src/hb-icu.c

   1 /*
   2  * Copyright (C) 2009  Red Hat, Inc.
   3  * Copyright (C) 2009  Keith Stribley <devel@thanlwinsoft.org>
   4  *
   5  *  This is part of HarfBuzz, a text shaping library.
   6  *
   7  * Permission is hereby granted, without written agreement and without
   8  * license or royalty fees, to use, copy, modify, and distribute this
   9  * software and its documentation for any purpose, provided that the
  10  * above copyright notice and the following two paragraphs appear in
  11  * all copies of this software.
  12  *
  13  * IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE TO ANY PARTY FOR
  14  * DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES
  15  * ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN
  16  * IF THE COPYRIGHT HOLDER HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
  17  * DAMAGE.
  18  *
  19  * THE COPYRIGHT HOLDER SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING,
  20  * BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
  21  * FITNESS FOR A PARTICULAR PURPOSE.  THE SOFTWARE PROVIDED HEREUNDER IS
  22  * ON AN "AS IS" BASIS, AND THE COPYRIGHT HOLDER HAS NO OBLIGATION TO
  23  * PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
  24  *
  25  * Red Hat Author(s): Behdad Esfahbod
  26  */
  27
  28 #include "hb-private.h"
  29
  30 #include "hb-icu.h"
  31
  32 #include "hb-unicode-private.h"
  33
  34 #include <unicode/uversion.h>
  35 #include <unicode/uchar.h>
  36 #include <unicode/uscript.h>
  37
  38 HB_BEGIN_DECLS
  39
  40
  41 static hb_codepoint_t hb_icu_get_mirroring (hb_codepoint_t unicode) { return u_charMirror(unicode); }
  42 static unsigned int hb_icu_get_combining_class (hb_codepoint_t unicode) { return u_getCombiningClass (unicode); }
  43
  44 static unsigned int
  45 hb_icu_get_eastasian_width (hb_codepoint_t unicode)
  46 {
  47   switch (u_getIntPropertyValue(unicode, UCHAR_EAST_ASIAN_WIDTH))
  48   {
  49   case U_EA_WIDE:
  50   case U_EA_FULLWIDTH:
  51     return 2;
  52   case U_EA_NEUTRAL:
  53   case U_EA_AMBIGUOUS:
  54   case U_EA_HALFWIDTH:
  55   case U_EA_NARROW:
  56     return 1;
  57   }
  58   return 1;
  59 }
  60
  61 static hb_category_t
  62 hb_icu_get_general_category (hb_codepoint_t unicode)
  63 {
  64   switch (u_getIntPropertyValue(unicode, UCHAR_GENERAL_CATEGORY))
  65   {
  66   case U_UNASSIGNED:                    return HB_CATEGORY_UNASSIGNED;
  67
  68   case U_UPPERCASE_LETTER:              return HB_CATEGORY_UPPERCASE_LETTER;    /* Lu */
  69   case U_LOWERCASE_LETTER:              return HB_CATEGORY_LOWERCASE_LETTER;    /* Ll */
  70   case U_TITLECASE_LETTER:              return HB_CATEGORY_TITLECASE_LETTER;    /* Lt */
  71   case U_MODIFIER_LETTER:               return HB_CATEGORY_MODIFIER_LETTER;     /* Lm */
  72   case U_OTHER_LETTER:                  return HB_CATEGORY_OTHER_LETTER;        /* Lo */
  73
