1 /* 2 * Copyright (C) 1984-2012 Mark Nudelman 3 * 4 * You may distribute under the terms of either the GNU General Public 5 * License or the Less License, as specified in the README file. 6 * 7 * For more information, see the README file. 8 */ 9 10 11 /* 12 * Functions to define the character set 13 * and do things specific to the character set. 14 */ 15 16 #include "less.h" 17 #if HAVE_LOCALE 18 #include <locale.h> 19 #include <ctype.h> 20 #include <langinfo.h> 21 #endif 22 23 #include "charset.h" 24 25 public int utf_mode = 0; 26 27 /* 28 * Predefined character sets, 29 * selected by the LESSCHARSET environment variable. 30 */ 31 struct charset { 32 char *name; 33 int *p_flag; 34 char *desc; 35 } charsets[] = { 36 { "ascii", NULL, "8bcccbcc18b95.b" }, 37 { "utf-8", &utf_mode, "8bcccbcc18b95.b126.bb" }, 38 { "iso8859", NULL, "8bcccbcc18b95.33b." }, 39 { "latin3", NULL, "8bcccbcc18b95.33b5.b8.b15.b4.b12.b18.b12.b." }, 40 { "arabic", NULL, "8bcccbcc18b95.33b.3b.7b2.13b.3b.b26.5b19.b" }, 41 { "greek", NULL, "8bcccbcc18b95.33b4.2b4.b3.b35.b44.b" }, 42 { "greek2005", NULL, "8bcccbcc18b95.33b14.b35.b44.b" }, 43 { "hebrew", NULL, "8bcccbcc18b95.33b.b29.32b28.2b2.b" }, 44 { "koi8-r", NULL, "8bcccbcc18b95.b." }, 45 { "KOI8-T", NULL, "8bcccbcc18b95.b8.b6.b8.b.b.5b7.3b4.b4.b3.b.b.3b." }, 46 { "georgianps", NULL, "8bcccbcc18b95.3b11.4b12.2b." }, 47 { "tcvn", NULL, "b..b...bcccbccbbb7.8b95.b48.5b." }, 48 { "TIS-620", NULL, "8bcccbcc18b95.b.4b.11b7.8b." }, 49 { "next", NULL, "8bcccbcc18b95.bb125.bb" }, 50 { "dos", NULL, "8bcccbcc12bc5b95.b." }, 51 { "windows-1251", NULL, "8bcccbcc12bc5b95.b24.b." }, 52 { "windows-1252", NULL, "8bcccbcc12bc5b95.b.b11.b.2b12.b." }, 53 { "windows-1255", NULL, "8bcccbcc12bc5b95.b.b8.b.5b9.b.4b." }, 54 { "ebcdic", NULL, "5bc6bcc7bcc41b.9b7.9b5.b..8b6.10b6.b9.7b9.8b8.17b3.3b9.7b9.8b8.6b10.b.b.b." }, 55 { "IBM-1047", NULL, "4cbcbc3b9cbccbccbb4c6bcc5b3cbbc4bc4bccbc191.b" }, 56 { NULL, NULL, NULL } 57 }; 58 59 /* 60 * Support "locale charmap"/nl_langinfo(CODESET) values, as well as others. 61 */ 62 struct cs_alias { 63 char *name; 64 char *oname; 65 } cs_aliases[] = { 66 { "UTF-8", "utf-8" }, 67 { "ANSI_X3.4-1968", "ascii" }, 68 { "US-ASCII", "ascii" }, 69 { "latin1", "iso8859" }, 70 { "ISO-8859-1", "iso8859" }, 71 { "latin9", "iso8859" }, 72 { "ISO-8859-15", "iso8859" }, 73 { "latin2", "iso8859" }, 74 { "ISO-8859-2", "iso8859" }, 75 { "ISO-8859-3", "latin3" }, 76 { "latin4", "iso8859" }, 77 { "ISO-8859-4", "iso8859" }, 78 { "cyrillic", "iso8859" }, 79 { "ISO-8859-5", "iso8859" }, 80 { "ISO-8859-6", "arabic" }, 81 { "ISO-8859-7", "greek" }, 82 { "IBM9005", "greek2005" }, 83 { "ISO-8859-8", "hebrew" }, 84 { "latin5", "iso8859" }, 85 { "ISO-8859-9", "iso8859" }, 86 { "latin6", "iso8859" }, 87 { "ISO-8859-10", "iso8859" }, 88 { "latin7", "iso8859" }, 89 { "ISO-8859-13", "iso8859" }, 90 { "latin8", "iso8859" }, 91 { "ISO-8859-14", "iso8859" }, 92 { "latin10", "iso8859" }, 93 { "ISO-8859-16", "iso8859" }, 94 { "IBM437", "dos" }, 95 { "EBCDIC-US", "ebcdic" }, 96 { "IBM1047", "IBM-1047" }, 97 { "KOI8-R", "koi8-r" }, 98 { "KOI8-U", "koi8-r" }, 99 { "GEORGIAN-PS", "georgianps" }, 100 { "TCVN5712-1", "tcvn" }, 101 { "NEXTSTEP", "next" }, 102 { "windows", "windows-1252" }, /* backward compatibility */ 103 { "CP1251", "windows-1251" }, 104 { "CP1252", "windows-1252" }, 105 { "CP1255", "windows-1255" }, 106 { NULL, NULL } 107 }; 108 109 #define IS_BINARY_CHAR 01 110 #define IS_CONTROL_CHAR 02 111 112 static char chardef[256]; 113 static char *binfmt = NULL; 114 static char *utfbinfmt = NULL; 115 public int binattr = AT_STANDOUT; 116 117 118 /* 119 * Define a charset, given a description string. 120 * The string consists of 256 letters, 121 * one for each character in the charset. 122 * If the string is shorter than 256 letters, missing letters 123 * are taken to be identical to the last one. 124 * A decimal number followed by a letter is taken to be a 125 * repetition of the letter. 