mdw@git.distorted.org.uk Git - sgt/charset/blob - mimeenc.c

   1 /*
   2  * mimeenc.c - translate our internal character set codes to and
   3  * from MIME standard character-set names.
   4  *
   5  */
   6
   7 #include <ctype.h>
   8 #include "charset.h"
   9 #include "internal.h"
  10
  11 static const struct {
  12     const char *name;
  13     int charset;
  14 } mimeencs[] = {
  15     /*
  16      * Most of these names are taken from
  17      *
  18      *   http://www.iana.org/assignments/character-sets
  19      *
  20      * Where multiple encoding names map to the same encoding id
  21      * (such as the variety of aliases for ISO-8859-1), the first
  22      * is considered canonical and will be returned when
  23      * translating the id to a string.
  24      *
  25      * I also list here a few names which aren't in the above web
  26      * page, but which I've seen in the wild in real mail. These
  27      * are marked with a comment saying WILD.
  28      */
  29
  30     { "US-ASCII", CS_ASCII },
  31     { "ANSI_X3.4-1968", CS_ASCII },
  32     { "iso-ir-6", CS_ASCII },
  33     { "ANSI_X3.4-1986", CS_ASCII },
  34     { "ISO_646.irv:1991", CS_ASCII },
  35     { "ASCII", CS_ASCII },
  36     { "ISO646-US", CS_ASCII },
  37     { "us", CS_ASCII },
  38     { "IBM367", CS_ASCII },
  39     { "cp367", CS_ASCII },
  40     { "csASCII", CS_ASCII },
  41     { "646", CS_ASCII },               /* WILD */
  42
  43     { "ISO-8859-1", CS_ISO8859_1 },
  44     { "ISO8859-1", CS_ISO8859_1 },     /* WILD */
  45     { "iso-ir-100", CS_ISO8859_1 },
  46     { "ISO_8859-1", CS_ISO8859_1 },
  47     { "ISO_8859-1:1987", CS_ISO8859_1 },
  48     { "latin1", CS_ISO8859_1 },
  49     { "l1", CS_ISO8859_1 },
  50     { "IBM819", CS_ISO8859_1 },
  51     { "CP819", CS_ISO8859_1 },
  52     { "csISOLatin1", CS_ISO8859_1 },
  53
  54     { "ISO-8859-2", CS_ISO8859_2 },
  55     { "ISO8859-2", CS_ISO8859_2 },     /* WILD */
  56     { "ISO_8859-2:1987", CS_ISO8859_2 },
  57     { "iso-ir-101", CS_ISO8859_2 },
  58     { "ISO_8859-2", CS_ISO8859_2 },
  59     { "latin2", CS_ISO8859_2 },
  60     { "l2", CS_ISO8859_2 },
  61     { "csISOLatin2", CS_ISO8859_2 },
  62
  63     { "ISO-8859-3", CS_ISO8859_3 },
  64     { "ISO8859-3", CS_ISO8859_3 },     /* WILD */
  65     { "ISO_8859-3:1988", CS_ISO8859_3 },
  66     { "iso-ir-109", CS_ISO8859_3 },
  67     { "ISO_8859-3", CS_ISO8859_3 },
  68     { "latin3", CS_ISO8859_3 },
  69     { "l3", CS_ISO8859_3 },
  70     { "csISOLatin3", CS_ISO8859_3 },
  71
  72     { "ISO-8859-4", CS_ISO8859_4 },
  73     { "ISO8859-4", CS_ISO8859_4 },     /* WILD */
  74     { "ISO_8859-4:1988", CS_ISO8859_4 },
  75     { "iso-ir-110", CS_ISO8859_4 },
  76     { "ISO_8859-4", CS_ISO8859_4 },
  77     { "latin4", CS_ISO8859_4 },
  78     { "l4", CS_ISO8859_4 },
  79     { "csISOLatin4", CS_ISO8859_4 },
  80
  81     { "ISO-8859-5", CS_ISO8859_5 },
  82     { "ISO8859-5", CS_ISO8859_5 },     /* WILD */
  83     { "ISO_8859-5:1988", CS_ISO8859_5 },
  84     { "iso-ir-144", CS_ISO8859_5 },
  85     { "ISO_8859-5", CS_ISO8859_5 },
  86     { "cyrillic", CS_ISO8859_5 },
  87     { "csISOLatinCyrillic", CS_ISO8859_5 },
  88
  89     { "ISO-8859-6", CS_ISO8859_6 },
  90     { "ISO8859-6", CS_ISO8859_6 },     /* WILD */
  91     { "ISO_8859-6:1987", CS_ISO8859_6 },
  92     { "iso-ir-127", CS_ISO8859_6 },
  93     { "ISO_8859-6", CS_ISO8859_6 },
  94     { "ECMA-114", CS_ISO8859_6 },
  95     { "ASMO-708", CS_ISO8859_6 },
  96     { "arabic", CS_ISO8859_6 },
  97     { "csISOLatinArabic", CS_ISO8859_6 },
  98
  99     { "ISO-8859-7", CS_ISO8859_7 },
