harfbuzz/src/hb-ot-tag.cc

622 lines
17 KiB
C++
Raw Normal View History

/*
2011-04-21 23:14:28 +02:00
* Copyright © 2009 Red Hat, Inc.
* Copyright © 2011 Google, Inc.
*
2010-04-22 06:11:43 +02:00
* This is part of HarfBuzz, a text shaping library.
*
* Permission is hereby granted, without written agreement and without
* license or royalty fees, to use, copy, modify, and distribute this
* software and its documentation for any purpose, provided that the
* above copyright notice and the following two paragraphs appear in
* all copies of this software.
*
* IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE TO ANY PARTY FOR
* DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES
* ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN
* IF THE COPYRIGHT HOLDER HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
* DAMAGE.
*
* THE COPYRIGHT HOLDER SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING,
* BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
* FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS
* ON AN "AS IS" BASIS, AND THE COPYRIGHT HOLDER HAS NO OBLIGATION TO
* PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
*
* Red Hat Author(s): Behdad Esfahbod
* Google Author(s): Behdad Esfahbod, Roozbeh Pournader
*/
#include "hb.hh"
#ifndef HB_NO_OT_TAG
/* hb_script_t */
static hb_tag_t
hb_ot_old_tag_from_script (hb_script_t script)
{
2012-11-02 21:38:55 +01:00
/* This seems to be accurate as of end of 2012. */
2018-10-12 01:39:07 +02:00
switch ((hb_tag_t) script)
{
case HB_SCRIPT_INVALID: return HB_OT_TAG_DEFAULT_SCRIPT;
case HB_SCRIPT_MATH: return HB_OT_TAG_MATH_SCRIPT;
/* KATAKANA and HIRAGANA both map to 'kana' */
case HB_SCRIPT_HIRAGANA: return HB_TAG('k','a','n','a');
/* Spaces at the end are preserved, unlike ISO 15924 */
case HB_SCRIPT_LAO: return HB_TAG('l','a','o',' ');
case HB_SCRIPT_YI: return HB_TAG('y','i',' ',' ');
/* Unicode-5.0 additions */
case HB_SCRIPT_NKO: return HB_TAG('n','k','o',' ');
/* Unicode-5.1 additions */
case HB_SCRIPT_VAI: return HB_TAG('v','a','i',' ');
}
/* Else, just change first char to lowercase and return */
return ((hb_tag_t) script) | 0x20000000u;
}
static hb_script_t
hb_ot_old_tag_to_script (hb_tag_t tag)
{
if (unlikely (tag == HB_OT_TAG_DEFAULT_SCRIPT))
return HB_SCRIPT_INVALID;
if (unlikely (tag == HB_OT_TAG_MATH_SCRIPT))
return HB_SCRIPT_MATH;
/* This side of the conversion is fully algorithmic. */
/* Any spaces at the end of the tag are replaced by repeating the last
* letter. Eg 'nko ' -> 'Nkoo' */
if (unlikely ((tag & 0x0000FF00u) == 0x00002000u))
tag |= (tag >> 8) & 0x0000FF00u; /* Copy second letter to third */
if (unlikely ((tag & 0x000000FFu) == 0x00000020u))
tag |= (tag >> 8) & 0x000000FFu; /* Copy third letter to fourth */
/* Change first char to uppercase and return */
return (hb_script_t) (tag & ~0x20000000u);
}
static hb_tag_t
hb_ot_new_tag_from_script (hb_script_t script)
{
switch ((hb_tag_t) script) {
case HB_SCRIPT_BENGALI: return HB_TAG('b','n','g','2');
case HB_SCRIPT_DEVANAGARI: return HB_TAG('d','e','v','2');
case HB_SCRIPT_GUJARATI: return HB_TAG('g','j','r','2');
case HB_SCRIPT_GURMUKHI: return HB_TAG('g','u','r','2');
case HB_SCRIPT_KANNADA: return HB_TAG('k','n','d','2');
case HB_SCRIPT_MALAYALAM: return HB_TAG('m','l','m','2');
case HB_SCRIPT_ORIYA: return HB_TAG('o','r','y','2');
case HB_SCRIPT_TAMIL: return HB_TAG('t','m','l','2');
case HB_SCRIPT_TELUGU: return HB_TAG('t','e','l','2');
case HB_SCRIPT_MYANMAR: return HB_TAG('m','y','m','2');
}
return HB_OT_TAG_DEFAULT_SCRIPT;
}
static hb_script_t
hb_ot_new_tag_to_script (hb_tag_t tag)
{
switch (tag) {
case HB_TAG('b','n','g','2'): return HB_SCRIPT_BENGALI;
case HB_TAG('d','e','v','2'): return HB_SCRIPT_DEVANAGARI;
case HB_TAG('g','j','r','2'): return HB_SCRIPT_GUJARATI;
case HB_TAG('g','u','r','2'): return HB_SCRIPT_GURMUKHI;
case HB_TAG('k','n','d','2'): return HB_SCRIPT_KANNADA;
case HB_TAG('m','l','m','2'): return HB_SCRIPT_MALAYALAM;
case HB_TAG('o','r','y','2'): return HB_SCRIPT_ORIYA;
case HB_TAG('t','m','l','2'): return HB_SCRIPT_TAMIL;
case HB_TAG('t','e','l','2'): return HB_SCRIPT_TELUGU;
case HB_TAG('m','y','m','2'): return HB_SCRIPT_MYANMAR;
}
return HB_SCRIPT_UNKNOWN;
}
#ifndef HB_DISABLE_DEPRECATED
void
hb_ot_tags_from_script (hb_script_t script,
hb_tag_t *script_tag_1,
hb_tag_t *script_tag_2)
{
unsigned int count = 2;
hb_tag_t tags[2];
2018-07-24 03:19:23 +02:00
hb_ot_tags_from_script_and_language (script, HB_LANGUAGE_INVALID, &count, tags, nullptr, nullptr);
*script_tag_1 = count > 0 ? tags[0] : HB_OT_TAG_DEFAULT_SCRIPT;
*script_tag_2 = count > 1 ? tags[1] : HB_OT_TAG_DEFAULT_SCRIPT;
}
#endif
/*
* Complete list at:
* https://docs.microsoft.com/en-us/typography/opentype/spec/scripttags
*
* Most of the script tags are the same as the ISO 15924 tag but lowercased.
* So we just do that, and handle the exceptional cases in a switch.