126 * 127 * Each letter is one of: 128 * . normal character 129 * b binary character 130 * c control character 131 */ 132 static void 133 ichardef(s) 134 char *s; 135 { 136 register char *cp; 137 register int n; 138 register char v; 139 140 n = 0; 141 v = 0; 142 cp = chardef; 143 while (*s != '\0') 144 { 145 switch (*s++) 146 { 147 case '.': 148 v = 0; 149 break; 150 case 'c': 151 v = IS_CONTROL_CHAR; 152 break; 153 case 'b': 154 v = IS_BINARY_CHAR|IS_CONTROL_CHAR; 155 break; 156 157 case '0': case '1': case '2': case '3': case '4': 158 case '5': case '6': case '7': case '8': case '9': 159 n = (10 * n) + (s[-1] - '0'); 160 continue; 161 162 default: 163 error("invalid chardef", NULL_PARG); 164 quit(QUIT_ERROR); 165 /*NOTREACHED*/ 166 } 167 168 do 169 { 170 if (cp >= chardef + sizeof(chardef)) 171 { 172 error("chardef longer than 256", NULL_PARG); 173 quit(QUIT_ERROR); 174 /*NOTREACHED*/ 175 } 176 *cp++ = v; 177 } while (--n > 0); 178 n = 0; 179 } 180 181 while (cp < chardef + sizeof(chardef)) 182 *cp++ = v; 183 } 184 185 /* 186 * Define a charset, given a charset name. 187 * The valid charset names are listed in the "charsets" array. 188 */ 189 static int 190 icharset(name, no_error) 191 register char *name; 192 int no_error; 193 { 194 register struct charset *p; 195 register struct cs_alias *a; 196 197 if (name == NULL || *name == '\0') 198 return (0); 199 200 /* First see if the name is an alias. */ 201 for (a = cs_aliases; a->name != NULL; a++) 202 { 203 if (strcmp(name, a->name) == 0) 204 { 205 name = a->oname; 206 break; 207 } 208 } 209 210 for (p = charsets; p->name != NULL; p++) 211 { 212 if (strcmp(name, p->name) == 0) 213 { 214 ichardef(p->desc); 215 if (p->p_flag != NULL) 216 *(p->p_flag) = 1; 217 return (1); 218 } 219 } 220 221 if (!no_error) { 222 error("invalid charset name", NULL_PARG); 223 quit(QUIT_ERROR); 224 } 225 return (0); 226 } 227 228 #if HAVE_LOCALE 229 /* 230 * Define a charset, given a locale name. 231 */ 232 static void 233 ilocale() 234 { 235 register int c; 236 237 for (c = 0; c < (int) sizeof(chardef); c++) 238 { 239 if (isprint(c)) 240 chardef[c] = 0; 241 else if (iscntrl(c)) 242 chardef[c] = IS_CONTROL_CHAR; 243 else 244 chardef[c] = IS_BINARY_CHAR|IS_CONTROL_CHAR; 245 } 246 } 247 #endif 248 249 /* 250 * Define the printing format for control (or binary utf) chars. 251 */ 252 static void 253 setbinfmt(s, fmtvarptr, default_fmt) 254 char *s; 255 char **fmtvarptr; 256 char *default_fmt; 257 { 258 if (s && utf_mode) 259 { 260 /* It would be too hard to account for width otherwise. */ 261 char *t = s; 262 while (*t) 263 { 264 if (*t < ' ' || *t > '~') 265 { 266 s = default_fmt; 267 goto attr; 268 } 269 t++; 270 } 271 } 272 273 /* %n is evil */ 274 if (s == NULL || *s == '\0' || 275 (*s == '*' && (s[1] == '\0' || s[2] == '\0' || strchr(s + 2, 'n'))) || 276 (*s != '*' && strchr(s, 'n'))) 277 s = default_fmt; 278 279 /* 280 * Select the attributes if it starts with "*". 281 */ 282 attr: 283 if (*s == '*') 284 { 285 switch (s[1]) 286 { 287 case 'd': binattr = AT_BOLD; break; 288 case 'k': binattr = AT_BLINK; break; 289 case 's': binattr = AT_STANDOUT; break; 290 case 'u': binattr = AT_UNDERLINE; break; 291 default: binattr = AT_NORMAL; break; 292 } 293 s += 2; 294 } 295 *fmtvarptr = s; 296 } 297 298 /* 299 * 300 */ 301 static void 302 set_charset() 303 { 304 char *s; 305 306 /* 307 * See if environment variable LESSCHARSET is defined. 308 */ 309 s = lgetenv("LESSCHARSET"); 310 if (icharset(s, 0)) 311 return; 312 313 /* 314 * LESSCHARSET is not defined: try LESSCHARDEF. 315 */ 316 s = lgetenv("LESSCHARDEF"); 317 if (s != NULL && *s != '\0') 318 { 319 ichardef(s); 320 return; 321 } 322 323 #if HAVE_LOCALE 324 #ifdef CODESET 325 /* 326 * Try using the codeset name as the charset name. 327 */ 328 s = nl_langinfo(CODESET); 329 if (icharset(s, 1)) 330 return; 331 #endif 332 #endif 333 334 #if HAVE_STRSTR 335 /* 336 * Check whether LC_ALL, LC_CTYPE or LANG look like UTF-8 is used. 337 */ 338 if ((s = lgetenv("LC_ALL")) != NULL || 339 (s = lgetenv("LC_CTYPE")) != NULL || 340 (s = lgetenv("LANG")) != NULL) 341 { 342 if ( strstr(s, "UTF-8") != NULL || strstr(s, "utf-8") != NULL 343 || strstr(s, "UTF8") != NULL || strstr(s, "utf8") != NULL) 344 if (icharset("utf-8", 1)) 345 return; 346 } 347 #endif 348 349 #if HAVE_LOCALE 350 /* 351 * Get character definitions from locale functions, 352 * rather than from predefined charset entry. 353 */ 354 ilocale(); 355 #if MSDOS_COMPILER 356 /* 357 * Default to "dos". 358 */ 359 (void) icharset("dos", 1); 360 #else 361 /* 362 * Default to "latin1". 363 */ 364 (void) icharset("latin1", 1); 365 #endif 366 #endif 367 } 368 369 /* 370 * Initialize charset data structures. 371 */ 372 public void 373 init_charset() 374 { 375 char *s; 376 377 #if HAVE_LOCALE 378 setlocale(LC_ALL, ""); 379 #endif 380 381 set_charset(); 382 383 s = lgetenv("LESSBINFMT"); 384 setbinfmt(s, &binfmt, "*s<%02X>"); 385 386 s = lgetenv("LESSUTFBINFMT"); 387 setbinfmt(s, &utfbinfmt, "<U+%04lX>"); 388 } 389 390 /* 391 * Is a given character a "binary" character? 