 100     { "ISO8859-7", CS_ISO8859_7 },     /* WILD */
 101     { "ISO_8859-7:1987", CS_ISO8859_7 },
 102     { "iso-ir-126", CS_ISO8859_7 },
 103     { "ISO_8859-7", CS_ISO8859_7 },
 104     { "ELOT_928", CS_ISO8859_7 },
 105     { "ECMA-118", CS_ISO8859_7 },
 106     { "greek", CS_ISO8859_7 },
 107     { "greek8", CS_ISO8859_7 },
 108     { "csISOLatinGreek", CS_ISO8859_7 },
 109
 110     { "ISO-8859-8", CS_ISO8859_8 },
 111     { "ISO8859-8", CS_ISO8859_8 },     /* WILD */
 112     { "ISO_8859-8:1988", CS_ISO8859_8 },
 113     { "iso-ir-138", CS_ISO8859_8 },
 114     { "ISO_8859-8", CS_ISO8859_8 },
 115     { "hebrew", CS_ISO8859_8 },
 116     { "csISOLatinHebrew", CS_ISO8859_8 },
 117
 118     { "ISO-8859-9", CS_ISO8859_9 },
 119     { "ISO8859-9", CS_ISO8859_9 },     /* WILD */
 120     { "ISO_8859-9:1989", CS_ISO8859_9 },
 121     { "iso-ir-148", CS_ISO8859_9 },
 122     { "ISO_8859-9", CS_ISO8859_9 },
 123     { "latin5", CS_ISO8859_9 },
 124     { "l5", CS_ISO8859_9 },
 125     { "csISOLatin5", CS_ISO8859_9 },
 126
 127     { "ISO-8859-10", CS_ISO8859_10 },
 128     { "ISO8859-10", CS_ISO8859_10 },   /* WILD */
 129     { "iso-ir-157", CS_ISO8859_10 },
 130     { "l6", CS_ISO8859_10 },
 131     { "ISO_8859-10:1992", CS_ISO8859_10 },
 132     { "csISOLatin6", CS_ISO8859_10 },
 133     { "latin6", CS_ISO8859_10 },
 134
 135     { "ISO-8859-13", CS_ISO8859_13 },
 136     { "ISO8859-13", CS_ISO8859_13 },   /* WILD */
 137
 138     { "ISO-8859-14", CS_ISO8859_14 },
 139     { "ISO8859-14", CS_ISO8859_14 },   /* WILD */
 140     { "iso-ir-199", CS_ISO8859_14 },
 141     { "ISO_8859-14:1998", CS_ISO8859_14 },
 142     { "ISO_8859-14", CS_ISO8859_14 },
 143     { "latin8", CS_ISO8859_14 },
 144     { "iso-celtic", CS_ISO8859_14 },
 145     { "l8", CS_ISO8859_14 },
 146
 147     { "ISO-8859-15", CS_ISO8859_15 },
 148     { "ISO8859-15", CS_ISO8859_15 },   /* WILD */
 149     { "ISO_8859-15", CS_ISO8859_15 },
 150     { "Latin-9", CS_ISO8859_15 },
 151
 152     { "ISO-8859-16", CS_ISO8859_16 },
 153     { "ISO8859-16", CS_ISO8859_16 },   /* WILD */
 154     { "iso-ir-226", CS_ISO8859_16 },
 155     { "ISO_8859-16", CS_ISO8859_16 },
 156     { "ISO_8859-16:2001", CS_ISO8859_16 },
 157     { "latin10", CS_ISO8859_16 },
 158     { "l10", CS_ISO8859_16 },
 159
 160     { "IBM437", CS_CP437 },
 161     { "cp437", CS_CP437 },
 162     { "437", CS_CP437 },
 163     { "csPC8CodePage437", CS_CP437 },
 164
 165     { "IBM850", CS_CP850 },
 166     { "cp850", CS_CP850 },
 167     { "850", CS_CP850 },
 168     { "csPC850Multilingual", CS_CP850 },
 169
 170     { "windows-1250", CS_CP1250 },
 171     { "win-1250", CS_CP1250 },         /* WILD */
 172
 173     { "windows-1251", CS_CP1251 },
 174     { "win-1251", CS_CP1251 },         /* WILD */
 175
 176     { "windows-1252", CS_CP1252 },
 177     { "win-1252", CS_CP1252 },         /* WILD */
 178
 179     { "windows-1253", CS_CP1253 },
 180     { "win-1253", CS_CP1253 },         /* WILD */
 181
 182     { "windows-1254", CS_CP1254 },
 183     { "win-1254", CS_CP1254 },         /* WILD */
 184
 185     { "windows-1255", CS_CP1255 },
 186     { "win-1255", CS_CP1255 },         /* WILD */
 187
 188     { "windows-1256", CS_CP1256 },
 189     { "win-1256", CS_CP1256 },         /* WILD */
 190
 191     { "windows-1257", CS_CP1257 },
 192     { "win-1257", CS_CP1257 },         /* WILD */
 193
 194     { "windows-1258", CS_CP1258 },
 195     { "win-1258", CS_CP1258 },         /* WILD */
 196
 197     { "KOI8-R", CS_KOI8_R },
 198     { "csKOI8R", CS_KOI8_R },
 199
 200     { "KOI8-U", CS_KOI8_U },
 201
 202     { "KOI8-RU", CS_KOI8_RU },         /* WILD */
 203
