1 /* ======================================================================== 2 * Copyright 1988-2008 University of Washington 3 * 4 * Licensed under the Apache License, Version 2.0 (the "License"); 5 * you may not use this file except in compliance with the License. 6 * You may obtain a copy of the License at 7 * 8 * http://www.apache.org/licenses/LICENSE-2.0 9 * 10 * 11 * ======================================================================== 12 */ 13 14 /* 15 * Program: UTF-8 routines 16 * 17 * Author: Mark Crispin 18 * Networks and Distributed Computing 19 * Computing & Communications 20 * University of Washington 21 * Administration Building, AG-44 22 * Seattle, WA 98195 23 * Internet: MRC@CAC.Washington.EDU 24 * 25 * Date: 11 June 1997 26 * Last Edited: 17 January 2008 27 */ 28 29 /* UTF-8 size and conversion routines from UCS-2 values (thus in the BMP). 30 * Don't use these if UTF-16 data (surrogate pairs) are an issue. 31 * For UCS-4 values, use the utf8_size() and utf8_put() functions. 32 */ 33 34 #define UTF8_SIZE_BMP(c) ((c & 0xff80) ? ((c & 0xf800) ? 3 : 2) : 1) 35 36 #define UTF8_PUT_BMP(b,c) { \ 37 if (c & 0xff80) { /* non-ASCII? */ \ 38 if (c & 0xf800) { /* three byte code */ \ 39 *b++ = 0xe0 | (c >> 12); \ 40 *b++ = 0x80 | ((c >> 6) & 0x3f); \ 41 } \ 42 else *b++ = 0xc0 | ((c >> 6) & 0x3f); \ 43 *b++ = 0x80 | (c & 0x3f); \ 44 } \ 45 else *b++ = c; \ 46 } 47 48 /* utf8_text() flag values */ 49 50 #define U8T_CASECANON 2 /* canonicalize case */ 51 #define U8T_DECOMPOSE 4 /* decompose */ 52 /* full canonicalization */ 53 #define U8T_CANONICAL (U8T_CASECANON | U8T_DECOMPOSE) 54 55 56 /* utf8_get() return values */ 57 58 /* 0x0000 - 0xffff BMP plane */ 59 #define U8GM_NONBMP 0xffff0000 /* mask for non-BMP values */ 60 /* 0x10000 - 0x10ffff extended planes */ 61 /* 0x110000 - 0x7ffffff non-Unicode */ 62 #define U8G_ERROR 0x80000000 /* error flag */ 63 #define U8G_BADCONT U8G_ERROR+1 /* continuation when not in progress */ 64 #define U8G_INCMPLT U8G_ERROR+2 /* incomplete UTF-8 character */ 65 #define U8G_NOTUTF8 U8G_ERROR+3 /* not a valid UTF-8 octet */ 66 #define U8G_ENDSTRG U8G_ERROR+4 /* end of string */ 67 #define U8G_ENDSTRI U8G_ERROR+5 /* end of string w/ incomplete UTF-8 char */ 68 #define U8G_SURROGA U8G_ERROR+6 /* surrogate codepoint */ 69 #define U8G_NOTUNIC U8G_ERROR+7 /* non-Unicode codepoint */ 70 71 72 /* ucs4_width() return values */ 73 74 #define U4W_ERROR 0x80000000 /* error flags */ 75 #define U4W_NOTUNCD U4W_ERROR+1 /* not a Unicode char */ 76 #define U4W_PRIVATE U4W_ERROR+2 /* private-space plane */ 77 #define U4W_SSPCHAR U4W_ERROR+3 /* Supplementary Special-purpose Plane */ 78 #define U4W_UNASSGN U4W_ERROR+4 /* unassigned space plane */ 79 #define U4W_CONTROL U4W_ERROR+5 /* C0/C1 control */ 80 #define U4W_CTLSRGT U4W_CONTROL /* in case legacy code references this */ 81 82 /* ISO-2022 engine states */ 83 84 #define I2S_CHAR 0 /* character */ 85 #define I2S_ESC 1 /* previous character was ESC */ 86 #define I2S_MUL 2 /* previous character was multi-byte code */ 