*/
static void
hb_ot_all_tags_from_script (hb_script_t script,
unsigned int *count /* IN/OUT */,
hb_tag_t *tags /* OUT */)
{
unsigned int i = 0;
hb_tag_t new_tag = hb_ot_new_tag_from_script (script);
if (unlikely (new_tag != HB_OT_TAG_DEFAULT_SCRIPT))
2018-10-11 23:15:22 +02:00
{
2019-06-11 23:55:31 +02:00
/* HB_SCRIPT_MYANMAR maps to 'mym2', but there is no 'mym3'. */
if (new_tag != HB_TAG('m','y','m','2'))
tags[i++] = new_tag | '3';
2018-10-11 23:15:22 +02:00
if (*count > i)
tags[i++] = new_tag;
}
if (*count > i)
{
hb_tag_t old_tag = hb_ot_old_tag_from_script (script);
if (old_tag != HB_OT_TAG_DEFAULT_SCRIPT)
tags[i++] = old_tag;
}
*count = i;
}
/**
* hb_ot_tag_to_script:
* @tag: a script tag
*
* Converts a script tag to an #hb_script_t.
*
* Return value: The #hb_script_t corresponding to @tag.
*
**/
2010-05-21 12:15:07 +02:00
hb_script_t
hb_ot_tag_to_script (hb_tag_t tag)
{
2018-10-11 23:15:22 +02:00
unsigned char digit = tag & 0x000000FFu;
if (unlikely (digit == '2' || digit == '3'))
return hb_ot_new_tag_to_script (tag & 0xFFFFFF32);
2010-05-21 12:15:07 +02:00
return hb_ot_old_tag_to_script (tag);
}
2010-05-21 12:15:07 +02:00
/* hb_language_t */
static inline bool
Autogenerate the BCP 47 to OpenType mappings The new script, gen-tag-table.py, generates `ot_languages` automatically from the [OpenType language system tag registry][ot] and the [IANA Language Subtag Registry][bcp47] with some manual modifications. If an OpenType tag maps to a BCP 47 macrolanguage, all the macrolanguage's individual languages are mapped to the same OpenType tag, except for individual languages with their own OpenType mappings. Deprecated BCP 47 tags are canonicalized. [ot]: https://docs.microsoft.com/en-us/typography/opentype/spec/languagetags [bcp47]: https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry Some OpenType tags correspond to multiple ISO 639 codes. The mapping from ISO 639 codes lists OpenType tags in priority order, such that more specific or more likely tags appear first. Some OpenType tags have no corresponding ISO 639 code in the registry so their mappings use BCP 47 subtags besides the language. For example, any BCP 47 tag with a fonipa variant subtag is mapped to 'IPPH', and 'IPPH' is mapped back to und-fonipa. Other OpenType tags have no corresponding ISO 639 code because it is not clear what they are for. HarfBuzz just ignores these tags. One such ignored tag is 'ZHP ' (Chinese Phonetic). It probably means zh-Latn. However, it is used in Microsoft JhengHei and Microsoft YaHei with the script tag 'hani', implying that it is not a romanization scheme after all. It would be simple enough to add this mapping to gen-tag-table.py once a definitive mapping is determined. The manual modifications are mainly either obvious mappings that the OpenType registry omits or mappings for compatibility with previous versions of HarfBuzz. Some of the old mappings were discarded, though, for homophonous language names. For example, OpenType maps 'KUI ' to kxu; previous versions of HarfBuzz also mapped it to kvd, because kvd and kxu both happen to be called "Kui". gen-tag-table.py also generates a function to convert multi-subtag tags like el-polyton and zh-HK to OpenType tags, replacing `ot_languages_zh` and the hard-coded list of special cases in `hb_ot_tags_from_language`. It also generates a function to convert OpenType tags to BCP 47, replacing the hard-coded list of special cases in `hb_ot_tag_to_language`.
2017-12-09 04:45:52 +01:00
subtag_matches (const char *lang_str,
const char *limit,
const char *subtag,
2022-05-18 01:31:18 +02:00
unsigned subtag_len)
{
if (likely ((unsigned) (limit - lang_str) < subtag_len))
return false;
do {
const char *s = strstr (lang_str, subtag);
Autogenerate the BCP 47 to OpenType mappings The new script, gen-tag-table.py, generates `ot_languages` automatically from the [OpenType language system tag registry][ot] and the [IANA Language Subtag Registry][bcp47] with some manual modifications. If an OpenType tag maps to a BCP 47 macrolanguage, all the macrolanguage's individual languages are mapped to the same OpenType tag, except for individual languages with their own OpenType mappings. Deprecated BCP 47 tags are canonicalized. [ot]: https://docs.microsoft.com/en-us/typography/opentype/spec/languagetags [bcp47]: https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry Some OpenType tags correspond to multiple ISO 639 codes. The mapping from ISO 639 codes lists OpenType tags in priority order, such that more specific or more likely tags appear first. Some OpenType tags have no corresponding ISO 639 code in the registry so their mappings use BCP 47 subtags besides the language. For example, any BCP 47 tag with a fonipa variant subtag is mapped to 'IPPH', and 'IPPH' is mapped back to und-fonipa. Other OpenType tags have no corresponding ISO 639 code because it is not clear what they are for. HarfBuzz just ignores these tags. One such ignored tag is 'ZHP ' (Chinese Phonetic). It probably means zh-Latn. However, it is used in Microsoft JhengHei and Microsoft YaHei with the script tag 'hani', implying that it is not a romanization scheme after all. It would be simple enough to add this mapping to gen-tag-table.py once a definitive mapping is determined. The manual modifications are mainly either obvious mappings that the OpenType registry omits or mappings for compatibility with previous versions of HarfBuzz. Some of the old mappings were discarded, though, for homophonous language names. For example, OpenType maps 'KUI ' to kxu; previous versions of HarfBuzz also mapped it to kvd, because kvd and kxu both happen to be called "Kui". gen-tag-table.py also generates a function to convert multi-subtag tags like el-polyton and zh-HK to OpenType tags, replacing `ot_languages_zh` and the hard-coded list of special cases in `hb_ot_tags_from_language`. It also generates a function to convert OpenType tags to BCP 47, replacing the hard-coded list of special cases in `hb_ot_tag_to_language`.
