucs4 = map->ent[i].bmp;
glyph = FT_Get_Char_Index (face, map->ent[i].encode);
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
if (glyph &&
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
FcFreeTypeCheckGlyph (face, ucs4, glyph, blanks, &advance, using_strike))
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
{
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
/*
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
* ignore glyphs with zero advance. They’re
* combining characters, and while their behaviour
* isn’t well defined for monospaced applications in
* Unicode, there are many fonts which include
* zero-width combining characters in otherwise
* monospaced fonts.
*/
if (advance)
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
{
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
if (!has_advance)
{
has_advance = FcTrue;
advance_one = advance;
}
else if (!APPROXIMATELY_EQUAL (advance, advance_one))
{
if (fixed_advance)
{
dual_advance = FcTrue;
fixed_advance = FcFalse;
advance_two = advance;
}
else if (!APPROXIMATELY_EQUAL (advance, advance_two))
dual_advance = FcFalse;
}
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
}
Add new spacing value FC_DUAL (dual-width, as some CJK fonts). (bug #111) When checking for monospace and dual-width fonts, allow roughly a 3% variance in the advances.
2003-09-06 21:40:41 +02:00
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
leaf = FcCharSetFindLeafCreate (fcs, ucs4);
if (!leaf)
goto bail1;
leaf->map[(ucs4 & 0xff) >> 5] |= (1 << (ucs4 & 0x1f));
#ifdef CHECK
if (ucs4 > font_max)
font_max = ucs4;
#endif
}
}
}
else
{
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
page = ~0;
leaf = NULL;
ucs4 = FT_Get_First_Char (face, &glyph);
while (glyph != 0)
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
{
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
if (FcFreeTypeCheckGlyph (face, ucs4, glyph, blanks, &advance, using_strike))
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
{
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
if (advance)
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
{
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
if (!has_advance)
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
{
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
has_advance = FcTrue;
advance_one = advance;
}
else if (!APPROXIMATELY_EQUAL (advance, advance_one))
{
if (fixed_advance)
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
{
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
dual_advance = FcTrue;
fixed_advance = FcFalse;
advance_two = advance;
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
}
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
else if (!APPROXIMATELY_EQUAL (advance, advance_two))
dual_advance = FcFalse;
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
}
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
}
Add new spacing value FC_DUAL (dual-width, as some CJK fonts). (bug #111) When checking for monospace and dual-width fonts, allow roughly a 3% variance in the advances.
2003-09-06 21:40:41 +02:00
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
if ((ucs4 >> 8) != page)
{
page = (ucs4 >> 8);
leaf = FcCharSetFindLeafCreate (fcs, ucs4);
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
if (!leaf)
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
goto bail1;
}
off = ucs4 & 0xff;
leaf->map[off >> 5] |= (1 << (off & 0x1f));
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
#ifdef CHECK
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
if (ucs4 > font_max)
font_max = ucs4;
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
#endif
}
Replace character discovery loop with simpler, faster version. The existing loop for discovering which characters map to glyphs is ugly and inefficient. The replacement is functionally identical, but far cleaner and faster.
2006-09-07 23:29:35 +02:00
ucs4 = FT_Get_Next_Char (face, ucs4, &glyph);
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
}
#ifdef CHECK
for (ucs4 = 0; ucs4 < 0x10000; ucs4++)
{
FcBool FT_Has, FC_Has;
FT_Has = FT_Get_Char_Index (face, ucs4) != 0;
FC_Has = FcCharSetHasChar (fcs, ucs4);
if (FT_Has != FC_Has)
{
printf ("0x%08x FT says %d FC says %d\n", ucs4, FT_Has, FC_Has);
}
}
#endif
}
}
Add autoconf checks for FT_Has_PS_Glyph_Names
2003-06-26 10:19:11 +02:00
#if HAVE_FT_HAS_PS_GLYPH_NAMES
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
/*
* Add mapping from PS glyph names if available
*/
if (FcFreeTypeUseNames (face))
{
Work around FreeType bug when glyph name buffer is too small. Recent versions of FreeType do not correctly deal with glyph name buffers that are too small; work around this by declaring a buffer that can hold any PS name (127 bytes).