87 #define I2S_INT 3 /* previous character was intermediate */ 88 89 90 /* ISO-2022 Gn selections */ 91 92 #define I2C_G0 0 /* G0 */ 93 #define I2C_G1 1 /* G1 */ 94 #define I2C_G2 2 /* G2 */ 95 #define I2C_G3 3 /* G3 */ 96 #define I2C_SG2 (2 << 2) /* single shift G2 */ 97 #define I2C_SG3 (3 << 2) /* single shift G2 */ 98 99 100 /* ISO-2022 octet definitions */ 101 102 #define I2C_ESC 0x1b /* ESCape */ 103 104 /* Intermediate character */ 105 #define I2C_STRUCTURE 0x20 /* announce code structure */ 106 #define I2C_C0 0x21 /* C0 */ 107 #define I2C_C1 0x22 /* C1 */ 108 #define I2C_CONTROL 0x23 /* single control function */ 109 #define I2C_MULTI 0x24 /* multi-byte character set */ 110 #define I2C_OTHER 0x25 /* other coding system */ 111 #define I2C_REVISED 0x26 /* revised registration */ 112 #define I2C_G0_94 0x28 /* G0 94-character set */ 113 #define I2C_G1_94 0x29 /* G1 94-character set */ 114 #define I2C_G2_94 0x2A /* G2 94-character set */ 115 #define I2C_G3_94 0x2B /* G3 94-character set */ 116 #define I2C_G0_96 0x2C /* (not in ISO-2022) G0 96-character set */ 117 #define I2C_G1_96 0x2D /* G1 96-character set */ 118 #define I2C_G2_96 0x2E /* G2 96-character set */ 119 #define I2C_G3_96 0x2F /* G3 96-character set */ 120 121 /* Locking shifts */ 122 #define I2C_SI 0x0f /* lock shift to G0 (Shift In) */ 123 #define I2C_SO 0x0e /* lock shift to G1 (Shift Out) */ 124 /* prefixed by ESC */ 125 #define I2C_LS2 0x6e /* lock shift to G2 */ 126 #define I2C_LS3 0x6f /* lock shift to G3 */ 127 #define I2C_LS1R 0x7e /* lock shift GR to G1 */ 128 #define I2C_LS2R 0x7d /* lock shift GR to G2 */ 129 #define I2C_LS3R 0x7c /* lock shift GR to G3 */ 130 131 /* Single shifts */ 132 #define I2C_SS2_ALT 0x8e /* single shift to G2 (SS2) */ 133 #define I2C_SS3_ALT 0x8f /* single shift to G3 (SS3) */ 134 #define I2C_SS2_ALT_7 0x19 /* single shift to G2 (SS2) */ 135 #define I2C_SS3_ALT_7 0x1d /* single shift to G3 (SS3) */ 136 /* prefixed by ESC */ 137 #define I2C_SS2 0x4e /* single shift to G2 (SS2) */ 138 #define I2C_SS3 0x4f /* single shift to G3 (SS3) */ 139 140 /* 94 character sets */ 141 142 /* 4/0 ISO 646 IRV */ 143 #define I2CS_94_BRITISH 0x41 /* 4/1 ISO 646 British */ 144 #define I2CS_94_ASCII 0x42 /* 4/2 ISO 646 USA (ASCII) */ 145 /* 4/3 NATS Finland/Sweden (primary) */ 146 /* 4/4 NATS Finland/Sweden (secondary) */ 147 /* 4/5 NATS Denmark/Norway (primary) */ 148 /* 4/6 NATS Denmark/Norway (secondary) */ 149 /* 4/7 ISO 646 Swedish SEN 850200 */ 150 /* 4/8 ISO 646 Swedish names */ 151 #define I2CS_94_JIS_BUGROM 0x48 /* 4/8 some buggy software does this */ 152 #define I2CS_94_JIS_KANA 0x49 /* 4/9 JIS X 0201-1976 right half */ 153 #define I2CS_94_JIS_ROMAN 0x4a /* 4/a JIS X 0201-1976 left half */ 154 /* 4/b ISO 646 German */ 155 /* 4/c ISO 646 Portuguese (Olivetti) */ 156 /* 4/d ISO 6438 African */ 157 /* 4/e ISO 5427 Cyrillic (Honeywell-Bull) */ 158 /* 4/f DIN 31624 extended bibliography */ 159 /* 5/0 ISO 5426-1980 Bibliography */ 160 /* 5/1 ISO 5427-1981 Cyrillic*/ 161 /* 5/2 ISO 646 French (withdrawn) */ 