392 */ 393 public int 394 binary_char(c) 395 LWCHAR c; 396 { 397 if (utf_mode) 398 return (is_ubin_char(c)); 399 c &= 0377; 400 return (chardef[c] & IS_BINARY_CHAR); 401 } 402 403 /* 404 * Is a given character a "control" character? 405 */ 406 public int 407 control_char(c) 408 LWCHAR c; 409 { 410 c &= 0377; 411 return (chardef[c] & IS_CONTROL_CHAR); 412 } 413 414 /* 415 * Return the printable form of a character. 416 * For example, in the "ascii" charset '\3' is printed as "^C". 417 */ 418 public char * 419 prchar(c) 420 LWCHAR c; 421 { 422 /* {{ This buffer can be overrun if LESSBINFMT is a long string. }} */ 423 static char buf[32]; 424 425 c &= 0377; 426 if ((c < 128 || !utf_mode) && !control_char(c)) 427 SNPRINTF1(buf, sizeof(buf), "%c", (int) c); 428 else if (c == ESC) 429 strcpy(buf, "ESC"); 430 #if IS_EBCDIC_HOST 431 else if (!binary_char(c) && c < 64) 432 SNPRINTF1(buf, sizeof(buf), "^%c", 433 /* 434 * This array roughly inverts CONTROL() #defined in less.h, 435 * and should be kept in sync with CONTROL() and IBM-1047. 436 */ 437 "@ABC.I.?...KLMNO" 438 "PQRS.JH.XY.." 439 "\\]^_" 440 "......W[.....EFG" 441 "..V....D....TU.Z"[c]); 442 #else 443 else if (c < 128 && !control_char(c ^ 0100)) 444 SNPRINTF1(buf, sizeof(buf), "^%c", (int) (c ^ 0100)); 445 #endif 446 else 447 SNPRINTF1(buf, sizeof(buf), binfmt, c); 448 return (buf); 449 } 450 451 /* 452 * Return the printable form of a UTF-8 character. 453 */ 454 public char * 455 prutfchar(ch) 456 LWCHAR ch; 457 { 458 static char buf[32]; 459 460 if (ch == ESC) 461 strcpy(buf, "ESC"); 462 else if (ch < 128 && control_char(ch)) 463 { 464 if (!control_char(ch ^ 0100)) 465 SNPRINTF1(buf, sizeof(buf), "^%c", ((char) ch) ^ 0100); 466 else 467 SNPRINTF1(buf, sizeof(buf), binfmt, (char) ch); 468 } else if (is_ubin_char(ch)) 469 SNPRINTF1(buf, sizeof(buf), utfbinfmt, ch); 470 else 471 { 472 int len; 473 if (ch >= 0x80000000) 474 { 475 len = 3; 476 ch = 0xFFFD; 477 } else 478 { 479 len = (ch < 0x80) ? 1 480 : (ch < 0x800) ? 2 481 : (ch < 0x10000) ? 3 482 : (ch < 0x200000) ? 4 483 : (ch < 0x4000000) ? 5 484 : 6; 485 } 486 buf[len] = '\0'; 487 if (len == 1) 488 *buf = (char) ch; 489 else 490 { 491 *buf = ((1 << len) - 1) << (8 - len); 492 while (--len > 0) 493 { 494 buf[len] = (char) (0x80 | (ch & 0x3F)); 495 ch >>= 6; 496 } 497 *buf |= ch; 498 } 499 } 500 return (buf); 501 } 502 503 /* 504 * Get the length of a UTF-8 character in bytes. 505 */ 506 public int 507 utf_len(ch) 508 char ch; 509 { 510 if ((ch & 0x80) == 0) 511 return 1; 512 if ((ch & 0xE0) == 0xC0) 513 return 2; 514 if ((ch & 0xF0) == 0xE0) 515 return 3; 516 if ((ch & 0xF8) == 0xF0) 517 return 4; 518 if ((ch & 0xFC) == 0xF8) 519 return 5; 520 if ((ch & 0xFE) == 0xFC) 521 return 6; 522 /* Invalid UTF-8 encoding. */ 523 return 1; 524 } 525 526 /* 527 * Is a UTF-8 character well-formed? 528 */ 529 public int 530 is_utf8_well_formed(s) 531 unsigned char *s; 532 { 533 int i; 534 int len; 535 536 if (IS_UTF8_INVALID(s[0])) 537 return (0); 538 539 len = utf_len((char) s[0]); 540 if (len == 1) 541 return (1); 542 if (len == 2) 543 { 544 if (s[0] < 0xC2) 545 return (0); 546 } else 547 { 548 unsigned char mask; 549 mask = (~((1 << (8-len)) - 1)) & 0xFF; 550 if (s[0] == mask && (s[1] & mask) == 0x80) 551 return (0); 552 } 553 554 for (i = 1; i < len; i++) 555 if (!IS_UTF8_TRAIL(s[i])) 556 return (0); 557 return (1); 558 } 559 560 /* 561 * Get the value of a UTF-8 character. 562 */ 563 public LWCHAR 564 get_wchar(p) 565 char *p; 566 { 567 switch (utf_len(p[0])) 568 { 569 case 1: 570 default: 571 /* 0xxxxxxx */ 572 return (LWCHAR) 573 (p[0] & 0xFF); 574 case 2: 575 /* 110xxxxx 10xxxxxx */ 576 return (LWCHAR) ( 577 ((p[0] & 0x1F) << 6) | 578 (p[1] & 0x3F)); 579 case 3: 580 /* 1110xxxx 10xxxxxx 10xxxxxx */ 581 return (LWCHAR) ( 582 ((p[0] & 0x0F) << 12) | 583 ((p[1] & 0x3F) << 6) | 584 (p[2] & 0x3F)); 585 case 4: 586 /* 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */ 587 return (LWCHAR) ( 588 ((p[0] & 0x07) << 18) | 589 ((p[1] & 0x3F) << 12) | 590 ((p[2] & 0x3F) << 6) | 591 (p[3] & 0x3F)); 592 case 5: 593 /* 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx */ 594 return (LWCHAR) ( 595 ((p[0] & 0x03) << 24) | 596 ((p[1] & 0x3F) << 18) | 597 ((p[2] & 0x3F) << 12) | 598 ((p[3] & 0x3F) << 6) | 599 (p[4] & 0x3F)); 600 case 6: 601 /* 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx */ 602 return (LWCHAR) ( 603 ((p[0] & 0x01) << 30) | 604 ((p[1] & 0x3F) << 24) | 605 ((p[2] & 0x3F) << 18) | 606 ((p[3] & 0x3F) << 12) | 607 ((p[4] & 0x3F) << 6) | 608 (p[5] & 0x3F)); 609 } 610 } 611 612 /* 613 * Store a character into a UTF-8 string. 