2007-03-12 18:32:23 +01:00
FcChar8 name_buf[FC_GLYPHNAME_BUFLEN + 2];
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
Fix more gcc4 warnings: - Cast sizeof to int, to shut up signedness warnings in comparison. - Add consts where appropriate. reviewed by: Patrick Lam <plam@mit.edu>
2005-09-23 01:45:53 +02:00
for (glyph = 0; glyph < (FT_UInt) face->num_glyphs; glyph++)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
Work around FreeType bug when glyph name buffer is too small. Recent versions of FreeType do not correctly deal with glyph name buffers that are too small; work around this by declaring a buffer that can hold any PS name (127 bytes).
2007-03-12 18:32:23 +01:00
if (FT_Get_Glyph_Name (face, glyph, name_buf, FC_GLYPHNAME_BUFLEN+1) == 0)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
ucs4 = FcGlyphNameToUcs4 (name_buf);
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
if (ucs4 != 0xffff &&
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
FcFreeTypeCheckGlyph (face, ucs4, glyph, blanks, &advance, using_strike))
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
if (advance)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
Merge with HEAD and finish the GCC 4 cleanups (no more warnings!)
2005-09-11 04:16:09 +02:00
if (!has_advance)
{
has_advance = FcTrue;
advance_one = advance;
}
else if (!APPROXIMATELY_EQUAL (advance, advance_one))
{
if (fixed_advance)
{
dual_advance = FcTrue;
fixed_advance = FcFalse;
advance_two = advance;
}
else if (!APPROXIMATELY_EQUAL (advance, advance_two))
dual_advance = FcFalse;
}
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
}
leaf = FcCharSetFindLeafCreate (fcs, ucs4);
if (!leaf)
goto bail1;
leaf->map[(ucs4 & 0xff) >> 5] |= (1 << (ucs4 & 0x1f));
#ifdef CHECK
if (ucs4 > font_max)
font_max = ucs4;
#endif
}
}
}
}
Add autoconf checks for FT_Has_PS_Glyph_Names
2003-06-26 10:19:11 +02:00
#endif
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
#ifdef CHECK
printf ("%d glyphs %d encoded\n", (int) face->num_glyphs, FcCharSetCount (fcs));
for (ucs4 = 0; ucs4 <= font_max; ucs4++)
{
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
FcBool has_char = (glyph = FcFreeTypeCharIndex (face, ucs4)) != 0;
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
FcBool has_bit = FcCharSetHasChar (fcs, ucs4);
if (has_char && !has_bit)
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
{
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
if (!FcFreeTypeCheckGlyph (face, ucs4, glyph, blanks, &advance, using_strike))
Handle Adobe glyph names for fonts which include ADOBE_CUSTOM encodings
2003-05-05 00:53:49 +02:00
printf ("Bitmap missing broken char 0x%x\n", ucs4);
else
printf ("Bitmap missing char 0x%x\n", ucs4);
}
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
else if (!has_char && has_bit)
printf ("Bitmap extra char 0x%x\n", ucs4);
}
#endif
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
if (fixed_advance)
*spacing = FC_MONO;
Replace MIN/MAX/ABS macros which happen to have come from FreeType with fontconfig-specific ones (FC_*)
2004-05-06 04:28:37 +02:00
else if (dual_advance && APPROXIMATELY_EQUAL (2 * FC_MIN (advance_one, advance_two), FC_MAX (advance_one, advance_two)))
Add new spacing value FC_DUAL (dual-width, as some CJK fonts). (bug #111) When checking for monospace and dual-width fonts, allow roughly a 3% variance in the advances.
2003-09-06 21:40:41 +02:00
*spacing = FC_DUAL;
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
else
*spacing = FC_PROPORTIONAL;
Build fclang.h before building library This required compiling the charset funcs into fc-lang, which was done by refactoring code in fccharset.c and fcfreetype.c a bit Updated ethiopic orthographies Remove imake support Install empty local.conf file if none is present
2003-02-06 18:46:06 +01:00
return fcs;
bail1:
FcCharSetDestroy (fcs);
bail0:
return 0;
}
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
FcCharSet *
FcFreeTypeCharSetAndSpacing (FT_Face face, FcBlanks *blanks, int *spacing)
{
FcCharSet *cs;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
/*
* Check for bitmap-only ttf fonts that are missing the glyf table.