2017-12-09 04:45:52 +01:00
if (!s || s >= limit)
return false;
if (!ISALNUM (s[subtag_len]))
return true;
lang_str = s + subtag_len;
2018-10-18 17:18:42 +02:00
} while (true);
}
2022-05-17 22:57:08 +02:00
static bool
2022-05-18 01:31:18 +02:00
lang_matches (const char *lang_str,
const char *limit,
const char *spec,
unsigned spec_len)
{
2022-05-17 23:07:49 +02:00
if (likely ((unsigned) (limit - lang_str) < spec_len))
2022-05-17 22:57:08 +02:00
return false;
2022-05-17 22:57:08 +02:00
return strncmp (lang_str, spec, spec_len) == 0 &&
(lang_str[spec_len] == '\0' || lang_str[spec_len] == '-');
}
struct LangTag
{
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
hb_tag_t language;
hb_tag_t tag;
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
int cmp (hb_tag_t a) const
{
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
return a < this->language ? -1 : a > this->language ? +1 : 0;
}
int cmp (const LangTag *that) const
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
{ return cmp (that->language); }
};
Autogenerate the BCP 47 to OpenType mappings The new script, gen-tag-table.py, generates `ot_languages` automatically from the [OpenType language system tag registry][ot] and the [IANA Language Subtag Registry][bcp47] with some manual modifications. If an OpenType tag maps to a BCP 47 macrolanguage, all the macrolanguage's individual languages are mapped to the same OpenType tag, except for individual languages with their own OpenType mappings. Deprecated BCP 47 tags are canonicalized. [ot]: https://docs.microsoft.com/en-us/typography/opentype/spec/languagetags [bcp47]: https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry Some OpenType tags correspond to multiple ISO 639 codes. The mapping from ISO 639 codes lists OpenType tags in priority order, such that more specific or more likely tags appear first. Some OpenType tags have no corresponding ISO 639 code in the registry so their mappings use BCP 47 subtags besides the language. For example, any BCP 47 tag with a fonipa variant subtag is mapped to 'IPPH', and 'IPPH' is mapped back to und-fonipa. Other OpenType tags have no corresponding ISO 639 code because it is not clear what they are for. HarfBuzz just ignores these tags. One such ignored tag is 'ZHP ' (Chinese Phonetic). It probably means zh-Latn. However, it is used in Microsoft JhengHei and Microsoft YaHei with the script tag 'hani', implying that it is not a romanization scheme after all. It would be simple enough to add this mapping to gen-tag-table.py once a definitive mapping is determined. The manual modifications are mainly either obvious mappings that the OpenType registry omits or mappings for compatibility with previous versions of HarfBuzz. Some of the old mappings were discarded, though, for homophonous language names. For example, OpenType maps 'KUI ' to kxu; previous versions of HarfBuzz also mapped it to kvd, because kvd and kxu both happen to be called "Kui". gen-tag-table.py also generates a function to convert multi-subtag tags like el-polyton and zh-HK to OpenType tags, replacing `ot_languages_zh` and the hard-coded list of special cases in `hb_ot_tags_from_language`. It also generates a function to convert OpenType tags to BCP 47, replacing the hard-coded list of special cases in `hb_ot_tag_to_language`.
2017-12-09 04:45:52 +01:00
#include "hb-ot-tag-table.hh"
/* The corresponding languages IDs for the following IDs are unclear,
* overlap, or are architecturally weird. Needs more research. */
/*{"??", {HB_TAG('B','C','R',' ')}},*/ /* Bible Cree */
/*{"zh?", {HB_TAG('C','H','N',' ')}},*/ /* Chinese (seen in Microsoft fonts) */
/*{"ar-Syrc?", {HB_TAG('G','A','R',' ')}},*/ /* Garshuni */
/*{"??", {HB_TAG('N','G','R',' ')}},*/ /* Nagari */
/*{"??", {HB_TAG('Y','I','C',' ')}},*/ /* Yi Classic */
/*{"zh?", {HB_TAG('Z','H','P',' ')}},*/ /* Chinese Phonetic */
#ifndef HB_DISABLE_DEPRECATED
hb_tag_t
hb_ot_tag_from_language (hb_language_t language)
{
unsigned int count = 1;
hb_tag_t tags[1];
2018-07-24 03:19:23 +02:00
hb_ot_tags_from_script_and_language (HB_SCRIPT_UNKNOWN, language, nullptr, nullptr, &count, tags);
return count > 0 ? tags[0] : HB_OT_TAG_DEFAULT_LANGUAGE;
}
#endif
static void
hb_ot_tags_from_language (const char *lang_str,
const char *limit,
unsigned int *count,
hb_tag_t *tags)
{
const char *s;
Autogenerate the BCP 47 to OpenType mappings The new script, gen-tag-table.py, generates `ot_languages` automatically from the [OpenType language system tag registry][ot] and the [IANA Language Subtag Registry][bcp47] with some manual modifications. If an OpenType tag maps to a BCP 47 macrolanguage, all the macrolanguage's individual languages are mapped to the same OpenType tag, except for individual languages with their own OpenType mappings. Deprecated BCP 47 tags are canonicalized. [ot]: https://docs.microsoft.com/en-us/typography/opentype/spec/languagetags [bcp47]: https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry Some OpenType tags correspond to multiple ISO 639 codes. The mapping from ISO 639 codes lists OpenType tags in priority order, such that more specific or more likely tags appear first. Some OpenType tags have no corresponding ISO 639 code in the registry so their mappings use BCP 47 subtags besides the language. For example, any BCP 47 tag with a fonipa variant subtag is mapped to 'IPPH', and 'IPPH' is mapped back to und-fonipa. Other OpenType tags have no corresponding ISO 639 code because it is not clear what they are for. HarfBuzz just ignores these tags. One such ignored tag is 'ZHP ' (Chinese Phonetic). It probably means zh-Latn. However, it is used in Microsoft JhengHei and Microsoft YaHei with the script tag 'hani', implying that it is not a romanization scheme after all. It would be simple enough to add this mapping to gen-tag-table.py once a definitive mapping is determined. The manual modifications are mainly either obvious mappings that the OpenType registry omits or mappings for compatibility with previous versions of HarfBuzz. Some of the old mappings were discarded, though, for homophonous language names. For example, OpenType maps 'KUI ' to kxu; previous versions of HarfBuzz also mapped it to kvd, because kvd and kxu both happen to be called "Kui". gen-tag-table.py also generates a function to convert multi-subtag tags like el-polyton and zh-HK to OpenType tags, replacing `ot_languages_zh` and the hard-coded list of special cases in `hb_ot_tags_from_language`. It also generates a function to convert OpenType tags to BCP 47, replacing the hard-coded list of special cases in `hb_ot_tag_to_language`.