162 /* 5/3 ISO 5428-1980 Greek bibliography */ 163 /* 5/4 GB 1988-80 Chinese */ 164 /* 5/5 Latin-Greek (Honeywell-Bull) */ 165 /* 5/6 UK Viewdata/Teletext */ 166 /* 5/7 INIS (IRV subset) */ 167 /* 5/8 ISO 5428 Greek Bibliography */ 168 /* 5/9 ISO 646 Italian (Olivetti) */ 169 /* 5/a ISO 646 Spanish (Olivetti) */ 170 /* 5/b Greek (Olivetti) */ 171 /* 5/c Latin-Greek (Olivetti) */ 172 /* 5/d INIS non-standard extension */ 173 /* 5/e INIS Cyrillic extension */ 174 /* 5/f Arabic CODAR-U IERA */ 175 /* 6/0 ISO 646 Norwegian */ 176 /* 6/1 Norwegian version 2 (withdrawn) */ 177 /* 6/2 Videotex supplementary */ 178 /* 6/3 Videotex supplementary #2 */ 179 /* 6/4 Videotex supplementary #3 */ 180 /* 6/5 APL */ 181 /* 6/6 ISO 646 French */ 182 /* 6/7 ISO 646 Portuguese (IBM) */ 183 /* 6/8 ISO 646 Spanish (IBM) */ 184 /* 6/9 ISO 646 Hungarian */ 185 /* 6/a Greek ELOT (withdrawn) */ 186 /* 6/b ISO 9036 Arabic 7-bit */ 187 /* 6/c ISO 646 IRV supplementary set */ 188 /* 6/d JIS C6229-1984 OCR-A */ 189 /* 6/e JIS C6229-1984 OCR-B */ 190 /* 6/f JIS C6229-1984 OCR-B additional */ 191 /* 7/0 JIS C6229-1984 hand-printed */ 192 /* 7/1 JIS C6229-1984 additional hand-printd */ 193 /* 7/2 JIS C6229-1984 katakana hand-printed */ 194 /* 7/3 E13B Japanese graphic */ 195 /* 7/4 Supplementary Videotex (withdrawn) */ 196 /* 7/5 Teletex primary CCITT T.61 */ 197 /* 7/6 Teletex secondary CCITT T.61 */ 198 /* 7/7 CSA Z 243.4-1985 Alternate primary #1 */ 199 /* 7/8 CSA Z 243.4-1985 Alternate primary #2 */ 200 /* 7/9 Mosaic CCITT T.101 */ 201 /* 7/a Serbocroatian/Slovenian Latin */ 202 /* 7/b Serbocroatian Cyrillic */ 203 /* 7/c Supplementary CCITT T.101 */ 204 /* 7/d Macedonian Cyrillic */ 205 206 /* 94 character sets - second intermediate byte */ 207 208 /* 4/0 Greek primary CCITT */ 209 /* 4/1 Cuba */ 210 /* 4/2 ISO/IEC 646 invariant */ 211 /* 4/3 Irish Gaelic 7-bit */ 212 /* 4/4 Turkmen */ 213 214 215 /* 94x94 character sets */ 216 217 #define I2CS_94x94_JIS_OLD 0x40 /* 4/0 JIS X 0208-1978 */ 218 #define I2CS_94x94_GB 0x41 /* 4/1 GB 2312 */ 219 #define I2CS_94x94_JIS_NEW 0x42 /* 4/2 JIS X 0208-1983 */ 220 #define I2CS_94x94_KSC 0x43 /* 4/3 KSC 5601 */ 221 #define I2CS_94x94_JIS_EXT 0x44 /* 4/4 JIS X 0212-1990 */ 222 /* 4/5 CCITT Chinese */ 223 /* 4/6 Blisssymbol Graphic */ 224 #define I2CS_94x94_CNS1 0x47 /* 4/7 CNS 11643 plane 1 */ 225 #define I2CS_94x94_CNS2 0x48 /* 4/8 CNS 11643 plane 2 */ 226 #define I2CS_94x94_CNS3 0x49 /* 4/9 CNS 11643 plane 3 */ 227 #define I2CS_94x94_CNS4 0x4a /* 4/a CNS 11643 plane 4 */ 228 #define I2CS_94x94_CNS5 0x4b /* 4/b CNS 11643 plane 5 */ 229 #define I2CS_94x94_CNS6 0x4c /* 4/c CNS 11643 plane 6 */ 230 #define I2CS_94x94_CNS7 0x4d /* 4/d CNS 11643 plane 7 */ 231 /* 4/e DPRK (North Korea) KGCII */ 232 /* 4/f JGCII plane 1 */ 233 /* 5/0 JGCII plane 2 */ 234 235 /* 96 character sets */ 236 237 #define I2CS_96_ISO8859_1 0x41 /* 4/1 Latin-1 (Western Europe) */ 238 #define I2CS_96_ISO8859_2 0x42 /* 4/2 Latin-2 (Czech, Slovak) */ 239 #define I2CS_96_ISO8859_3 0x43 /* 4/3 Latin-3 (Dutch, Turkish) */ 