  74   case U_NON_SPACING_MARK:              return HB_CATEGORY_NON_SPACING_MARK;    /* Mn */
  75   case U_ENCLOSING_MARK:                return HB_CATEGORY_ENCLOSING_MARK;      /* Me */
  76   case U_COMBINING_SPACING_MARK:        return HB_CATEGORY_COMBINING_MARK;      /* Mc */
  77
  78   case U_DECIMAL_DIGIT_NUMBER:          return HB_CATEGORY_DECIMAL_NUMBER;      /* Nd */
  79   case U_LETTER_NUMBER:                 return HB_CATEGORY_LETTER_NUMBER;       /* Nl */
  80   case U_OTHER_NUMBER:                  return HB_CATEGORY_OTHER_NUMBER;        /* No */
  81
  82   case U_SPACE_SEPARATOR:               return HB_CATEGORY_SPACE_SEPARATOR;     /* Zs */
  83   case U_LINE_SEPARATOR:                return HB_CATEGORY_LINE_SEPARATOR;      /* Zl */
  84   case U_PARAGRAPH_SEPARATOR:           return HB_CATEGORY_PARAGRAPH_SEPARATOR; /* Zp */
  85
  86   case U_CONTROL_CHAR:                  return HB_CATEGORY_CONTROL;             /* Cc */
  87   case U_FORMAT_CHAR:                   return HB_CATEGORY_FORMAT;              /* Cf */
  88   case U_PRIVATE_USE_CHAR:              return HB_CATEGORY_PRIVATE_USE;         /* Co */
  89   case U_SURROGATE:                     return HB_CATEGORY_SURROGATE;           /* Cs */
  90
  91
  92   case U_DASH_PUNCTUATION:              return HB_CATEGORY_DASH_PUNCTUATION;    /* Pd */
  93   case U_START_PUNCTUATION:             return HB_CATEGORY_OPEN_PUNCTUATION;    /* Ps */
  94   case U_END_PUNCTUATION:               return HB_CATEGORY_CLOSE_PUNCTUATION;   /* Pe */
  95   case U_CONNECTOR_PUNCTUATION:         return HB_CATEGORY_CONNECT_PUNCTUATION; /* Pc */
  96   case U_OTHER_PUNCTUATION:             return HB_CATEGORY_OTHER_PUNCTUATION;   /* Po */
  97
  98   case U_MATH_SYMBOL:                   return HB_CATEGORY_MATH_SYMBOL;         /* Sm */
  99   case U_CURRENCY_SYMBOL:               return HB_CATEGORY_CURRENCY_SYMBOL;     /* Sc */
 100   case U_MODIFIER_SYMBOL:               return HB_CATEGORY_MODIFIER_SYMBOL;     /* Sk */
 101   case U_OTHER_SYMBOL:                  return HB_CATEGORY_OTHER_SYMBOL;        /* So */
 102
 103   case U_INITIAL_PUNCTUATION:           return HB_CATEGORY_INITIAL_PUNCTUATION; /* Pi */
 104   case U_FINAL_PUNCTUATION:             return HB_CATEGORY_FINAL_PUNCTUATION;   /* Pf */
 105   }
 106
 107   return HB_CATEGORY_UNASSIGNED;
 108 }
 109
 110 static hb_script_t
 111 hb_icu_get_script (hb_codepoint_t unicode)
 112 {
 113   UErrorCode status = U_ZERO_ERROR;
 114   UScriptCode scriptCode = uscript_getScript(unicode, &status);
 115   switch ((int) scriptCode)
 116   {
 117 #define CHECK_ICU_VERSION(major, minor) \
 118         U_ICU_VERSION_MAJOR_NUM > (major) || (U_ICU_VERSION_MAJOR_NUM == (major) && U_ICU_VERSION_MINOR_NUM >= (minor))
 119 #define MATCH_SCRIPT(C) case USCRIPT_##C: return HB_SCRIPT_##C
 120 #define MATCH_SCRIPT2(C1, C2) case USCRIPT_##C1: return HB_SCRIPT_##C2
 121   MATCH_SCRIPT (INVALID_CODE);