2017-12-09 04:45:52 +01:00
/* Check for matches of multiple subtags. */
if (hb_ot_tags_from_complex_language (lang_str, limit, count, tags))
return;
Autogenerate the BCP 47 to OpenType mappings The new script, gen-tag-table.py, generates `ot_languages` automatically from the [OpenType language system tag registry][ot] and the [IANA Language Subtag Registry][bcp47] with some manual modifications. If an OpenType tag maps to a BCP 47 macrolanguage, all the macrolanguage's individual languages are mapped to the same OpenType tag, except for individual languages with their own OpenType mappings. Deprecated BCP 47 tags are canonicalized. [ot]: https://docs.microsoft.com/en-us/typography/opentype/spec/languagetags [bcp47]: https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry Some OpenType tags correspond to multiple ISO 639 codes. The mapping from ISO 639 codes lists OpenType tags in priority order, such that more specific or more likely tags appear first. Some OpenType tags have no corresponding ISO 639 code in the registry so their mappings use BCP 47 subtags besides the language. For example, any BCP 47 tag with a fonipa variant subtag is mapped to 'IPPH', and 'IPPH' is mapped back to und-fonipa. Other OpenType tags have no corresponding ISO 639 code because it is not clear what they are for. HarfBuzz just ignores these tags. One such ignored tag is 'ZHP ' (Chinese Phonetic). It probably means zh-Latn. However, it is used in Microsoft JhengHei and Microsoft YaHei with the script tag 'hani', implying that it is not a romanization scheme after all. It would be simple enough to add this mapping to gen-tag-table.py once a definitive mapping is determined. The manual modifications are mainly either obvious mappings that the OpenType registry omits or mappings for compatibility with previous versions of HarfBuzz. Some of the old mappings were discarded, though, for homophonous language names. For example, OpenType maps 'KUI ' to kxu; previous versions of HarfBuzz also mapped it to kvd, because kvd and kxu both happen to be called "Kui". gen-tag-table.py also generates a function to convert multi-subtag tags like el-polyton and zh-HK to OpenType tags, replacing `ot_languages_zh` and the hard-coded list of special cases in `hb_ot_tags_from_language`. It also generates a function to convert OpenType tags to BCP 47, replacing the hard-coded list of special cases in `hb_ot_tag_to_language`.
2017-12-09 04:45:52 +01:00
/* Find a language matching in the first component. */
s = strchr (lang_str, '-');
{
if (s && limit - lang_str >= 6)
{
const char *extlang_end = strchr (s + 1, '-');
/* If there is an extended language tag, use it. */
if (3 == (extlang_end ? extlang_end - s - 1 : strlen (s + 1)) &&
ISALPHA (s[1]))
lang_str = s + 1;
}
[ot-tags] Speed up hb_ot_tags_from_language() Part of https://github.com/harfbuzz/harfbuzz/issues/3591 "After that, bulk of the time I suppose is spent in binary-searching the language table. I suggest we split the language table in 2-letter and 3-letter tags, to speed-up the vast majority of cases that are 2-letter." benchmark-ot, before: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 112 ns 111 ns 6286271 BM_hb_ot_tags_from_script_and_language/COMMON en_US 60.6 ns 60.4 ns 11671176 BM_hb_ot_tags_from_script_and_language/LATIN en_US 61.3 ns 61.1 ns 11442645 BM_hb_ot_tags_from_script_and_language/COMMON none 4.75 ns 4.74 ns 146997235 BM_hb_ot_tags_from_script_and_language/LATIN none 4.65 ns 4.64 ns 150938747 After: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 89.5 ns 89.2 ns 7747649 BM_hb_ot_tags_from_script_and_language/COMMON en_US 38.5 ns 38.4 ns 18199432 BM_hb_ot_tags_from_script_and_language/LATIN en_US 39.0 ns 38.9 ns 18049238 BM_hb_ot_tags_from_script_and_language/COMMON none 4.53 ns 4.52 ns 154895110 BM_hb_ot_tags_from_script_and_language/LATIN none 4.54 ns 4.52 ns 154762105
2022-05-17 22:28:28 +02:00
const LangTag *ot_languages = nullptr;
unsigned ot_languages_len = 0;
const char *dash = strchr (lang_str, '-');
unsigned first_len = dash ? dash - lang_str : limit - lang_str;
if (first_len == 2)
{
ot_languages = ot_languages2;
ot_languages_len = ARRAY_LENGTH (ot_languages2);
}
else if (first_len == 3)
{
ot_languages = ot_languages3;
ot_languages_len = ARRAY_LENGTH (ot_languages3);
}
[ot-tag] Cache last bsearch result Part of https://github.com/harfbuzz/harfbuzz/issues/3591 Humm. Looks like not all of the fat is bsearch overhead now. I cached the last bsearch result, but most of the time is still there. I'm baffled. Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.08 ns 8.05 ns 84500482 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 42.2 ns 42.1 ns 16722006 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43461527 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.5 ns 42448505 BM_hb_ot_tags_from_script_and_language/COMMON none 4.34 ns 4.33 ns 161290530 BM_hb_ot_tags_from_script_and_language/LATIN none 4.34 ns 4.33 ns 162339799 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.13 ns 8.11 ns 80438134 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 40.0 ns 39.9 ns 17487939 BM_hb_ot_tags_from_script_and_language/COMMON en_US 12.7 ns 12.7 ns 55124394 BM_hb_ot_tags_from_script_and_language/LATIN en_US 13.1 ns 13.0 ns 53660125 BM_hb_ot_tags_from_script_and_language/COMMON none 4.61 ns 4.60 ns 151394104 BM_hb_ot_tags_from_script_and_language/LATIN none 4.70 ns 4.68 ns 150402847
2022-05-18 00:21:02 +02:00
hb_tag_t lang_tag = hb_tag_from_string (lang_str, first_len);
static hb_atomic_int_t last_tag_idx; /* Poor man's cache. */
unsigned tag_idx = last_tag_idx.get_relaxed ();