* In that case, pick a size and look for glyphs in that size instead
*/
Bug 23336 - unable to display bitmap-only (SFNT) TrueType or OpenType Force to find out a size for bitmap-only ttf to avoid the blank glyphs in the font. Patch from Bug Fly
2012-03-07 09:32:14 +01:00
if (!(face->face_flags & FT_FACE_FLAG_SCALABLE) &&
face->num_fixed_sizes > 0 &&
FT_Get_Sfnt_Table (face, ft_sfnt_head))
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
{
Bug 23336 - unable to display bitmap-only (SFNT) TrueType or OpenType Force to find out a size for bitmap-only ttf to avoid the blank glyphs in the font. Patch from Bug Fly
2012-03-07 09:32:14 +01:00
FT_Int strike_index = 0;
int i;
/* Select the face closest to 16 pixels tall */
for (i = 1; i < face->num_fixed_sizes; i++) {
if (abs (face->available_sizes[i].height - 16) <
abs (face->available_sizes[strike_index].height - 16))
strike_index = i;
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
}
Bug 23336 - unable to display bitmap-only (SFNT) TrueType or OpenType Force to find out a size for bitmap-only ttf to avoid the blank glyphs in the font. Patch from Bug Fly
2012-03-07 09:32:14 +01:00
cs = FcFreeTypeCharSetAndSpacingForSize (face, blanks, spacing, strike_index);
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
}
Bug 23336 - unable to display bitmap-only (SFNT) TrueType or OpenType Force to find out a size for bitmap-only ttf to avoid the blank glyphs in the font. Patch from Bug Fly
2012-03-07 09:32:14 +01:00
else
cs = FcFreeTypeCharSetAndSpacingForSize (face, blanks, spacing, -1);
Work around for bitmap-only TrueType fonts that are missing the glyf table. Bitmap-only TrueType fonts without a glyf table will not load a glyph when FT_LOAD_NO_SCALE is set. Work around this by identifying TrueType fonts that have no glyphs and select a single strike to measure the glyph map with.
2008-05-04 02:19:43 +02:00
return cs;
}
Set spacing to mono if every encoded glyph is the same width
2003-03-28 18:08:35 +01:00
FcCharSet *
FcFreeTypeCharSet (FT_Face face, FcBlanks *blanks)
{
int spacing;
return FcFreeTypeCharSetAndSpacing (face, blanks, &spacing);
}
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
#define TTAG_GPOS FT_MAKE_TAG( 'G', 'P', 'O', 'S' )
#define TTAG_GSUB FT_MAKE_TAG( 'G', 'S', 'U', 'B' )
#define TTAG_SILF FT_MAKE_TAG( 'S', 'i', 'l', 'f')
Rework GSUB/GPOS script parsing to survive broken fonts. Thanks for the broken font go to Manish Singh
2005-03-06 00:34:57 +01:00
#define OTLAYOUT_HEAD "otlayout:"
#define OTLAYOUT_HEAD_LEN 9
#define OTLAYOUT_ID_LEN 4
/* space + head + id */
#define OTLAYOUT_LEN (1 + OTLAYOUT_HEAD_LEN + OTLAYOUT_ID_LEN)
Include space and remove numbers from valid script tags. This ensures that tags like 'lao ' work while rejecting those which have any digits. Eliminate a spurious debugging variable (len)
2005-03-06 00:50:55 +01:00
/*
* This is a bit generous; the registry has only lower case and space
* except for 'DFLT'.
*/
#define FcIsSpace(x) (040 == (x))
Accept digits as part of OpenType script tags They've been used since 2005.
2013-02-15 15:48:38 +01:00
#define FcIsDigit(c) (('0' <= (c) && (c) <= '9'))
#define FcIsValidScript(x) (FcIsLower(x) || FcIsUpper (x) || FcIsDigit(x) || FcIsSpace(x))
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
static void
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
addtag(FcChar8 *complex_, FT_ULong tag)
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
{
Rework GSUB/GPOS script parsing to survive broken fonts. Thanks for the broken font go to Manish Singh
2005-03-06 00:34:57 +01:00
FcChar8 tagstring[OTLAYOUT_ID_LEN + 1];
tagstring[0] = (FcChar8)(tag >> 24),
tagstring[1] = (FcChar8)(tag >> 16),
tagstring[2] = (FcChar8)(tag >> 8),
tagstring[3] = (FcChar8)(tag);
tagstring[4] = '\0';
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
Accept digits as part of OpenType script tags They've been used since 2005.