 204     { "JIS_X0201", CS_JISX0201 },
 205     { "X0201", CS_JISX0201 },
 206     { "csHalfWidthKatakana", CS_JISX0201 },
 207
 208     { "macintosh", CS_MAC_ROMAN_OLD },
 209     { "mac", CS_MAC_ROMAN_OLD },
 210     { "csMacintosh", CS_MAC_ROMAN_OLD },
 211
 212     { "VISCII", CS_VISCII },
 213     { "csVISCII", CS_VISCII },
 214
 215     { "hp-roman8", CS_HP_ROMAN8 },
 216     { "roman8", CS_HP_ROMAN8 },
 217     { "r8", CS_HP_ROMAN8 },
 218     { "csHPRoman8", CS_HP_ROMAN8 },
 219
 220     { "DEC-MCS", CS_DEC_MCS },
 221     { "dec", CS_DEC_MCS },
 222     { "csDECMCS", CS_DEC_MCS },
 223
 224     { "UTF-8", CS_UTF8 },
 225
 226     { "UTF-7", CS_UTF7 },
 227     { "UNICODE-1-1-UTF-7", CS_UTF7 },
 228     { "csUnicode11UTF7", CS_UTF7 },
 229
 230     /*
 231      * Quite why the EUC-CN encoding is known to MIME by the name
 232      * of its underlying character set, I'm not entirely sure, but
 233      * it is. Shrug.
 234      */
 235     { "GB2312", CS_EUC_CN },
 236     { "csGB2312", CS_EUC_CN },
 237
 238     { "EUC-KR", CS_EUC_KR },
 239     { "csEUCKR", CS_EUC_KR },
 240
 241     { "EUC-JP", CS_EUC_JP },
 242     { "csEUCPkdFmtJapanese", CS_EUC_JP },
 243     { "Extended_UNIX_Code_Packed_Format_for_Japanese", CS_EUC_JP },
 244
 245     { "ISO-2022-JP", CS_ISO2022_JP },
 246     { "csISO2022JP", CS_ISO2022_JP },
 247
 248     { "ISO-2022-KR", CS_ISO2022_KR },
 249     { "csISO2022KR", CS_ISO2022_KR },
 250
 251     { "Big5", CS_BIG5 },
 252     { "csBig5", CS_BIG5 },
 253     /* The following two names are not listed on the above web page, but
 254      * I've seen them in the wild in real e-mail. */
 255     { "Big-5", CS_BIG5 },
 256     { "ChineseBig5", CS_BIG5 },
 257
 258     { "Shift_JIS", CS_SHIFT_JIS },
 259     { "MS_Kanji", CS_SHIFT_JIS },
 260     { "csShiftJIS", CS_SHIFT_JIS },
 261
 262     { "HZ-GB-2312", CS_HZ },
 263
 264     { "UTF-16BE", CS_UTF16BE },
 265
 266     { "UTF-16LE", CS_UTF16LE },
 267
 268     { "UTF-16", CS_UTF16 },
 269
 270     /*
 271      * This bit is fiddly and possibly technically incorrect; but
 272      * rumour has it that the KSC 5601 encoding is a subset of
 273      * Microsoft CP949, and that MS products tend to announce CP949
 274      * as KSC 5601 in much the same way they seem willing to
 275      * announce CP1252 as its subset ISO 8859-1. So I cheat
 276      * shamelessly here by letting KSC 5601 map to CP949.
 277      */
 278     { "KS_C_5601-1987", CS_CP949 },
 279     { "iso-ir-149", CS_CP949 },
 280     { "KS_C_5601-1989", CS_CP949 },
 281     { "KSC_5601", CS_CP949 },
 282     { "korean", CS_CP949 },
 283     { "csKSC56011987", CS_CP949 },
 284     { "KSC5601", CS_CP949 },           /* WILD */
 285
 286 #if 0
 287     { "ISO-2022-JP-2", CS_ISO2022_JP_2 },
 288     { "csISO2022JP2", CS_ISO2022_JP_2 },
 289 #endif
 290 };
 291
 292 const char *charset_to_mimeenc(int charset)
 293 {
 294     int i;
 295
 296     for (i = 0; i < (int)lenof(mimeencs); i++)
 297         if (charset == mimeencs[i].charset)
 298             return mimeencs[i].name;
 299
 300     return NULL;                       /* not found */
 301 }
 302
 303 int charset_from_mimeenc(const char *name)
 304 {
 305     int i;
 306
 307     for (i = 0; i < (int)lenof(mimeencs); i++) {
 308         const char *p, *q;
 309         p = name;
 310         q = mimeencs[i].name;
 311         while (*p || *q) {
 312             if (tolower(*p) != tolower(*q))
 313                 break;
 314             p++; q++;
 315         }
 316         if (!*p && !*q)
 317             return mimeencs[i].charset;
 318     }
 319
 320     return CS_NONE;                    /* not found */
 321 }