240 #define I2CS_96_ISO8859_4 0x44 /* 4/4 Latin-4 (Scandinavian) */ 241 /* 4/5 CSA Z 243.4-1985 */ 242 #define I2CS_96_ISO8859_7 0x46 /* 4/6 Greek */ 243 #define I2CS_96_ISO8859_6 0x47 /* 4/7 Arabic */ 244 #define I2CS_96_ISO8859_8 0x48 /* 4/8 Hebrew */ 245 /* 4/9 Czechoslovak CSN 369103 */ 246 /* 4/a Supplementary Latin and non-alpha */ 247 /* 4/b Technical */ 248 #define I2CS_96_ISO8859_5 0x4c /* 4/c Cyrillic */ 249 #define I2CS_96_ISO8859_9 0x4d /* 4/d Latin-5 (Finnish, Portuguese) */ 250 /* 4/e ISO 6937-2 residual */ 251 /* 4/f Basic Cyrillic */ 252 /* 5/0 Supplementary Latin 1, 2 and 5 */ 253 /* 5/1 Basic Box */ 254 /* 5/2 Supplementary ISO/IEC 6937 : 1992 */ 255 /* 5/3 CCITT Hebrew supplementary */ 256 #define I2CS_96_TIS620 0x54 /* 5/4 TIS 620 */ 257 /* 5/5 Arabic/French/German */ 258 #define I2CS_96_ISO8859_10 0x56 /* 5/6 Latin-6 (Northern Europe) */ 259 /* 5/7 ??? */ 260 /* 5/8 Sami (Lappish) supplementary */ 261 #define I2CS_96_ISO8859_13 0x59 /* 5/9 Latin-7 (Baltic) */ 262 #define I2CS_96_VSCII 0x5a /* 5/a Vietnamese */ 263 /* 5/b Technical #1 IEC 1289 */ 264 #define I2CS_96_ISO8859_14 0x5c /* 5/c Latin-8 (Celtic) */ 265 /* 5/d Sami supplementary Latin */ 266 /* 5/e Latin/Hebrew */ 267 /* 5/f Celtic supplementary Latin */ 268 /* 6/0 Uralic supplementary Cyrillic */ 269 /* 6/1 Volgaic supplementary Cyrillic */ 270 #define I2CS_96_ISO8859_15 0x62 /* 6/2 Latin-9 (Euro) */ 271 /* 6/3 Latin-1 with Euro */ 272 /* 6/4 Latin-4 with Euro */ 273 /* 6/5 Latin-7 with Euro */ 274 #define I2CS_96_ISO8859_16 0x66 /* 6/6 Latin-10 (Balkan) */ 275 /* 6/7 Ogham */ 276 /* 6/8 Sami supplementary Latin #2 */ 277 /* 7/d Supplementary Mosaic for CCITT 101 */ 278 279 /* 96x96 character sets */ 280 281 /* Types of character sets */ 282 283 #define I2CS_94 0x000 /* 94 character set */ 284 #define I2CS_96 0x100 /* 96 character set */ 285 #define I2CS_MUL 0x200 /* multi-byte */ 286 #define I2CS_94x94 (I2CS_MUL | I2CS_94) 287 #define I2CS_96x96 (I2CS_MUL | I2CS_96) 288 289 290 /* Character set identifiers stored in Gn */ 291 292 #define I2CS_BRITISH (I2CS_94 | I2CS_94_BRITISH) 293 #define I2CS_ASCII (I2CS_94 | I2CS_94_ASCII) 294 #define I2CS_JIS_BUGROM (I2CS_94 | I2CS_94_JIS_BUGROM) 295 #define I2CS_JIS_KANA (I2CS_94 | I2CS_94_JIS_KANA) 296 #define I2CS_JIS_ROMAN (I2CS_94 | I2CS_94_JIS_ROMAN) 297 #define I2CS_JIS_OLD (I2CS_94x94 | I2CS_94x94_JIS_OLD) 298 #define I2CS_GB (I2CS_94x94 | I2CS_94x94_GB) 299 #define I2CS_JIS_NEW (I2CS_94x94 | I2CS_94x94_JIS_NEW) 300 #define I2CS_KSC (I2CS_94x94 | I2CS_94x94_KSC) 301 #define I2CS_JIS_EXT (I2CS_94x94 | I2CS_94x94_JIS_EXT) 302 #define I2CS_CNS1 (I2CS_94x94 | I2CS_94x94_CNS1) 303 #define I2CS_CNS2 (I2CS_94x94 | I2CS_94x94_CNS2) 304 #define I2CS_CNS3 (I2CS_94x94 | I2CS_94x94_CNS3) 305 #define I2CS_CNS4 (I2CS_94x94 | I2CS_94x94_CNS4) 306 #define I2CS_CNS5 (I2CS_94x94 | I2CS_94x94_CNS5) 307 #define I2CS_CNS6 (I2CS_94x94 | I2CS_94x94_CNS6) 308 #define I2CS_CNS7 (I2CS_94x94 | I2CS_94x94_CNS7) 309 #define I2CS_ISO8859_1 (I2CS_96 | I2CS_96_ISO8859_1) 310 #define I2CS_ISO8859_2 (I2CS_96 | I2CS_96_ISO8859_2) 311 #define I2CS_ISO8859_3 (I2CS_96 | I2CS_96_ISO8859_3) 312 #define I2CS_ISO8859_4 (I2CS_96 | I2CS_96_ISO8859_4) 313 #define I2CS_ISO8859_7 (I2CS_96 | I2CS_96_ISO8859_7) 314 #define I2CS_ISO8859_6 (I2CS_96 | I2CS_96_ISO8859_6) 315 #define I2CS_ISO8859_8 (I2CS_96 | I2CS_96_ISO8859_8) 316 #define I2CS_ISO8859_5 (I2CS_96 | I2CS_96_ISO8859_5) 317 #define I2CS_ISO8859_9 (I2CS_96 | I2CS_96_ISO8859_9) 318 #define I2CS_TIS620 (I2CS_96 | I2CS_96_TIS620) 319 #define I2CS_ISO8859_10 (I2CS_96 | I2CS_96_ISO8859_10) 320 #define I2CS_ISO8859_13 (I2CS_96 | I2CS_96_ISO8859_13) 321 #define I2CS_VSCII (I2CS_96 | I2CS_96_VSCII) 322 #define I2CS_ISO8859_14 (I2CS_96 | I2CS_96_ISO8859_14) 323 #define I2CS_ISO8859_15 (I2CS_96 | I2CS_96_ISO8859_15) 324 #define I2CS_ISO8859_16 (I2CS_96 | I2CS_96_ISO8859_16) 325 326 327 /* Miscellaneous ISO 2022 definitions */ 328 329 #define EUC_CS2 0x8e /* single shift CS2 */ 330 #define EUC_CS3 0x8f /* single shift CS3 */ 331 332 #define BITS7 0x7f /* 7-bit value mask */ 333 #define BIT8 0x80 /* 8th bit mask */ 334 335 /* The following saves us from having to have yet more charset tables */ 336 337 /* Unicode codepoints */ 338 339 #define UCS2_C0CONTROL 0x00 /* first C0 control */ 340 #define UCS2_C0CONTROLEND 0x1F /* last C0 control */ 341 #define UCS2_C1CONTROL 0x80 /* first C1 control */ 342 #define UCS2_C1CONTROLEND 0x9F /* last C1 control */ 343 344 /* ISO 646 substituted Unicode codepoints */ 345 #define UCS2_POUNDSTERLING 0x00a3 346 #define UCS2_YEN 0x00a5 347 #define UCS2_OVERLINE 0x203e 348 #define UCS2_EURO 0x20ac 349 #define UCS2_KATAKANA 0xff61 /* first katakana codepoint */ 350 #define UCS2_BOM 0xfeff /* byte order mark */ 351 #define UCS2_BOGON 0xfffd /* replacement character */ 352 /* next two codepoints are not Unicode chars */ 353 #define UCS2_BOMCHECK 0xfffe /* used to check byte order with UCS2_BOM */ 354 #define UCS2_NOTCHAR 0xffff /* not a character */ 355 356 #define UCS4_BMPBASE 0x0000 /* Basic Multilingual Plane */ 357 #define UCS4_SMPBASE 0x10000 /* Supplementary Multilinugual Plane */ 358 #define UCS4_SIPBASE 0x20000 /* Supplementary Ideographic Plane */ 359 /* EastAsianWidth says plane 3 is wide */ 360 #define UCS4_UNABASE 0x40000 /* unassigned space */ 361 #define UCS4_SSPBASE 0xe0000 /* Supplementary Special-purpose Plane */ 362 #define UCS4_PVTBASE 0xf0000 /* private-space (two planes) */ 363 #define UCS4_MAXUNICODE 0x10ffff/* highest Unicode codepoint */ 364 365 #define UTF16_BASE 0x10000 /* base of codepoints needing surrogates */ 366 #define UTF16_SHIFT 10 /* surrogate shift */ 367 #define UTF16_MASK 0x3ff /* surrogate mask */ 368 #define UTF16_SURR 0xd800 /* UTF-16 surrogate area */ 369 #define UTF16_SURRH 0xd800 /* UTF-16 first high surrogate */ 370 #define UTF16_SURRHEND 0xdbff /* UTF-16 last high surrogate */ 371 #define UTF16_SURRL 0xdc00 /* UTF-16 first low surrogate */ 372 #define UTF16_SURRLEND 0xdfff /* UTF-16 last low surrogate */ 373 #define UTF16_MAXSURR 0xdfff /* end of UTF-16 surrogates */ 374 375 376 /* UBOGON is used to represent a codepoint in a character set which does not 377 * map to Unicode. It is also used for mapping failures, e.g. incomplete 378 * shift sequences. This name has the same text width as 0x????, for 379 * convenience in the mapping tables. 