 122   MATCH_SCRIPT (COMMON);             /* Zyyy */
 123   MATCH_SCRIPT (INHERITED);          /* Qaai */
 124   MATCH_SCRIPT (ARABIC);             /* Arab */
 125   MATCH_SCRIPT (ARMENIAN);           /* Armn */
 126   MATCH_SCRIPT (BENGALI);            /* Beng */
 127   MATCH_SCRIPT (BOPOMOFO);           /* Bopo */
 128   MATCH_SCRIPT (CHEROKEE);           /* Cher */
 129   MATCH_SCRIPT (COPTIC);             /* Qaac */
 130   MATCH_SCRIPT (CYRILLIC);           /* Cyrl (Cyrs) */
 131   MATCH_SCRIPT (DESERET);            /* Dsrt */
 132   MATCH_SCRIPT (DEVANAGARI);         /* Deva */
 133   MATCH_SCRIPT (ETHIOPIC);           /* Ethi */
 134   MATCH_SCRIPT (GEORGIAN);           /* Geor (Geon); Geoa) */
 135   MATCH_SCRIPT (GOTHIC);             /* Goth */
 136   MATCH_SCRIPT (GREEK);              /* Grek */
 137   MATCH_SCRIPT (GUJARATI);           /* Gujr */
 138   MATCH_SCRIPT (GURMUKHI);           /* Guru */
 139   MATCH_SCRIPT (HAN);                /* Hani */
 140   MATCH_SCRIPT (HANGUL);             /* Hang */
 141   MATCH_SCRIPT (HEBREW);             /* Hebr */
 142   MATCH_SCRIPT (HIRAGANA);           /* Hira */
 143   MATCH_SCRIPT (KANNADA);            /* Knda */
 144   MATCH_SCRIPT (KATAKANA);           /* Kana */
 145   MATCH_SCRIPT (KHMER);              /* Khmr */
 146   MATCH_SCRIPT (LAO);                /* Laoo */
 147   MATCH_SCRIPT (LATIN);              /* Latn (Latf); Latg) */
 148   MATCH_SCRIPT (MALAYALAM);          /* Mlym */
 149   MATCH_SCRIPT (MONGOLIAN);          /* Mong */
 150   MATCH_SCRIPT (MYANMAR);            /* Mymr */
 151   MATCH_SCRIPT (OGHAM);              /* Ogam */
 152   MATCH_SCRIPT (OLD_ITALIC);         /* Ital */
 153   MATCH_SCRIPT (ORIYA);              /* Orya */
 154   MATCH_SCRIPT (RUNIC);              /* Runr */
 155   MATCH_SCRIPT (SINHALA);            /* Sinh */
 156   MATCH_SCRIPT (SYRIAC);             /* Syrc (Syrj, Syrn); Syre) */
 157   MATCH_SCRIPT (TAMIL);              /* Taml */
 158   MATCH_SCRIPT (TELUGU);             /* Telu */
 159   MATCH_SCRIPT (THAANA);             /* Thaa */
 160   MATCH_SCRIPT (THAI);               /* Thai */
 161   MATCH_SCRIPT (TIBETAN);            /* Tibt */
 162   MATCH_SCRIPT (CANADIAN_ABORIGINAL);/* Cans */
 163   MATCH_SCRIPT (YI);                 /* Yiii */
 164   MATCH_SCRIPT (TAGALOG);            /* Tglg */
 165   MATCH_SCRIPT (HANUNOO);            /* Hano */
 166   MATCH_SCRIPT (BUHID);              /* Buhd */
 167   MATCH_SCRIPT (TAGBANWA);           /* Tagb */
 168
 169   /* Unicode-4.0 additions */
 170   MATCH_SCRIPT (BRAILLE);            /* Brai */
 171   MATCH_SCRIPT (CYPRIOT);            /* Cprt */
 172   MATCH_SCRIPT (LIMBU);              /* Limb */
 173   MATCH_SCRIPT (OSMANYA);            /* Osma */
 174   MATCH_SCRIPT (SHAVIAN);            /* Shaw */
 175   MATCH_SCRIPT (LINEAR_B);           /* Linb */
 176   MATCH_SCRIPT (TAI_LE);             /* Tale */
 177   MATCH_SCRIPT (UGARITIC);           /* Ugar */