614 */ 615 public void 616 put_wchar(pp, ch) 617 char **pp; 618 LWCHAR ch; 619 { 620 if (!utf_mode || ch < 0x80) 621 { 622 /* 0xxxxxxx */ 623 *(*pp)++ = (char) ch; 624 } else if (ch < 0x800) 625 { 626 /* 110xxxxx 10xxxxxx */ 627 *(*pp)++ = (char) (0xC0 | ((ch >> 6) & 0x1F)); 628 *(*pp)++ = (char) (0x80 | (ch & 0x3F)); 629 } else if (ch < 0x10000) 630 { 631 /* 1110xxxx 10xxxxxx 10xxxxxx */ 632 *(*pp)++ = (char) (0xE0 | ((ch >> 12) & 0x0F)); 633 *(*pp)++ = (char) (0x80 | ((ch >> 6) & 0x3F)); 634 *(*pp)++ = (char) (0x80 | (ch & 0x3F)); 635 } else if (ch < 0x200000) 636 { 637 /* 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */ 638 *(*pp)++ = (char) (0xF0 | ((ch >> 18) & 0x07)); 639 *(*pp)++ = (char) (0x80 | ((ch >> 12) & 0x3F)); 640 *(*pp)++ = (char) (0x80 | ((ch >> 6) & 0x3F)); 641 *(*pp)++ = (char) (0x80 | (ch & 0x3F)); 642 } else if (ch < 0x4000000) 643 { 644 /* 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx */ 645 *(*pp)++ = (char) (0xF0 | ((ch >> 24) & 0x03)); 646 *(*pp)++ = (char) (0x80 | ((ch >> 18) & 0x3F)); 647 *(*pp)++ = (char) (0x80 | ((ch >> 12) & 0x3F)); 648 *(*pp)++ = (char) (0x80 | ((ch >> 6) & 0x3F)); 649 *(*pp)++ = (char) (0x80 | (ch & 0x3F)); 650 } else 651 { 652 /* 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx */ 653 *(*pp)++ = (char) (0xF0 | ((ch >> 30) & 0x01)); 654 *(*pp)++ = (char) (0x80 | ((ch >> 24) & 0x3F)); 655 *(*pp)++ = (char) (0x80 | ((ch >> 18) & 0x3F)); 656 *(*pp)++ = (char) (0x80 | ((ch >> 12) & 0x3F)); 657 *(*pp)++ = (char) (0x80 | ((ch >> 6) & 0x3F)); 658 *(*pp)++ = (char) (0x80 | (ch & 0x3F)); 659 } 660 } 661 662 /* 663 * Step forward or backward one character in a string. 664 */ 665 public LWCHAR 666 step_char(pp, dir, limit) 667 char **pp; 668 signed int dir; 669 char *limit; 670 { 671 LWCHAR ch; 672 int len; 673 char *p = *pp; 674 675 if (!utf_mode) 676 { 677 /* It's easy if chars are one byte. */ 678 if (dir > 0) 679 ch = (LWCHAR) ((p < limit) ? *p++ : 0); 680 else 681 ch = (LWCHAR) ((p > limit) ? *--p : 0); 682 } else if (dir > 0) 683 { 684 len = utf_len(*p); 685 if (p + len > limit) 686 { 687 ch = 0; 688 p = limit; 689 } else 690 { 691 ch = get_wchar(p); 692 p += len; 693 } 694 } else 695 { 696 while (p > limit && IS_UTF8_TRAIL(p[-1])) 697 p--; 698 if (p > limit) 699 ch = get_wchar(--p); 700 else 701 ch = 0; 702 } 703 *pp = p; 704 return ch; 705 } 706 707 /* 708 * Unicode characters data 709 */ 710 struct wchar_range { LWCHAR first, last; }; 711 712 /* 713 * Characters with general category values 714 * Mn: Mark, Nonspacing 715 * Me: Mark, Enclosing 716 * Last synched with 717 * <http://www.unicode.org/Public/5.0.0/ucd/UnicodeData-5.0.0d7.txt> 718 * dated 2005-11-30T00:58:48Z 719 */ 720 static struct wchar_range comp_table[] = { 721 { 0x0300, 0x036F} /* Mn */, { 0x0483, 0x0486} /* Mn */, 722 { 0x0488, 0x0489} /* Me */, 723 { 0x0591, 0x05BD} /* Mn */, { 0x05BF, 0x05BF} /* Mn */, 724 { 0x05C1, 0x05C2} /* Mn */, { 0x05C4, 0x05C5} /* Mn */, 725 { 0x05C7, 0x05C7} /* Mn */, { 0x0610, 0x0615} /* Mn */, 726 { 0x064B, 0x065E} /* Mn */, { 0x0670, 0x0670} /* Mn */, 727 { 0x06D6, 0x06DC} /* Mn */, 728 { 0x06DE, 0x06DE} /* Me */, 729 { 0x06DF, 0x06E4} /* Mn */, { 0x06E7, 0x06E8} /* Mn */, 730 { 0x06EA, 0x06ED} /* Mn */, { 0x0711, 0x0711} /* Mn */, 731 { 0x0730, 0x074A} /* Mn */, { 0x07A6, 0x07B0} /* Mn */, 732 { 0x07EB, 0x07F3} /* Mn */, { 0x0901, 0x0902} /* Mn */, 733 { 0x093C, 0x093C} /* Mn */, { 0x0941, 0x0948} /* Mn */, 734 { 0x094D, 0x094D} /* Mn */, { 0x0951, 0x0954} /* Mn */, 735 { 0x0962, 0x0963} /* Mn */, { 0x0981, 0x0981} /* Mn */, 736 { 0x09BC, 0x09BC} /* Mn */, { 0x09C1, 0x09C4} /* Mn */, 737 { 0x09CD, 0x09CD} /* Mn */, { 0x09E2, 0x09E3} /* Mn */, 738 { 0x0A01, 0x0A02} /* Mn */, { 0x0A3C, 0x0A3C} /* Mn */, 739 { 0x0A41, 0x0A42} /* Mn */, { 0x0A47, 0x0A48} /* Mn */, 740 { 0x0A4B, 0x0A4D} /* Mn */, { 0x0A70, 0x0A71} /* Mn */, 741 { 0x0A81, 0x0A82} /* Mn */, { 0x0ABC, 0x0ABC} /* Mn */, 742 { 0x0AC1, 0x0AC5} /* Mn */, { 0x0AC7, 0x0AC8} /* Mn */, 743 { 0x0ACD, 0x0ACD} /* Mn */, { 0x0AE2, 0x0AE3} /* Mn */, 744 { 0x0B01, 0x0B01} /* Mn */, { 0x0B3C, 0x0B3C} /* Mn */, 745 { 0x0B3F, 0x0B3F} /* Mn */, { 0x0B41, 0x0B43} /* Mn */, 746 { 0x0B4D, 0x0B4D} /* Mn */, { 0x0B56, 0x0B56} /* Mn */, 747 { 0x0B82, 0x0B82} /* Mn */, { 0x0BC0, 0x0BC0} /* Mn */, 748 { 0x0BCD, 0x0BCD} /* Mn */, { 0x0C3E, 0x0C40} /* Mn */, 749 { 0x0C46, 0x0C48} /* Mn */, { 0x0C4A, 0x0C4D} /* Mn */, 750 { 0x0C55, 0x0C56} /* Mn */, { 0x0CBC, 0x0CBC} /* Mn */, 751 { 0x0CBF, 0x0CBF} /* Mn */, { 0x0CC6, 0x0CC6} /* Mn */, 752 { 0x0CCC, 0x0CCD} /* Mn */, { 0x0CE2, 0x0CE3} /* Mn */, 753 { 0x0D41, 0x0D43} /* Mn */, { 0x0D4D, 0x0D4D} /* Mn */, 754 { 0x0DCA, 0x0DCA} /* Mn */, { 0x0DD2, 0x0DD4} /* Mn */, 755 { 0x0DD6, 0x0DD6} /* Mn */, { 0x0E31, 0x0E31} /* Mn */, 756 { 0x0E34, 0x0E3A} /* Mn */, { 0x0E47, 0x0E4E} /* Mn */, 757 { 0x0EB1, 0x0EB1} /* Mn */, { 0x0EB4, 0x0EB9} /* Mn */, 758 { 0x0EBB, 0x0EBC} /* Mn */, { 0x0EC8, 0x0ECD} /* Mn */, 759 { 0x0F18, 0x0F19} /* Mn */, { 0x0F35, 0x0F35} /* Mn */, 760 { 0x0F37, 0x0F37} /* Mn */, { 0x0F39, 0x0F39} /* Mn */, 761 { 0x0F71, 0x0F7E} /* Mn */, { 0x0F80, 0x0F84} /* Mn */, 762 { 0x0F86, 0x0F87} /* Mn */, { 0x0F90, 0x0F97} /* Mn */, 763 { 0x0F99, 0x0FBC} /* Mn */, { 0x0FC6, 0x0FC6} /* Mn */, 764 { 0x102D, 0x1030} /* Mn */, { 0x1032, 0x1032} /* Mn */, 765 { 0x1036, 0x1037} /* Mn */, { 0x1039, 0x1039} /* Mn */, 766 { 0x1058, 0x1059} /* Mn */, { 0x135F, 0x135F} /* Mn */, 767 { 0x1712, 0x1714} /* Mn */, { 0x1732, 0x1734} /* Mn */, 768 { 0x1752, 0x1753} /* Mn */, { 0x1772, 0x1773} /* Mn */, 769 { 0x17B7, 0x17BD} /* Mn */, { 0x17C6, 0x17C6} /* Mn */, 770 { 0x17C9, 0x17D3} /* Mn */, { 0x17DD, 0x17DD} /* Mn */, 771 { 0x180B, 0x180D} /* Mn */, { 0x18A9, 0x18A9} /* Mn */, 772 { 0x1920, 0x1922} /* Mn */, { 0x1927, 0x1928} /* Mn */, 773 { 0x1932, 0x1932} /* Mn */, { 0x1939, 0x193B} /* Mn */, 774 { 0x1A17, 0x1A18} /* Mn */, { 0x1B00, 0x1B03} /* Mn */, 775 { 0x1B34, 0x1B34} /* Mn */, { 0x1B36, 0x1B3A} /* Mn */, 776 { 0x1B3C, 0x1B3C} /* Mn */, { 0x1B42, 0x1B42} /* Mn */, 777 { 0x1B6B, 0x1B73} /* Mn */, { 0x1DC0, 0x1DCA} /* Mn */, 778 { 0x1DFE, 0x1DFF} /* Mn */, { 0x20D0, 0x20DC} /* Mn */, 779 { 0x20DD, 0x20E0} /* Me */, 780 { 0x20E1, 0x20E1} /* Mn */, 781 { 0x20E2, 0x20E4} /* Me */, 782 { 0x20E5, 0x20EF} /* Mn */, { 0x302A, 0x302F} /* Mn */, 783 { 0x3099, 0x309A} /* Mn */, { 0xA806, 0xA806} /* Mn */, 784 { 0xA80B, 0xA80B} /* Mn */, { 0xA825, 0xA826} /* Mn */, 785 { 0xFB1E, 0xFB1E} /* Mn */, { 0xFE00, 0xFE0F} /* Mn */, 786 { 0xFE20, 0xFE23} /* Mn */, { 0x10A01, 0x10A03} /* Mn */, 787 { 0x10A05, 0x10A06} /* Mn */, { 0x10A0C, 0x10A0F} /* Mn */, 788 { 0x10A38, 0x10A3A} /* Mn */, { 0x10A3F, 0x10A3F} /* Mn */, 789 { 0x1D167, 0x1D169} /* Mn */, { 0x1D17B, 0x1D182} /* Mn */, 790 { 0x1D185, 0x1D18B} /* Mn */, { 0x1D1AA, 0x1D1AD} /* Mn */, 791 { 0x1D242, 0x1D244} /* Mn */, { 0xE0100, 0xE01EF} /* Mn */, 792 }; 793 794 /* 795 * Special pairs, not ranges. 796 */ 797 static struct wchar_range comb_table[] = { 798 {0x0644,0x0622}, {0x0644,0x0623}, {0x0644,0x0625}, {0x0644,0x0627}, 799 }; 800 801 /* 802 * Characters with general category values 803 * Cc: Other, Control 804 * Cf: Other, Format 805 * Cs: Other, Surrogate 806 * Co: Other, Private Use 807 * Cn: Other, Not Assigned 808 * Zl: Separator, Line 809 * Zp: Separator, Paragraph 810 * Last synched with 811 * <http://www.unicode.org/Public/5.0.0/ucd/UnicodeData-5.0.0d7.txt> 812 * dated 2005-11-30T00:58:48Z 813 */ 814 static struct wchar_range ubin_table[] = { 815 { 0x0000, 0x0007} /* Cc */, 816 { 0x000B, 0x000C} /* Cc */, 817 { 0x000E, 0x001A} /* Cc */, 818 { 0x001C, 0x001F} /* Cc */, 819 { 0x007F, 0x009F} /* Cc */, 820 #if 0 821 { 0x00AD, 0x00AD} /* Cf */, 822 #endif 823 { 0x0370, 0x0373} /* Cn */, { 0x0376, 0x0379} /* Cn */, 824 { 0x037F, 0x0383} /* Cn */, { 0x038B, 0x038B} /* Cn */, 825 { 0x038D, 0x038D} /* Cn */, { 0x03A2, 0x03A2} /* Cn */, 826 { 0x03CF, 0x03CF} /* Cn */, { 0x0487, 0x0487} /* Cn */, 827 { 0x0514, 0x0530} /* Cn */, { 0x0557, 0x0558} /* Cn */, 828 { 0x0560, 0x0560} /* Cn */, { 0x0588, 0x0588} /* Cn */, 829 { 0x058B, 0x0590} /* Cn */, { 0x05C8, 0x05CF} /* Cn */, 830 { 0x05EB, 0x05EF} /* Cn */, { 0x05F5, 0x05FF} /* Cn */, 831 #if 0 832 { 0x0600, 0x0603} /* Cf */, 833 #endif 834 { 0x0604, 0x060A} /* Cn */, { 0x0616, 0x061A} /* Cn */, 835 { 0x061C, 0x061D} /* Cn */, { 0x0620, 0x0620} /* Cn */, 836 { 0x063B, 0x063F} /* Cn */, { 0x065F, 0x065F} /* Cn */, 837 #if 0 838 { 0x06DD, 0x06DD} /* Cf */, 839 #endif 840 { 0x070E, 0x070E} /* Cn */, 841 #if 0 842 { 0x070F, 0x070F} /* Cf */, 843 #endif 844 { 0x074B, 0x074C} /* Cn */, { 0x076E, 0x077F} /* Cn */, 845 { 0x07B2, 0x07BF} /* Cn */, { 0x07FB, 0x0900} /* Cn */, 846 { 0x093A, 0x093B} /* Cn */, { 0x094E, 0x094F} /* Cn */, 847 { 0x0955, 0x0957} /* Cn */, { 