380 * 381 * NOCHAR is used to represent a codepoint in Unicode which does not map to 382 * the target character set in a reverse mapping table. This name has the 383 * same text width as 0x???? in case we ever add static reverse mapping tables. 384 */ 385 386 #define UBOGON UCS2_BOGON 387 #define NOCHAR UCS2_NOTCHAR 388 389 /* Codepoints in non-Unicode character sets */ 390 391 /* Codepoints in ISO 646 character sets */ 392 393 /* British ASCII codepoints */ 394 395 #define BRITISH_POUNDSTERLING 0x23 396 397 /* JIS Roman codepoints */ 398 399 #define JISROMAN_YEN 0x5c 400 #define JISROMAN_OVERLINE 0x7e 401 402 403 /* Hankaku katakana codepoints & parameters 404 * 405 * In earlier versions, MAX_KANA_7 and MAX_KANA_8 were the maximum codepoint 406 * values. Although this made sense, it was confusing with the "max ku" and 407 * "max ten" values used in the double-byte tables; there are 1-origin, but 408 * the calculated values used for "ku" and "ten" are 0-origin (derived by 409 * substracting the "base"). What this all meant is that for double byte 410 * characters the limit test is of the form (value < max_ku), but for single 411 * byte characters (which used the same cell to hold the max ku) the limit 412 * test was (value <= max_ku). 413 * 414 * By making MAX_KANA_[78] be maximum+1, the same (value < max_ku) limit test 415 * is used throughout. - 6/15/2006 416 */ 417 418 #define MIN_KANA_7 0x21 419 #define MAX_KANA_7 0x60 /* maximum value + 1 */ 420 #define KANA_7 (UCS2_KATAKANA - MIN_KANA_7) 421 #define MIN_KANA_8 (MIN_KANA_7 | BIT8) 422 #define MAX_KANA_8 (MAX_KANA_7 | BIT8) 423 #define KANA_8 (UCS2_KATAKANA - MIN_KANA_8) 424 425 /* Charset scripts */ 426 427 /* The term "script" is used here in a very loose sense, enough to make 428 * purists cringe. Basically, the idea is to give the main program some 429 * idea of how it should treat the characters of text in a charset with 430 * respect to font, drawing routines, etc. 431 * 432 * In some cases, "script" is associated with a charset; in other cases, 433 * it's more closely tied to a language. 434 */ 435 436 #define SC_UNICODE 0x1 /* Unicode */ 437 #define SC_LATIN_1 0x10 /* Western Europe */ 438 #define SC_LATIN_2 0x20 /* Eastern Europe */ 439 #define SC_LATIN_3 0x40 /* Southern Europe */ 440 #define SC_LATIN_4 0x80 /* Northern Europe */ 441 #define SC_LATIN_5 0x100 /* Turkish */ 442 #define SC_LATIN_6 0x200 /* Nordic */ 443 #define SC_LATIN_7 0x400 /* Baltic */ 444 #define SC_LATIN_8 0x800 /* Celtic */ 445 #define SC_LATIN_9 0x1000 /* Euro */ 446 #define SC_LATIN_0 SC_LATIN_9 /* colloquial name for Latin-9 */ 447 #define SC_ARABIC 0x2000 448 #define SC_CYRILLIC 0x4000 449 #define SC_GREEK 0x8000 450 #define SC_HEBREW 0x10000 451 #define SC_THAI 0x20000 452 #define