[ot-tag] Cache last bsearch result Part of https://github.com/harfbuzz/harfbuzz/issues/3591 Humm. Looks like not all of the fat is bsearch overhead now. I cached the last bsearch result, but most of the time is still there. I'm baffled. Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.08 ns 8.05 ns 84500482 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 42.2 ns 42.1 ns 16722006 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43461527 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.5 ns 42448505 BM_hb_ot_tags_from_script_and_language/COMMON none 4.34 ns 4.33 ns 161290530 BM_hb_ot_tags_from_script_and_language/LATIN none 4.34 ns 4.33 ns 162339799 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.13 ns 8.11 ns 80438134 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 40.0 ns 39.9 ns 17487939 BM_hb_ot_tags_from_script_and_language/COMMON en_US 12.7 ns 12.7 ns 55124394 BM_hb_ot_tags_from_script_and_language/LATIN en_US 13.1 ns 13.0 ns 53660125 BM_hb_ot_tags_from_script_and_language/COMMON none 4.61 ns 4.60 ns 151394104 BM_hb_ot_tags_from_script_and_language/LATIN none 4.70 ns 4.68 ns 150402847
2022-05-18 00:21:02 +02:00
if (likely (tag_idx < ot_languages_len && ot_languages[tag_idx].language == lang_tag) ||
[ot-tag] Cache last bsearch result Part of https://github.com/harfbuzz/harfbuzz/issues/3591 Humm. Looks like not all of the fat is bsearch overhead now. I cached the last bsearch result, but most of the time is still there. I'm baffled. Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.08 ns 8.05 ns 84500482 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 42.2 ns 42.1 ns 16722006 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43461527 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.5 ns 42448505 BM_hb_ot_tags_from_script_and_language/COMMON none 4.34 ns 4.33 ns 161290530 BM_hb_ot_tags_from_script_and_language/LATIN none 4.34 ns 4.33 ns 162339799 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.13 ns 8.11 ns 80438134 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 40.0 ns 39.9 ns 17487939 BM_hb_ot_tags_from_script_and_language/COMMON en_US 12.7 ns 12.7 ns 55124394 BM_hb_ot_tags_from_script_and_language/LATIN en_US 13.1 ns 13.0 ns 53660125 BM_hb_ot_tags_from_script_and_language/COMMON none 4.61 ns 4.60 ns 151394104 BM_hb_ot_tags_from_script_and_language/LATIN none 4.70 ns 4.68 ns 150402847
2022-05-18 00:21:02 +02:00
hb_sorted_array (ot_languages, ot_languages_len).bfind (lang_tag, &tag_idx))
{
last_tag_idx.set_relaxed (tag_idx);
unsigned int i;
while (tag_idx != 0 &&
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
ot_languages[tag_idx].language == ot_languages[tag_idx - 1].language)
tag_idx--;
for (i = 0;
i < *count &&
[ot-tags] Speed up hb_ot_tags_from_language() Part of https://github.com/harfbuzz/harfbuzz/issues/3591 "After that, bulk of the time I suppose is spent in binary-searching the language table. I suggest we split the language table in 2-letter and 3-letter tags, to speed-up the vast majority of cases that are 2-letter." benchmark-ot, before: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 112 ns 111 ns 6286271 BM_hb_ot_tags_from_script_and_language/COMMON en_US 60.6 ns 60.4 ns 11671176 BM_hb_ot_tags_from_script_and_language/LATIN en_US 61.3 ns 61.1 ns 11442645 BM_hb_ot_tags_from_script_and_language/COMMON none 4.75 ns 4.74 ns 146997235 BM_hb_ot_tags_from_script_and_language/LATIN none 4.65 ns 4.64 ns 150938747 After: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 89.5 ns 89.2 ns 7747649 BM_hb_ot_tags_from_script_and_language/COMMON en_US 38.5 ns 38.4 ns 18199432 BM_hb_ot_tags_from_script_and_language/LATIN en_US 39.0 ns 38.9 ns 18049238 BM_hb_ot_tags_from_script_and_language/COMMON none 4.53 ns 4.52 ns 154895110 BM_hb_ot_tags_from_script_and_language/LATIN none 4.54 ns 4.52 ns 154762105
2022-05-17 22:28:28 +02:00
tag_idx + i < ot_languages_len &&
ot_languages[tag_idx + i].tag != HB_TAG_NONE &&
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
ot_languages[tag_idx + i].language == ot_languages[tag_idx].language;
i++)
tags[i] = ot_languages[tag_idx + i].tag;
*count = i;
return;
}
}
if (!s)
s = lang_str + strlen (lang_str);
if (s - lang_str == 3) {
/* Assume it's ISO-639-3 and upper-case and use it. */
tags[0] = hb_tag_from_string (lang_str, s - lang_str) & ~0x20202000u;
*count = 1;
return;
}
*count = 0;
}
static bool
parse_private_use_subtag (const char *private_use_subtag,
unsigned int *count,
hb_tag_t *tags,
const char *prefix,
unsigned char (*normalize) (unsigned char))
{
#ifdef HB_NO_LANGUAGE_PRIVATE_SUBTAG
return false;
#endif
2019-06-20 01:36:09 +02:00
if (!(private_use_subtag && count && tags && *count)) return false;
const char *s = strstr (private_use_subtag, prefix);
if (!s) return false;
char tag[4];
int i;
s += strlen (prefix);
if (s[0] == '-') {
s += 1;
char c;
for (i = 0; i < 8 && ISHEX (s[i]); i++)
{
c = FROMHEX (s[i]);
if (i % 2 == 0)
tag[i / 2] = c << 4;
else
tag[i / 2] += c;
}
if (i != 8) return false;
} else {
for (i = 0; i < 4 && ISALNUM (s[i]); i++)
tag[i] = normalize (s[i]);
if (!i) return false;
for (; i < 4; i++)
tag[i] = ' ';
}
2019-06-20 01:36:09 +02:00
tags[0] = HB_TAG (tag[0], tag[1], tag[2], tag[3]);
if ((tags[0] & 0xDFDFDFDF) == HB_OT_TAG_DEFAULT_SCRIPT)
tags[0] ^= ~0xDFDFDFDF;
*count = 1;
return true;
}
/**
2018-07-24 03:19:23 +02:00
* hb_ot_tags_from_script_and_language:
* @script: an #hb_script_t to convert.
* @language: an #hb_language_t to convert.
* @script_count: (inout) (optional): maximum number of script tags to retrieve (IN)
* and actual number of script tags retrieved (OUT)
* @script_tags: (out) (optional): array of size at least @script_count to store the
* script tag results
* @language_count: (inout) (optional): maximum number of language tags to retrieve
* (IN) and actual number of language tags retrieved (OUT)
* @language_tags: (out) (optional): array of size at least @language_count to store
* the language tag results
*
* Converts an #hb_script_t and an #hb_language_t to script and language tags.