0x0971, 0x097A} /* Cn */, 848 { 0x0980, 0x0980} /* Cn */, { 0x0984, 0x0984} /* Cn */, 849 { 0x098D, 0x098E} /* Cn */, { 0x0991, 0x0992} /* Cn */, 850 { 0x09A9, 0x09A9} /* Cn */, { 0x09B1, 0x09B1} /* Cn */, 851 { 0x09B3, 0x09B5} /* Cn */, { 0x09BA, 0x09BB} /* Cn */, 852 { 0x09C5, 0x09C6} /* Cn */, { 0x09C9, 0x09CA} /* Cn */, 853 { 0x09CF, 0x09D6} /* Cn */, { 0x09D8, 0x09DB} /* Cn */, 854 { 0x09DE, 0x09DE} /* Cn */, { 0x09E4, 0x09E5} /* Cn */, 855 { 0x09FB, 0x0A00} /* Cn */, { 0x0A04, 0x0A04} /* Cn */, 856 { 0x0A0B, 0x0A0E} /* Cn */, { 0x0A11, 0x0A12} /* Cn */, 857 { 0x0A29, 0x0A29} /* Cn */, { 0x0A31, 0x0A31} /* Cn */, 858 { 0x0A34, 0x0A34} /* Cn */, { 0x0A37, 0x0A37} /* Cn */, 859 { 0x0A3A, 0x0A3B} /* Cn */, { 0x0A3D, 0x0A3D} /* Cn */, 860 { 0x0A43, 0x0A46} /* Cn */, { 0x0A49, 0x0A4A} /* Cn */, 861 { 0x0A4E, 0x0A58} /* Cn */, { 0x0A5D, 0x0A5D} /* Cn */, 862 { 0x0A5F, 0x0A65} /* Cn */, { 0x0A75, 0x0A80} /* Cn */, 863 { 0x0A84, 0x0A84} /* Cn */, { 0x0A8E, 0x0A8E} /* Cn */, 864 { 0x0A92, 0x0A92} /* Cn */, { 0x0AA9, 0x0AA9} /* Cn */, 865 { 0x0AB1, 0x0AB1} /* Cn */, { 0x0AB4, 0x0AB4} /* Cn */, 866 { 0x0ABA, 0x0ABB} /* Cn */, { 0x0AC6, 0x0AC6} /* Cn */, 867 { 0x0ACA, 0x0ACA} /* Cn */, { 0x0ACE, 0x0ACF} /* Cn */, 868 { 0x0AD1, 0x0ADF} /* Cn */, { 0x0AE4, 0x0AE5} /* Cn */, 869 { 0x0AF0, 0x0AF0} /* Cn */, { 0x0AF2, 0x0B00} /* Cn */, 870 { 0x0B04, 0x0B04} /* Cn */, { 0x0B0D, 0x0B0E} /* Cn */, 871 { 0x0B11, 0x0B12} /* Cn */, { 0x0B29, 0x0B29} /* Cn */, 872 { 0x0B31, 0x0B31} /* Cn */, { 0x0B34, 0x0B34} /* Cn */, 873 { 0x0B3A, 0x0B3B} /* Cn */, { 0x0B44, 0x0B46} /* Cn */, 874 { 0x0B49, 0x0B4A} /* Cn */, { 0x0B4E, 0x0B55} /* Cn */, 875 { 0x0B58, 0x0B5B} /* Cn */, { 0x0B5E, 0x0B5E} /* Cn */, 876 { 0x0B62, 0x0B65} /* Cn */, { 0x0B72, 0x0B81} /* Cn */, 877 { 0x0B84, 0x0B84} /* Cn */, { 0x0B8B, 0x0B8D} /* Cn */, 878 { 0x0B91, 0x0B91} /* Cn */, { 0x0B96, 0x0B98} /* Cn */, 879 { 0x0B9B, 0x0B9B} /* Cn */, { 0x0B9D, 0x0B9D} /* Cn */, 880 { 0x0BA0, 0x0BA2} /* Cn */, { 0x0BA5, 0x0BA7} /* Cn */, 881 { 0x0BAB, 0x0BAD} /* Cn */, { 0x0BBA, 0x0BBD} /* Cn */, 882 { 0x0BC3, 0x0BC5} /* Cn */, { 0x0BC9, 0x0BC9} /* Cn */, 883 { 0x0BCE, 0x0BD6} /* Cn */, { 0x0BD8, 0x0BE5} /* Cn */, 884 { 0x0BFB, 0x0C00} /* Cn */, { 0x0C04, 0x0C04} /* Cn */, 885 { 0x0C0D, 0x0C0D} /* Cn */, { 0x0C11, 0x0C11} /* Cn */, 886 { 0x0C29, 0x0C29} /* Cn */, { 0x0C34, 0x0C34} /* Cn */, 887 { 0x0C3A, 0x0C3D} /* Cn */, { 0x0C45, 0x0C45} /* Cn */, 888 { 0x0C49, 0x0C49} /* Cn */, { 0x0C4E, 0x0C54} /* Cn */, 889 { 0x0C57, 0x0C5F} /* Cn */, { 0x0C62, 0x0C65} /* Cn */, 890 { 0x0C70, 0x0C81} /* Cn */, { 0x0C84, 0x0C84} /* Cn */, 891 { 0x0C8D, 0x0C8D} /* Cn */, { 0x0C91, 0x0C91} /* Cn */, 892 { 0x0CA9, 0x0CA9} /* Cn */, { 0x0CB4, 0x0CB4} /* Cn */, 893 { 0x0CBA, 0x0CBB} /* Cn */, { 0x0CC5, 0x0CC5} /* Cn */, 894 { 0x0CC9, 0x0CC9} /* Cn */, { 0x0CCE, 0x0CD4} /* Cn */, 895 { 0x0CD7, 0x0CDD} /* Cn */, { 0x0CDF, 0x0CDF} /* Cn */, 896 { 0x0CE4, 0x0CE5} /* Cn */, { 0x0CF0, 0x0CF0} /* Cn */, 897 { 0x0CF3, 0x0D01} /* Cn */, { 0x0D04, 0x0D04} /* Cn */, 898 { 0x0D0D, 0x0D0D} /* Cn */, { 0x0D11, 0x0D11} /* Cn */, 899 { 0x0D29, 0x0D29} /* Cn */, { 0x0D3A, 0x0D3D} /* Cn */, 900 { 0x0D44, 0x0D45} /* Cn */, { 0x0D49, 0x0D49} /* Cn */, 901 { 0x0D4E, 0x0D56} /* Cn */, { 0x0D58, 0x0D5F} /* Cn */, 902 { 0x0D62, 0x0D65} /* Cn */, { 0x0D70, 0x0D81} /* Cn */, 903 { 0x0D84, 0x0D84} /* Cn */, { 0x0D97, 0x0D99} /* Cn */, 904 { 0x0DB2, 0x0DB2} /* Cn */, { 0x0DBC, 0x0DBC} /* Cn */, 905 { 0x0DBE, 0x0DBF} /* Cn */, { 0x0DC7, 0x0DC9} /* Cn */, 906 { 0x0DCB, 0x0DCE} /* Cn */, { 0x0DD5, 0x0DD5} /* Cn */, 907 { 0x0DD7, 0x0DD7} /* Cn */, { 0x0DE0, 0x0DF1} /* Cn */, 908 { 0x0DF5, 0x0E00} /* Cn */, { 0x0E3B, 0x0E3E} /* Cn */, 909 { 0x0E5C, 0x0E80} /* Cn */, { 0x0E83, 0x0E83} /* Cn */, 910 { 0x0E85, 0x0E86} /* Cn */, { 0x0E89, 0x0E89} /* Cn */, 911 { 0x0E8B, 0x0E8C} /* Cn */, { 0x0E8E, 0x0E93} /* Cn */, 912 { 0x0E98, 0x0E98} /* Cn */, { 0x0EA0, 0x0EA0} /* Cn */, 913 { 0x0EA4, 0x0EA4} /* Cn */, { 0x0EA6, 0x0EA6} /* Cn */, 914 { 0x0EA8, 0x0EA9} /* Cn */, { 0x0EAC, 0x0EAC} /* Cn */, 915 { 0x0EBA, 0x0EBA} /* Cn */, { 0x0EBE, 0x0EBF} /* Cn */, 916 { 0x0EC5, 0x0EC5} /* Cn */, { 0x0EC7, 0x0EC7} /* Cn */, 917 { 0x0ECE, 0x0ECF} /* Cn */, { 0x0EDA, 0x0EDB} /* Cn */, 918 { 0x0EDE, 0x0EFF} /* Cn */, { 0x0F48, 0x0F48} /* Cn */, 919 { 0x0F6B, 0x0F70} /* Cn */, { 0x0F8C, 0x0F8F} /* Cn */, 920 { 0x0F98, 0x0F98} /* Cn */, { 0x0FBD, 0x0FBD} /* Cn */, 921 { 0x0FCD, 0x0FCE} /* Cn */, { 0x0FD2, 0x0FFF} /* Cn */, 922 { 0x1022, 