SC_UKRANIAN 0x40000 453 #define SC_LATIN_10 0x80000 /* Balkan */ 454 #define SC_VIETNAMESE 0x100000 455 #define SC_CHINESE_SIMPLIFIED 0x1000000 456 #define SC_CHINESE_TRADITIONAL 0x2000000 457 #define SC_JAPANESE 0x4000000 458 #define SC_KOREAN 0x8000000 459 460 461 /* Script table */ 462 463 typedef struct utf8_scent { 464 char *name; /* script name */ 465 char *description; /* script description */ 466 unsigned long script; /* script bitmask */ 467 } SCRIPT; 468 469 /* Character set table support */ 470 471 typedef struct utf8_csent { 472 char *name; /* charset name */ 473 unsigned short type; /* type of charset */ 474 unsigned short flags; /* charset flags */ 475 void *tab; /* additional data */ 476 unsigned long script; /* script(s) implemented by this charset */ 477 char *preferred; /* preferred charset over this one */ 478 } CHARSET; 479 480 481 struct utf8_eucparam { 482 unsigned int base_ku : 8; /* base row */ 483 unsigned int base_ten : 8; /* base column */ 484 unsigned int max_ku : 8; /* maximum row */ 485 unsigned int max_ten : 8; /* maximum column */ 486 void *tab; /* conversion table */ 487 }; 488 489 490 /* Charset types */ 491 492 #define CT_UNKNOWN 0 /* unknown 8-bit */ 493 #define CT_ASCII 1 /* 7-bit ASCII no table */ 494 #define CT_UCS2 2 /* 2 byte 16-bit Unicode no table */ 495 #define CT_UCS4 3 /* 4 byte 32-bit Unicode no table */ 496 #define CT_1BYTE0 10 /* 1 byte ISO 8859-1 no table */ 497 #define CT_1BYTE 11 /* 1 byte ASCII + table 0x80-0xff */ 498 #define CT_1BYTE8 12 /* 1 byte table 0x00 - 0xff */ 499 #define CT_EUC 100 /* 2 byte ASCII + utf8_eucparam base/CS2/CS3 */ 500 #define CT_DBYTE 101 /* 2 byte ASCII + utf8_eucparam */ 501 #define CT_DBYTE2 102 /* 2 byte ASCII + utf8_eucparam plane1/2 */ 502 #define CT_UTF16 1000 /* variable UTF-16 encoded Unicode no table */ 503 #define CT_UTF8 1001 /* variable UTF-8 encoded Unicode no table */ 504 #define CT_UTF7 1002 /* variable UTF-7 encoded Unicode no table */ 505 #define CT_2022 10000 /* variable ISO-2022 encoded no table */ 506 #define CT_SJIS 10001 /* 2 byte Shift-JIS encoded JIS no table */ 507 508 509 /* Character set flags */ 510 511 #define CF_PRIMARY 0x1 /* primary name for this charset */ 512 #define CF_DISPLAY 0x2 /* charset used in displays */ 513 #define CF_POSTING 0x4 /* charset used in email posting */ 514 #define CF_UNSUPRT 0x8 /* charset unsupported (can't convert to it) */ 515 #define CF_NOEMAIL 0x10 /* charset not used in email */ 516 517 518 /* UTF-7 engine states */ 519 520 #define U7_ASCII 0 /* ASCII character */ 521 #define U7_PLUS 1 /* plus seen */ 522 #define U7_UNICODE 2 /* Unicode characters */ 523 #define U7_MINUS 3 /* absorbed minus seen */ 524 525 /* Function prototypes */ 526 527 typedef unsigned long (*ucs4cn_t) (unsigned long c); 528 typedef unsigned long (*ucs4de_t) (unsigned long c,void **more); 529 530 SCRIPT *utf8_script (char *script); 531 const CHARSET *utf8_charset (char *charset); 532 char *utf8_badcharset (char *charset); 533 long utf8_text (SIZEDTEXT *text,char *charset,SIZEDTEXT *ret,long flags); 534 long utf8_text_cs (SIZEDTEXT *text,const CHARSET *cs,SIZEDTEXT *ret, 535 ucs4cn_t cv,ucs4de_t de); 536 long utf8_cstext (SIZEDTEXT *text,char *charset,SIZEDTEXT *ret, 537 unsigned long errch); 538 long utf8_cstocstext (SIZEDTEXT *text,char *sc,SIZEDTEXT *ret,char *dc, 539 unsigned long errch); 540 unsigned short *utf8_rmap (char *charset); 541 unsigned short *utf8_rmap_cs (const CHARSET *cs); 542 unsigned short *utf8_rmap_gen (const CHARSET *cs,unsigned short *oldmap); 543 long utf8_rmaptext (SIZEDTEXT *text,unsigned short *rmap,SIZEDTEXT *ret, 544 unsigned long errch,long iso2022jp); 545 unsigned long utf8_rmapsize (SIZEDTEXT *text,unsigned short *rmap, 546 unsigned long errch,long iso2022jp); 547 long ucs4_rmaptext (unsigned long *ucs4,unsigned long len,unsigned short *rmap, 548 SIZEDTEXT *ret,unsigned long errch); 549 long ucs4_rmaplen (unsigned long *ucs4,unsigned long len,unsigned short *rmap, 550 unsigned long errch); 551 long ucs4_rmapbuf (unsigned char *t,unsigned long *ucs4,unsigned long len, 552 unsigned short *rmap,unsigned long errch); 553 unsigned long utf8_get (unsigned char **s,unsigned long *i); 554 unsigned long utf8_get_raw (unsigned char **s,unsigned long *i); 555 unsigned long ucs4_cs_get (CHARSET *cs,unsigned char **s,unsigned long *i); 556 unsigned long *utf8_csvalidmap (char *charsets[]); 557 const CHARSET *utf8_infercharset (SIZEDTEXT *src); 558 long utf8_validate (unsigned char *s,unsigned long i); 559 void utf8_text_1byte0 (SIZEDTEXT *text,SIZEDTEXT *ret,ucs4cn_t cv,ucs4de_t de); 560 void utf8_text_1byte (SIZEDTEXT *text,SIZEDTEXT *ret,void *tab,ucs4cn_t cv, 561 ucs4de_t de); 562 void utf8_text_1byte8 (SIZEDTEXT *text,SIZEDTEXT *ret,void *tab,ucs4cn_t cv, 563 ucs4de_t de); 564 void utf8_text_euc (SIZEDTEXT *text,SIZEDTEXT *ret,void *tab,ucs4cn_t cv, 565 ucs4de_t de); 566 void utf8_text_dbyte (SIZEDTEXT *text,SIZEDTEXT *ret,void *tab,ucs4cn_t cv, 567 ucs4de_t de); 568 void utf8_text_dbyte2 (SIZEDTEXT *text,SIZEDTEXT *ret,void *tab,ucs4cn_t cv, 569 ucs4de_t de); 570 void utf8_text_sjis (SIZEDTEXT *text,SIZEDTEXT *ret,ucs4cn_t cv,ucs4de_t de); 571 void utf8_text_2022 (SIZEDTEXT *text,SIZEDTEXT *ret,ucs4cn_t cv,ucs4de_t de); 572 void utf8_text_utf7 (SIZEDTEXT *text,SIZEDTEXT *ret,ucs4cn_t cv,ucs4de_t de); 573 void utf8_text_utf8 (SIZEDTEXT *text,SIZEDTEXT *ret,ucs4cn_t cv,ucs4de_t de); 574 void utf8_text_ucs2 (SIZEDTEXT *text,SIZEDTEXT *ret,ucs4cn_t cv,ucs4de_t de); 575 void utf8_text_ucs4 (SIZEDTEXT *text,SIZEDTEXT *ret,ucs4cn_t cv,ucs4de_t de); 576 void utf8_text_utf16 (SIZEDTEXT *text,SIZEDTEXT *ret,ucs4cn_t cv,ucs4de_t de); 577 unsigned long utf8_size (unsigned long c); 578 unsigned char *utf8_put (unsigned char *s,unsigned long c); 579 unsigned long ucs4_titlecase (unsigned long c); 580 long ucs4_width (unsigned long c); 581 long utf8_strwidth (unsigned char *s); 582 long utf8_textwidth (SIZEDTEXT *utf8); 583 unsigned long ucs4_decompose (unsigned long c,void **more); 584 unsigned long ucs4_decompose_recursive (unsigned long c,void **more); 585