*
2018-10-18 14:58:17 +02:00
* Since: 2.0.0
**/
void
2018-07-24 03:19:23 +02:00
hb_ot_tags_from_script_and_language (hb_script_t script,
hb_language_t language,
unsigned int *script_count /* IN/OUT */,
hb_tag_t *script_tags /* OUT */,
unsigned int *language_count /* IN/OUT */,
hb_tag_t *language_tags /* OUT */)
{
bool needs_script = true;
if (language == HB_LANGUAGE_INVALID)
{
if (language_count && language_tags && *language_count)
*language_count = 0;
2018-10-11 20:30:15 +02:00
}
else
{
const char *lang_str, *s, *limit, *private_use_subtag;
bool needs_language;
lang_str = hb_language_to_string (language);
limit = nullptr;
private_use_subtag = nullptr;
if (lang_str[0] == 'x' && lang_str[1] == '-')
{
private_use_subtag = lang_str;
} else {
for (s = lang_str + 1; *s; s++)
{
if (s[-1] == '-' && s[1] == '-')
{
if (s[0] == 'x')
{
private_use_subtag = s;
if (!limit)
limit = s - 1;
break;
} else if (!limit)
{
limit = s - 1;
}
}
}
if (!limit)
limit = s;
}
2019-06-20 01:36:09 +02:00
needs_script = !parse_private_use_subtag (private_use_subtag, script_count, script_tags, "-hbsc", TOLOWER);
needs_language = !parse_private_use_subtag (private_use_subtag, language_count, language_tags, "-hbot", TOUPPER);
if (needs_language && language_count && language_tags && *language_count)
hb_ot_tags_from_language (lang_str, limit, language_count, language_tags);
}
if (needs_script && script_count && script_tags && *script_count)
hb_ot_all_tags_from_script (script, script_count, script_tags);
}
/**
* hb_ot_tag_to_language:
* @tag: an language tag
*
* Converts a language tag to an #hb_language_t.
*
* Return value: (transfer none) (nullable):
* The #hb_language_t corresponding to @tag.
*
* Since: 0.9.2
**/
hb_language_t
hb_ot_tag_to_language (hb_tag_t tag)
{
unsigned int i;
if (tag == HB_OT_TAG_DEFAULT_LANGUAGE)
2017-10-15 12:11:08 +02:00
return nullptr;
Autogenerate the BCP 47 to OpenType mappings The new script, gen-tag-table.py, generates `ot_languages` automatically from the [OpenType language system tag registry][ot] and the [IANA Language Subtag Registry][bcp47] with some manual modifications. If an OpenType tag maps to a BCP 47 macrolanguage, all the macrolanguage's individual languages are mapped to the same OpenType tag, except for individual languages with their own OpenType mappings. Deprecated BCP 47 tags are canonicalized. [ot]: https://docs.microsoft.com/en-us/typography/opentype/spec/languagetags [bcp47]: https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry Some OpenType tags correspond to multiple ISO 639 codes. The mapping from ISO 639 codes lists OpenType tags in priority order, such that more specific or more likely tags appear first. Some OpenType tags have no corresponding ISO 639 code in the registry so their mappings use BCP 47 subtags besides the language. For example, any BCP 47 tag with a fonipa variant subtag is mapped to 'IPPH', and 'IPPH' is mapped back to und-fonipa. Other OpenType tags have no corresponding ISO 639 code because it is not clear what they are for. HarfBuzz just ignores these tags. One such ignored tag is 'ZHP ' (Chinese Phonetic). It probably means zh-Latn. However, it is used in Microsoft JhengHei and Microsoft YaHei with the script tag 'hani', implying that it is not a romanization scheme after all. It would be simple enough to add this mapping to gen-tag-table.py once a definitive mapping is determined. The manual modifications are mainly either obvious mappings that the OpenType registry omits or mappings for compatibility with previous versions of HarfBuzz. Some of the old mappings were discarded, though, for homophonous language names. For example, OpenType maps 'KUI ' to kxu; previous versions of HarfBuzz also mapped it to kvd, because kvd and kxu both happen to be called "Kui". gen-tag-table.py also generates a function to convert multi-subtag tags like el-polyton and zh-HK to OpenType tags, replacing `ot_languages_zh` and the hard-coded list of special cases in `hb_ot_tags_from_language`. It also generates a function to convert OpenType tags to BCP 47, replacing the hard-coded list of special cases in `hb_ot_tag_to_language`.
2017-12-09 04:45:52 +01:00
{
hb_language_t disambiguated_tag = hb_ot_ambiguous_tag_to_language (tag);
if (disambiguated_tag != HB_LANGUAGE_INVALID)
return disambiguated_tag;
}
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
char buf[4];
[ot-tags] Speed up hb_ot_tags_from_language() Part of https://github.com/harfbuzz/harfbuzz/issues/3591 "After that, bulk of the time I suppose is spent in binary-searching the language table. I suggest we split the language table in 2-letter and 3-letter tags, to speed-up the vast majority of cases that are 2-letter." benchmark-ot, before: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 112 ns 111 ns 6286271 BM_hb_ot_tags_from_script_and_language/COMMON en_US 60.6 ns 60.4 ns 11671176 BM_hb_ot_tags_from_script_and_language/LATIN en_US 61.3 ns 61.1 ns 11442645 BM_hb_ot_tags_from_script_and_language/COMMON none 4.75 ns 4.74 ns 146997235 BM_hb_ot_tags_from_script_and_language/LATIN none 4.65 ns 4.64 ns 150938747 After: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 89.5 ns 89.2 ns 7747649 BM_hb_ot_tags_from_script_and_language/COMMON en_US 38.5 ns 38.4 ns 18199432 BM_hb_ot_tags_from_script_and_language/LATIN en_US 39.0 ns 38.9 ns 18049238 BM_hb_ot_tags_from_script_and_language/COMMON none 4.53 ns 4.52 ns 154895110 BM_hb_ot_tags_from_script_and_language/LATIN none 4.54 ns 4.52 ns 154762105
2022-05-17 22:28:28 +02:00
for (i = 0; i < ARRAY_LENGTH (ot_languages2); i++)
if (ot_languages2[i].tag == tag)
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
{
hb_tag_to_string (ot_languages2[i].language, buf);
return hb_language_from_string (buf, 2);
}
[ot-tags] Speed up hb_ot_tags_from_language() Part of https://github.com/harfbuzz/harfbuzz/issues/3591 "After that, bulk of the time I suppose is spent in binary-searching the language table. I suggest we split the language table in 2-letter and 3-letter tags, to speed-up the vast majority of cases that are 2-letter." benchmark-ot, before: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 112 ns 111 ns 6286271 BM_hb_ot_tags_from_script_and_language/COMMON en_US 60.6 ns 60.4 ns 11671176 BM_hb_ot_tags_from_script_and_language/LATIN en_US 61.3 ns 61.1 ns 11442645 BM_hb_ot_tags_from_script_and_language/COMMON none 4.75 ns 4.74 ns 146997235 BM_hb_ot_tags_from_script_and_language/LATIN none 4.65 ns 4.64 ns 150938747 After: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 89.5 ns 89.2 ns 7747649 BM_hb_ot_tags_from_script_and_language/COMMON en_US 38.5 ns 38.4 ns 18199432 BM_hb_ot_tags_from_script_and_language/LATIN en_US 39.0 ns 38.9 ns 18049238 BM_hb_ot_tags_from_script_and_language/COMMON none 4.53 ns 4.52 ns 154895110 BM_hb_ot_tags_from_script_and_language/LATIN none 4.54 ns 4.52 ns 154762105
2022-05-17 22:28:28 +02:00
for (i = 0; i < ARRAY_LENGTH (ot_languages3); i++)
if (ot_languages3[i].tag == tag)
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
{
hb_tag_to_string (ot_languages3[i].language, buf);
return hb_language_from_string (buf, 3);
}
/* Return a custom language in the form of "x-hbot-AABBCCDD".