0x1022} /* Cn */, { 0x1028, 0x1028} /* Cn */, 923 { 0x102B, 0x102B} /* Cn */, { 0x1033, 0x1035} /* Cn */, 924 { 0x103A, 0x103F} /* Cn */, { 0x105A, 0x109F} /* Cn */, 925 { 0x10C6, 0x10CF} /* Cn */, { 0x10FD, 0x10FF} /* Cn */, 926 { 0x115A, 0x115E} /* Cn */, { 0x11A3, 0x11A7} /* Cn */, 927 { 0x11FA, 0x11FF} /* Cn */, { 0x1249, 0x1249} /* Cn */, 928 { 0x124E, 0x124F} /* Cn */, { 0x1257, 0x1257} /* Cn */, 929 { 0x1259, 0x1259} /* Cn */, { 0x125E, 0x125F} /* Cn */, 930 { 0x1289, 0x1289} /* Cn */, { 0x128E, 0x128F} /* Cn */, 931 { 0x12B1, 0x12B1} /* Cn */, { 0x12B6, 0x12B7} /* Cn */, 932 { 0x12BF, 0x12BF} /* Cn */, { 0x12C1, 0x12C1} /* Cn */, 933 { 0x12C6, 0x12C7} /* Cn */, { 0x12D7, 0x12D7} /* Cn */, 934 { 0x1311, 0x1311} /* Cn */, { 0x1316, 0x1317} /* Cn */, 935 { 0x135B, 0x135E} /* Cn */, { 0x137D, 0x137F} /* Cn */, 936 { 0x139A, 0x139F} /* Cn */, { 0x13F5, 0x1400} /* Cn */, 937 { 0x1677, 0x167F} /* Cn */, { 0x169D, 0x169F} /* Cn */, 938 { 0x16F1, 0x16FF} /* Cn */, { 0x170D, 0x170D} /* Cn */, 939 { 0x1715, 0x171F} /* Cn */, { 0x1737, 0x173F} /* Cn */, 940 { 0x1754, 0x175F} /* Cn */, { 0x176D, 0x176D} /* Cn */, 941 { 0x1771, 0x1771} /* Cn */, { 0x1774, 0x177F} /* Cn */, 942 #if 0 943 { 0x17B4, 0x17B5} /* Cf */, 944 #endif 945 { 0x17DE, 0x17DF} /* Cn */, { 0x17EA, 0x17EF} /* Cn */, 946 { 0x17FA, 0x17FF} /* Cn */, { 0x180F, 0x180F} /* Cn */, 947 { 0x181A, 0x181F} /* Cn */, { 0x1878, 0x187F} /* Cn */, 948 { 0x18AA, 0x18FF} /* Cn */, { 0x191D, 0x191F} /* Cn */, 949 { 0x192C, 0x192F} /* Cn */, { 0x193C, 0x193F} /* Cn */, 950 { 0x1941, 0x1943} /* Cn */, { 0x196E, 0x196F} /* Cn */, 951 { 0x1975, 0x197F} /* Cn */, { 0x19AA, 0x19AF} /* Cn */, 952 { 0x19CA, 0x19CF} /* Cn */, { 0x19DA, 0x19DD} /* Cn */, 953 { 0x1A1C, 0x1A1D} /* Cn */, { 0x1A20, 0x1AFF} /* Cn */, 954 { 0x1B4C, 0x1B4F} /* Cn */, { 0x1B7D, 0x1CFF} /* Cn */, 955 { 0x1DCB, 0x1DFD} /* Cn */, { 0x1E9C, 0x1E9F} /* Cn */, 956 { 0x1EFA, 0x1EFF} /* Cn */, { 0x1F16, 0x1F17} /* Cn */, 957 { 0x1F1E, 0x1F1F} /* Cn */, { 0x1F46, 0x1F47} /* Cn */, 958 { 0x1F4E, 0x1F4F} /* Cn */, { 0x1F58, 0x1F58} /* Cn */, 959 { 0x1F5A, 0x1F5A} /* Cn */, { 0x1F5C, 0x1F5C} /* Cn */, 960 { 0x1F5E, 0x1F5E} /* Cn */, { 0x1F7E, 0x1F7F} /* Cn */, 961 { 0x1FB5, 0x1FB5} /* Cn */, { 0x1FC5, 0x1FC5} /* Cn */, 962 { 0x1FD4, 0x1FD5} /* Cn */, { 0x1FDC, 0x1FDC} /* Cn */, 963 { 0x1FF0, 0x1FF1} /* Cn */, { 0x1FF5, 0x1FF5} /* Cn */, 964 { 0x1FFF, 0x1FFF} /* Cn */, 965 { 0x200B, 0x200F} /* Cf */, 966 { 0x2028, 0x2028} /* Zl */, 967 { 0x2029, 0x2029} /* Zp */, 968 { 0x202A, 0x202E} /* Cf */, 969 { 0x2060, 0x2063} /* Cf */, 970 { 0x2064, 0x2069} /* Cn */, 971 { 0x206A, 0x206F} /* Cf */, 972 { 0x2072, 0x2073} /* Cn */, { 0x208F, 0x208F} /* Cn */, 973 { 0x2095, 0x209F} /* Cn */, { 0x20B6, 0x20CF} /* Cn */, 974 { 0x20F0, 0x20FF} /* Cn */, { 0x214F, 0x2152} /* Cn */, 975 { 0x2185, 0x218F} /* Cn */, { 0x23E8, 0x23FF} /* Cn */, 976 { 0x2427, 0x243F} /* Cn */, { 0x244B, 0x245F} /* Cn */, 977 { 0x269D, 0x269F} /* Cn */, { 0x26B3, 0x2700} /* Cn */, 978 { 0x2705, 0x2705} /* Cn */, { 0x270A, 0x270B} /* Cn */, 979 { 0x2728, 0x2728} /* Cn */, { 0x274C, 0x274C} /* Cn */, 980 { 0x274E, 0x274E} /* Cn */, { 0x2753, 0x2755} /* Cn */, 981 { 0x2757, 0x2757} /* Cn */, { 0x275F, 0x2760} /* Cn */, 982 { 0x2795, 0x2797} /* Cn */, { 0x27B0, 0x27B0} /* Cn */, 983 { 0x27BF, 0x27BF} /* Cn */, { 0x27CB, 0x27CF} /* Cn */, 984 { 0x27EC, 0x27EF} /* Cn */, { 0x2B1B, 0x2B1F} /* Cn */, 985 { 0x2B24, 0x2BFF} /* Cn */, { 0x2C2F, 0x2C2F} /* Cn */, 986 { 0x2C5F, 0x2C5F} /* Cn */, { 0x2C6D, 0x2C73} /* Cn */, 987 { 0x2C78, 0x2C7F} /* Cn */, { 0x2CEB, 0x2CF8} /* Cn */, 988 { 0x2D26, 0x2D2F} /* Cn */, { 0x2D66, 0x2D6E} /* Cn */, 989 { 0x2D70, 0x2D7F} /* Cn */, { 0x2D97, 0x2D9F} /* Cn */, 990 { 0x2DA7, 0x2DA7} /* Cn */, { 0x2DAF, 0x2DAF} /* Cn */, 991 { 0x2DB7, 0x2DB7} /* Cn */, { 0x2DBF, 0x2DBF} /* Cn */, 992 { 0x2DC7, 0x2DC7} /* Cn */, { 0x2DCF, 0x2DCF} /* Cn */, 993 { 0x2DD7, 0x2DD7} /* Cn */, { 0x2DDF, 0x2DFF} /* Cn */, 994 { 0x2E18, 0x2E1B} /* Cn */, { 0x2E1E, 0x2E7F} /* Cn */, 995 { 0x2E9A, 0x2E9A} /* Cn */, { 0x2EF4, 0x2EFF} /* Cn */, 996 { 0x2FD6, 0x2FEF} /* Cn */, { 0x2FFC, 0x2FFF} /* Cn */, 997 { 0x3040, 0x3040} /* Cn */, { 0x3097, 0x3098} /* Cn */, 998 { 0x3100, 0x3104} /* Cn */, { 0x312D, 0x3130} /* Cn */, 999 { 0x318F, 0x318F} /* Cn */, { 0x31B8, 0x31BF} /* Cn */, 1000 { 0x31D0, 0x31EF} /* Cn */, { 0x321F, 0x321F} /* Cn */, 1001 { 0x3244, 0x324F} /* Cn */, { 0x32FF, 0x32FF} /* Cn */, 1002 { 0x4DB6, 0x4DBF} /* Cn */, { 0x9FBC, 