2013-02-15 15:48:38 +01:00
/* skip tags which aren't alphanumeric, under the assumption that
Rework GSUB/GPOS script parsing to survive broken fonts. Thanks for the broken font go to Manish Singh
2005-03-06 00:34:57 +01:00
* they're probably broken
*/
Include space and remove numbers from valid script tags. This ensures that tags like 'lao ' work while rejecting those which have any digits. Eliminate a spurious debugging variable (len)
2005-03-06 00:50:55 +01:00
if (!FcIsValidScript(tagstring[0]) ||
!FcIsValidScript(tagstring[1]) ||
!FcIsValidScript(tagstring[2]) ||
!FcIsValidScript(tagstring[3]))
Rework GSUB/GPOS script parsing to survive broken fonts. Thanks for the broken font go to Manish Singh
2005-03-06 00:34:57 +01:00
return;
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
if (*complex_ != '\0')
strcat ((char *) complex_, " ");
Fix compiler warnings
2012-12-30 04:11:09 +01:00
strcat ((char *) complex_, OTLAYOUT_HEAD);
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
strcat ((char *) complex_, (char *) tagstring);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
}
static int
compareulong (const void *a, const void *b)
{
const FT_ULong *ua = (const FT_ULong *) a;
const FT_ULong *ub = (const FT_ULong *) b;
return *ua - *ub;
}
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
static int
GetScriptTags(FT_Face face, FT_ULong tabletag, FT_ULong **stags)
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
{
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
FT_ULong cur_offset, new_offset, base_offset;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
FT_Stream stream = face->stream;
FT_Error error;
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
FT_UShort n, p;
int script_count;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
if (!stream)
return 0;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
if (( error = ftglue_face_goto_table( face, tabletag, stream ) ))
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
return 0;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
base_offset = ftglue_stream_pos ( stream );
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
/* skip version */
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
if ( ftglue_stream_seek ( stream, base_offset + 4L ) || ftglue_stream_frame_enter( stream, 2L ) )
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
return 0;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Remove stuff we don't use, make get_{char,short,long} functions of ftglue macros to be inlined. Code cleanups (excess prototype, old-style function definition). reviewed by: plam
2006-03-08 03:30:43 +01:00
new_offset = GET_UShort() + base_offset;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
ftglue_stream_frame_exit( stream );
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
cur_offset = ftglue_stream_pos( stream );
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
if ( ftglue_stream_seek( stream, new_offset ) != FT_Err_Ok )
return 0;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
base_offset = ftglue_stream_pos( stream );
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
if ( ftglue_stream_frame_enter( stream, 2L ) )
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
return 0;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
script_count = GET_UShort ();
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
ftglue_stream_frame_exit( stream );
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
*stags = malloc(script_count * sizeof (FT_ULong));
if (!stags)
return 0;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
p = 0;
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
for ( n = 0; n < script_count; n++ )
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
{
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
if ( ftglue_stream_frame_enter( stream, 6L ) )
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
goto Fail;
Remove stuff we don't use, make get_{char,short,long} functions of ftglue macros to be inlined. Code cleanups (excess prototype, old-style function definition). reviewed by: plam
2006-03-08 03:30:43 +01:00
(*stags)[p] = GET_ULong ();
new_offset = GET_UShort () + base_offset;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
ftglue_stream_frame_exit( stream );
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
cur_offset = ftglue_stream_pos( stream );
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Fix memory leak (Coverity defect #2089). Ignore script if subtable is missing (Coverity defect #2088). Fix possible null pointer dereference (Coverity defect #784) and memory leak (Coverity defects #785, #786). Don't copy FcCharSet if we're going to throw it away anyway. (Reported by Kenichi Handa). reviewed by: plam
2006-04-12 16:36:36 +02:00
error = ftglue_stream_seek( stream, new_offset );
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
if ( error == FT_Err_Ok )
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
p++;
Get rid of the use of freetype internal headers in fcfreetype.c, since those headers will go away with freetype 2.2. Replace with public domain ftglue code from pango. Note that the patch removes some extra error checking in FT_Get_BDF_Property() and comments out the skipping of empty pcf fonts. reviewed by: plam
2005-11-19 23:32:13 +01:00
(void)ftglue_stream_seek( stream, cur_offset );
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
}
if (!p)
goto Fail;
Get rid of C++-style comments.