* If it's three letters long, also guess it's ISO 639-3 and lower-case and
* prepend it (if it's not a registered tag, the private use subtags will
* ensure that calling hb_ot_tag_from_language on the result will still return
* the same tag as the original tag).
*/
{
char buf[20];
char *str = buf;
if (ISALPHA (tag >> 24)
&& ISALPHA ((tag >> 16) & 0xFF)
&& ISALPHA ((tag >> 8) & 0xFF)
&& (tag & 0xFF) == ' ')
{
buf[0] = TOLOWER (tag >> 24);
buf[1] = TOLOWER ((tag >> 16) & 0xFF);
buf[2] = TOLOWER ((tag >> 8) & 0xFF);
buf[3] = '-';
str += 4;
}
snprintf (str, 16, "x-hbot-%08x", tag);
return hb_language_from_string (&*buf, -1);
}
}
2010-07-23 21:11:18 +02:00
2018-07-24 03:19:23 +02:00
/**
* hb_ot_tags_to_script_and_language:
* @script_tag: a script tag
* @language_tag: a language tag
* @script: (out) (optional): the #hb_script_t corresponding to @script_tag.
* @language: (out) (optional): the #hb_language_t corresponding to @script_tag and
* @language_tag.
2018-07-24 03:19:23 +02:00
*
* Converts a script tag and a language tag to an #hb_script_t and an
* #hb_language_t.
*
2018-10-18 14:58:17 +02:00
* Since: 2.0.0
2018-07-24 03:19:23 +02:00
**/
void
hb_ot_tags_to_script_and_language (hb_tag_t script_tag,
hb_tag_t language_tag,
hb_script_t *script /* OUT */,
hb_language_t *language /* OUT */)
{
hb_script_t script_out = hb_ot_tag_to_script (script_tag);
if (script)
*script = script_out;
if (language)
{
2018-07-24 03:19:23 +02:00
unsigned int script_count = 1;
hb_tag_t primary_script_tag[1];
hb_ot_tags_from_script_and_language (script_out,
HB_LANGUAGE_INVALID,
&script_count,
primary_script_tag,
nullptr, nullptr);
2018-07-24 03:19:23 +02:00
*language = hb_ot_tag_to_language (language_tag);
if (script_count == 0 || primary_script_tag[0] != script_tag)
{
2018-07-24 03:19:23 +02:00
unsigned char *buf;
const char *lang_str = hb_language_to_string (*language);
size_t len = strlen (lang_str);
buf = (unsigned char *) hb_malloc (len + 16);
if (unlikely (!buf))
{
2018-07-24 03:19:23 +02:00
*language = nullptr;
}
else
{
int shift;
2018-07-24 03:19:23 +02:00
memcpy (buf, lang_str, len);
if (lang_str[0] != 'x' || lang_str[1] != '-') {
buf[len++] = '-';
buf[len++] = 'x';
}
buf[len++] = '-';
buf[len++] = 'h';
buf[len++] = 'b';
buf[len++] = 's';
buf[len++] = 'c';
buf[len++] = '-';
for (shift = 28; shift >= 0; shift -= 4)
buf[len++] = TOHEX (script_tag >> shift);
2018-07-24 03:19:23 +02:00
*language = hb_language_from_string ((char *) buf, len);
hb_free (buf);
2018-07-24 03:19:23 +02:00
}
}
}
}
#ifdef MAIN
static inline void
test_langs_sorted ()
{
[ot-tags] Speed up hb_ot_tags_from_language() Part of https://github.com/harfbuzz/harfbuzz/issues/3591 "After that, bulk of the time I suppose is spent in binary-searching the language table. I suggest we split the language table in 2-letter and 3-letter tags, to speed-up the vast majority of cases that are 2-letter." benchmark-ot, before: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 112 ns 111 ns 6286271 BM_hb_ot_tags_from_script_and_language/COMMON en_US 60.6 ns 60.4 ns 11671176 BM_hb_ot_tags_from_script_and_language/LATIN en_US 61.3 ns 61.1 ns 11442645 BM_hb_ot_tags_from_script_and_language/COMMON none 4.75 ns 4.74 ns 146997235 BM_hb_ot_tags_from_script_and_language/LATIN none 4.65 ns 4.64 ns 150938747 After: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 89.5 ns 89.2 ns 7747649 BM_hb_ot_tags_from_script_and_language/COMMON en_US 38.5 ns 38.4 ns 18199432 BM_hb_ot_tags_from_script_and_language/LATIN en_US 39.0 ns 38.9 ns 18049238 BM_hb_ot_tags_from_script_and_language/COMMON none 4.53 ns 4.52 ns 154895110 BM_hb_ot_tags_from_script_and_language/LATIN none 4.54 ns 4.52 ns 154762105
2022-05-17 22:28:28 +02:00
for (unsigned int i = 1; i < ARRAY_LENGTH (ot_languages2); i++)
{
int c = ot_languages2[i].cmp (&ot_languages2[i - 1]);
if (c > 0)
{
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
fprintf (stderr, "ot_languages2 not sorted at index %d: %08x %d %08x\n",
[ot-tags] Speed up hb_ot_tags_from_language() Part of https://github.com/harfbuzz/harfbuzz/issues/3591 "After that, bulk of the time I suppose is spent in binary-searching the language table. I suggest we split the language table in 2-letter and 3-letter tags, to speed-up the vast majority of cases that are 2-letter." benchmark-ot, before: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 112 ns 111 ns 6286271 BM_hb_ot_tags_from_script_and_language/COMMON en_US 60.6 ns 60.4 ns 11671176 BM_hb_ot_tags_from_script_and_language/LATIN en_US 61.3 ns 61.1 ns 11442645 BM_hb_ot_tags_from_script_and_language/COMMON none 4.75 ns 4.74 ns 146997235 BM_hb_ot_tags_from_script_and_language/LATIN none 4.65 ns 4.64 ns 150938747 After: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 89.5 ns 89.2 ns 7747649 BM_hb_ot_tags_from_script_and_language/COMMON en_US 38.5 ns 38.4 ns 18199432 BM_hb_ot_tags_from_script_and_language/LATIN en_US 39.0 ns 38.9 ns 18049238 BM_hb_ot_tags_from_script_and_language/COMMON none 4.53 ns 4.52 ns 154895110 BM_hb_ot_tags_from_script_and_language/LATIN none 4.54 ns 4.52 ns 154762105