 178
 179   /* Unicode-4.1 additions */
 180   MATCH_SCRIPT (NEW_TAI_LUE);        /* Talu */
 181   MATCH_SCRIPT (BUGINESE);           /* Bugi */
 182   MATCH_SCRIPT (GLAGOLITIC);         /* Glag */
 183   MATCH_SCRIPT (TIFINAGH);           /* Tfng */
 184   MATCH_SCRIPT (SYLOTI_NAGRI);       /* Sylo */
 185   MATCH_SCRIPT (OLD_PERSIAN);        /* Xpeo */
 186   MATCH_SCRIPT (KHAROSHTHI);         /* Khar */
 187
 188   /* Unicode-5.0 additions */
 189   MATCH_SCRIPT (UNKNOWN);            /* Zzzz */
 190   MATCH_SCRIPT (BALINESE);           /* Bali */
 191   MATCH_SCRIPT (CUNEIFORM);          /* Xsux */
 192   MATCH_SCRIPT (PHOENICIAN);         /* Phnx */
 193   MATCH_SCRIPT (PHAGS_PA);           /* Phag */
 194   MATCH_SCRIPT (NKO);                /* Nkoo */
 195
 196   /* Unicode-5.1 additions */
 197   MATCH_SCRIPT (KAYAH_LI);           /* Kali */
 198   MATCH_SCRIPT (LEPCHA);             /* Lepc */
 199   MATCH_SCRIPT (REJANG);             /* Rjng */
 200   MATCH_SCRIPT (SUNDANESE);          /* Sund */
 201   MATCH_SCRIPT (SAURASHTRA);         /* Saur */
 202   MATCH_SCRIPT (CHAM);               /* Cham */
 203   MATCH_SCRIPT (OL_CHIKI);           /* Olck */
 204   MATCH_SCRIPT (VAI);                /* Vaii */
 205   MATCH_SCRIPT (CARIAN);             /* Cari */
 206   MATCH_SCRIPT (LYCIAN);             /* Lyci */
 207   MATCH_SCRIPT (LYDIAN);             /* Lydi */
 208
 209   /* Unicode-5.2 additions */
 210   MATCH_SCRIPT (AVESTAN);                /* Avst */
 211 #if CHECK_ICU_VERSION (4, 4)
 212   MATCH_SCRIPT (BAMUM);                  /* Bamu */
 213 #endif
 214   MATCH_SCRIPT (EGYPTIAN_HIEROGLYPHS);   /* Egyp */
 215   MATCH_SCRIPT (IMPERIAL_ARAMAIC);       /* Armi */
 216   MATCH_SCRIPT (INSCRIPTIONAL_PAHLAVI);  /* Phli */
 217   MATCH_SCRIPT (INSCRIPTIONAL_PARTHIAN); /* Prti */
 218   MATCH_SCRIPT (JAVANESE);               /* Java */
 219   MATCH_SCRIPT (KAITHI);                 /* Kthi */
 220   MATCH_SCRIPT2(LANNA, TAI_THAM);        /* Lana */
 221 #if CHECK_ICU_VERSION (4, 4)
 222   MATCH_SCRIPT (LISU);                   /* Lisu */
 223 #endif
 224   MATCH_SCRIPT (MEITEI_MAYEK);           /* Mtei */
 225 #if CHECK_ICU_VERSION (4, 4)
 226   MATCH_SCRIPT (OLD_SOUTH_ARABIAN);      /* Sarb */
 227 #endif
 228   MATCH_SCRIPT2(ORKHON, OLD_TURKIC);     /* Orkh */
 229   MATCH_SCRIPT (SAMARITAN);              /* Samr */
 230   MATCH_SCRIPT (TAI_VIET);               /* Tavt */
 231   }
 232   return HB_SCRIPT_UNKNOWN;
 233 }
 234
 235 static hb_unicode_funcs_t icu_ufuncs = {
 236   HB_REFERENCE_COUNT_INVALID, /* ref_count */
 237   TRUE, /* immutable */
 238   {
 239     hb_icu_get_general_category,
 240     hb_icu_get_combining_class,
 241     hb_icu_get_mirroring,
 242     hb_icu_get_script,
 243     hb_icu_get_eastasian_width
 244   }
 245 };
 246
 247 hb_unicode_funcs_t *
 248 hb_icu_get_unicode_funcs (void)
 249 {
 250   return &icu_ufuncs;
 251 }
 252
 253
 254 HB_END_DECLS