2006-03-03 07:35:53 +01:00
/* sort the tag list before returning it */
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
qsort(*stags, script_count, sizeof(FT_ULong), compareulong);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
return script_count;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
Fail:
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
free(*stags);
Takashi Iwai <tiwai@suse.de> reviewed by: plam Fix double-free on error case.
2006-02-24 17:41:34 +01:00
*stags = NULL;
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
return 0;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
}
static FcChar8 *
FcFontCapabilities(FT_Face face)
{
FcBool issilgraphitefont = 0;
FT_Error err;
FT_ULong len = 0;
FT_ULong *gsubtags=NULL, *gpostags=NULL;
Rework GSUB/GPOS script parsing to survive broken fonts. Thanks for the broken font go to Manish Singh
2005-03-06 00:34:57 +01:00
FT_UShort gsub_count=0, gpos_count=0;
FT_ULong maxsize;
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
FcChar8 *complex_ = NULL;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
int indx1 = 0, indx2 = 0;
err = FT_Load_Sfnt_Table(face, TTAG_SILF, 0, 0, &len);
issilgraphitefont = ( err == FT_Err_Ok);
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
gpos_count = GetScriptTags(face, TTAG_GPOS, &gpostags);
gsub_count = GetScriptTags(face, TTAG_GSUB, &gsubtags);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
if (!issilgraphitefont && !gsub_count && !gpos_count)
goto bail;
Remove all training whitespaces
2010-04-12 18:18:50 +02:00
maxsize = (((FT_ULong) gpos_count + (FT_ULong) gsub_count) * OTLAYOUT_LEN +
Rework GSUB/GPOS script parsing to survive broken fonts. Thanks for the broken font go to Manish Singh
2005-03-06 00:34:57 +01:00
(issilgraphitefont ? 13 : 0));
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
complex_ = malloc (sizeof (FcChar8) * maxsize);
if (!complex_)
Rework GSUB/GPOS script parsing to survive broken fonts. Thanks for the broken font go to Manish Singh
2005-03-06 00:34:57 +01:00
goto bail;
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
complex_[0] = '\0';
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
if (issilgraphitefont)
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
strcpy((char *) complex_, "ttable:Silf ");
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
while ((indx1 < gsub_count) || (indx2 < gpos_count)) {
if (indx1 == gsub_count) {
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
addtag(complex_, gpostags[indx2]);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
indx2++;
} else if ((indx2 == gpos_count) || (gsubtags[indx1] < gpostags[indx2])) {
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
addtag(complex_, gsubtags[indx1]);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
indx1++;
} else if (gsubtags[indx1] == gpostags[indx2]) {
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
addtag(complex_, gsubtags[indx1]);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
indx1++;
indx2++;
} else {
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
addtag(complex_, gpostags[indx2]);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
indx2++;
}
}
if (FcDebug () & FC_DBG_SCANV)
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
printf("complex_ features in this font: %s\n", complex_);
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
bail:
Remove unused ftglue code
2009-06-28 20:14:46 +02:00
free(gsubtags);
free(gpostags);
Don't use identifier named complex
2008-12-28 15:03:29 +01:00
return complex_;
Fix typo. Add detection for font capabilities (bug #105) reviewed by: Keith Packard <keithp@keithp.com>
2004-12-04 23:06:52 +01:00
}
Eliminate .so PLT entries for local symbols. (thanks to Arjan van de Ven) Using a simple shell script that processes the public headers, two header files are constructed that map public symbols to hidden internal aliases avoiding the assocated PLT entry for referring to a public symbol. A few mistakes in the FcPrivate/FcPublic annotations were also discovered through this process
2006-09-05 11:24:01 +02:00
#define __fcfreetype__
#include "fcaliastail.h"
Include fcftaliastail.h so that the freetype funcs are exported. This header file needs to be included at the end of every file that exports any freetype symbols.
2008-01-08 01:31:06 +01:00
#include "fcftaliastail.h"
Eliminate .so PLT entries for local symbols. (thanks to Arjan van de Ven) Using a simple shell script that processes the public headers, two header files are constructed that map public symbols to hidden internal aliases avoiding the assocated PLT entry for referring to a public symbol. A few mistakes in the FcPrivate/FcPublic annotations were also discovered through this process
2006-09-05 11:24:01 +02:00
#undef __fcfreetype__