2022-05-17 22:28:28 +02:00
i, ot_languages2[i-1].language, c, ot_languages2[i].language);
abort();
}
}
for (unsigned int i = 1; i < ARRAY_LENGTH (ot_languages3); i++)
{
[ot-tags] Speed up hb_ot_tags_from_language() Part of https://github.com/harfbuzz/harfbuzz/issues/3591 "After that, bulk of the time I suppose is spent in binary-searching the language table. I suggest we split the language table in 2-letter and 3-letter tags, to speed-up the vast majority of cases that are 2-letter." benchmark-ot, before: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 112 ns 111 ns 6286271 BM_hb_ot_tags_from_script_and_language/COMMON en_US 60.6 ns 60.4 ns 11671176 BM_hb_ot_tags_from_script_and_language/LATIN en_US 61.3 ns 61.1 ns 11442645 BM_hb_ot_tags_from_script_and_language/COMMON none 4.75 ns 4.74 ns 146997235 BM_hb_ot_tags_from_script_and_language/LATIN none 4.65 ns 4.64 ns 150938747 After: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 89.5 ns 89.2 ns 7747649 BM_hb_ot_tags_from_script_and_language/COMMON en_US 38.5 ns 38.4 ns 18199432 BM_hb_ot_tags_from_script_and_language/LATIN en_US 39.0 ns 38.9 ns 18049238 BM_hb_ot_tags_from_script_and_language/COMMON none 4.53 ns 4.52 ns 154895110 BM_hb_ot_tags_from_script_and_language/LATIN none 4.54 ns 4.52 ns 154762105
2022-05-17 22:28:28 +02:00
int c = ot_languages3[i].cmp (&ot_languages3[i - 1]);
if (c > 0)
{
[ot-tags] Further speed up language bsearch() Using an integer tag to bsearch, instead of string. Part of: https://github.com/harfbuzz/harfbuzz/issues/3591 Before: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 8.11 ns 8.08 ns 87067795 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 53.6 ns 53.5 ns 13042418 BM_hb_ot_tags_from_script_and_language/COMMON en_US 24.2 ns 24.1 ns 29052731 BM_hb_ot_tags_from_script_and_language/LATIN en_US 24.4 ns 24.3 ns 28736769 BM_hb_ot_tags_from_script_and_language/COMMON none 4.43 ns 4.41 ns 160370413 BM_hb_ot_tags_from_script_and_language/LATIN none 4.35 ns 4.34 ns 160578191 After: ------------------------------------------------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------------------------------------------------ BM_hb_ot_tags_from_script_and_language/COMMON abcd_XY 7.97 ns 7.95 ns 85208363 BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 41.7 ns 41.6 ns 16945817 BM_hb_ot_tags_from_script_and_language/COMMON en_US 16.1 ns 16.0 ns 43613523 BM_hb_ot_tags_from_script_and_language/LATIN en_US 16.5 ns 16.4 ns 42568107 BM_hb_ot_tags_from_script_and_language/COMMON none 4.30 ns 4.29 ns 164055469 BM_hb_ot_tags_from_script_and_language/LATIN none 4.29 ns 4.27 ns 163793591
2022-05-17 23:51:41 +02:00
fprintf (stderr, "ot_languages3 not sorted at index %d: %08x %d %08x\n",
[ot-tags] Speed up hb_ot_tags_from_language() Part of https://github.com/harfbuzz/harfbuzz/issues/3591 "After that, bulk of the time I suppose is spent in binary-searching the language table. I suggest we split the language table in 2-letter and 3-letter tags, to speed-up the vast majority of cases that are 2-letter." benchmark-ot, before: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 112 ns 111 ns 6286271 BM_hb_ot_tags_from_script_and_language/COMMON en_US 60.6 ns 60.4 ns 11671176 BM_hb_ot_tags_from_script_and_language/LATIN en_US 61.3 ns 61.1 ns 11442645 BM_hb_ot_tags_from_script_and_language/COMMON none 4.75 ns 4.74 ns 146997235 BM_hb_ot_tags_from_script_and_language/LATIN none 4.65 ns 4.64 ns 150938747 After: ---------------------------------------------------------------------------------------------- Benchmark Time CPU Iterations ---------------------------------------------------------------------------------------------- BM_hb_ot_tags_from_script_and_language/COMMON zh_CN 89.5 ns 89.2 ns 7747649 BM_hb_ot_tags_from_script_and_language/COMMON en_US 38.5 ns 38.4 ns 18199432 BM_hb_ot_tags_from_script_and_language/LATIN en_US 39.0 ns 38.9 ns 18049238 BM_hb_ot_tags_from_script_and_language/COMMON none 4.53 ns 4.52 ns 154895110 BM_hb_ot_tags_from_script_and_language/LATIN none 4.54 ns 4.52 ns 154762105
2022-05-17 22:28:28 +02:00
i, ot_languages3[i-1].language, c, ot_languages3[i].language);
abort();
}
}
}
int
main ()
{
test_langs_sorted ();
return 0;
}
2010-07-23 21:11:18 +02:00
#endif
#endif