0x9FFF} /* Cn */, 1003 { 0xA48D, 0xA48F} /* Cn */, { 0xA4C7, 0xA6FF} /* Cn */, 1004 { 0xA71B, 0xA71F} /* Cn */, { 0xA722, 0xA7FF} /* Cn */, 1005 { 0xA82C, 0xA83F} /* Cn */, { 0xA878, 0xABFF} /* Cn */, 1006 { 0xD7A4, 0xD7FF} /* Cn */, 1007 { 0xD800, 0xDFFF} /* Cs */, 1008 { 0xE000, 0xF8FF} /* Co */, 1009 { 0xFA2E, 0xFA2F} /* Cn */, { 0xFA6B, 0xFA6F} /* Cn */, 1010 { 0xFADA, 0xFAFF} /* Cn */, { 0xFB07, 0xFB12} /* Cn */, 1011 { 0xFB18, 0xFB1C} /* Cn */, { 0xFB37, 0xFB37} /* Cn */, 1012 { 0xFB3D, 0xFB3D} /* Cn */, { 0xFB3F, 0xFB3F} /* Cn */, 1013 { 0xFB42, 0xFB42} /* Cn */, { 0xFB45, 0xFB45} /* Cn */, 1014 { 0xFBB2, 0xFBD2} /* Cn */, { 0xFD40, 0xFD4F} /* Cn */, 1015 { 0xFD90, 0xFD91} /* Cn */, { 0xFDC8, 0xFDEF} /* Cn */, 1016 { 0xFDFE, 0xFDFF} /* Cn */, { 0xFE1A, 0xFE1F} /* Cn */, 1017 { 0xFE24, 0xFE2F} /* Cn */, { 0xFE53, 0xFE53} /* Cn */, 1018 { 0xFE67, 0xFE67} /* Cn */, { 0xFE6C, 0xFE6F} /* Cn */, 1019 { 0xFE75, 0xFE75} /* Cn */, { 0xFEFD, 0xFEFE} /* Cn */, 1020 { 0xFEFF, 0xFEFF} /* Cf */, 1021 { 0xFF00, 0xFF00} /* Cn */, { 0xFFBF, 0xFFC1} /* Cn */, 1022 { 0xFFC8, 0xFFC9} /* Cn */, { 0xFFD0, 0xFFD1} /* Cn */, 1023 { 0xFFD8, 0xFFD9} /* Cn */, { 0xFFDD, 0xFFDF} /* Cn */, 1024 { 0xFFE7, 0xFFE7} /* Cn */, { 0xFFEF, 0xFFF8} /* Cn */, 1025 { 0xFFF9, 0xFFFB} /* Cf */, 1026 { 0xFFFE, 0xFFFF} /* Cn */, { 0x1000C, 0x1000C} /* Cn */, 1027 { 0x10027, 0x10027} /* Cn */, { 0x1003B, 0x1003B} /* Cn */, 1028 { 0x1003E, 0x1003E} /* Cn */, { 0x1004E, 0x1004F} /* Cn */, 1029 { 0x1005E, 0x1007F} /* Cn */, { 0x100FB, 0x100FF} /* Cn */, 1030 { 0x10103, 0x10106} /* Cn */, { 0x10134, 0x10136} /* Cn */, 1031 { 0x1018B, 0x102FF} /* Cn */, { 0x1031F, 0x1031F} /* Cn */, 1032 { 0x10324, 0x1032F} /* Cn */, { 0x1034B, 0x1037F} /* Cn */, 1033 { 0x1039E, 0x1039E} /* Cn */, { 0x103C4, 0x103C7} /* Cn */, 1034 { 0x103D6, 0x103FF} /* Cn */, 1035 { 0x1049E, 0x1049F} /* Cn */, { 0x104AA, 0x107FF} /* Cn */, 1036 { 0x10806, 0x10807} /* Cn */, { 0x10809, 0x10809} /* Cn */, 1037 { 0x10836, 0x10836} /* Cn */, { 0x10839, 0x1083B} /* Cn */, 1038 { 0x1083D, 0x1083E} /* Cn */, { 0x10840, 0x108FF} /* Cn */, 1039 { 0x1091A, 0x1091E} /* Cn */, { 0x10920, 0x109FF} /* Cn */, 1040 { 0x10A04, 0x10A04} /* Cn */, { 0x10A07, 0x10A0B} /* Cn */, 1041 { 0x10A14, 0x10A14} /* Cn */, { 0x10A18, 0x10A18} /* Cn */, 1042 { 0x10A34, 0x10A37} /* Cn */, { 0x10A3B, 0x10A3E} /* Cn */, 1043 { 0x10A48, 0x10A4F} /* Cn */, { 0x10A59, 0x11FFF} /* Cn */, 1044 { 0x1236F, 0x123FF} /* Cn */, { 0x12463, 0x1246F} /* Cn */, 1045 { 0x12474, 0x1CFFF} /* Cn */, { 0x1D0F6, 0x1D0FF} /* Cn */, 1046 { 0x1D127, 0x1D129} /* Cn */, 1047 { 0x1D173, 0x1D17A} /* Cf */, 1048 { 0x1D1DE, 0x1D1FF} /* Cn */, { 0x1D246, 0x1D2FF} /* Cn */, 1049 { 0x1D357, 0x1D35F} /* Cn */, { 0x1D372, 0x1D3FF} /* Cn */, 1050 { 0x1D455, 0x1D455} /* Cn */, { 0x1D49D, 0x1D49D} /* Cn */, 1051 { 0x1D4A0, 0x1D4A1} /* Cn */, { 0x1D4A3, 0x1D4A4} /* Cn */, 1052 { 0x1D4A7, 0x1D4A8} /* Cn */, { 0x1D4AD, 0x1D4AD} /* Cn */, 1053 { 0x1D4BA, 0x1D4BA} /* Cn */, { 0x1D4BC, 0x1D4BC} /* Cn */, 1054 { 0x1D4C4, 0x1D4C4} /* Cn */, { 0x1D506, 0x1D506} /* Cn */, 1055 { 0x1D50B, 0x1D50C} /* Cn */, { 0x1D515, 0x1D515} /* Cn */, 1056 { 0x1D51D, 0x1D51D} /* Cn */, { 0x1D53A, 0x1D53A} /* Cn */, 1057 { 0x1D53F, 0x1D53F} /* Cn */, { 0x1D545, 0x1D545} /* Cn */, 1058 { 0x1D547, 0x1D549} /* Cn */, { 0x1D551, 0x1D551} /* Cn */, 1059 { 0x1D6A6, 0x1D6A7} /* Cn */, { 0x1D7CC, 0x1D7CD} /* Cn */, 1060 { 0x1D800, 0x1FFFF} /* Cn */, { 0x2A6D7, 0x2F7FF} /* Cn */, 1061 { 0x2FA1E, 0xE0000} /* Cn */, 1062 { 0xE0001, 0xE0001} /* Cf */, 1063 { 0xE0002, 0xE001F} /* Cn */, 1064 { 0xE0020, 0xE007F} /* Cf */, 1065 { 0xE0080, 0xE00FF} /* Cn */, { 0xE01F0, 0xEFFFF} /* Cn */, 1066 { 0xF0000, 0xFFFFD} /* Co */, 1067 { 0xFFFFE, 0xFFFFF} /* Cn */, 1068 {0x100000,0x10FFFD} /* Co */, 1069 {0x10FFFE,0x10FFFF} /* Cn */, 1070 {0x110000,0x7FFFFFFF} /* ISO 10646?? */ 1071 }; 1072 1073 /* 1074 * Double width characters 1075 * W: East Asian Wide 1076 * F: East Asian Full-width 1077 * Unassigned code points may be included when they allow ranges to be merged. 1078 * Last synched with 1079 * <http://www.unicode.org/Public/5.0.0/ucd/EastAsianWidth-5.0.0d2.txt> 1080 * dated 2005-11-08T01:32:56Z 1081 */ 1082 static struct wchar_range wide_table[] = { 1083 { 0x1100, 0x115F} /* W */, { 0x2329, 0x232A} /* W */, 1084 { 0x2E80, 0x2FFB} /* W */, 1085 { 0x3000, 0x3000} /* F */, 1086 { 0x3001, 0x303E} /* W */, { 0x3041, 0x4DB5} /* W */, 1087 { 0x4E00, 0x9FBB} /* W */, { 0xA000, 0xA4C6} /* W */, 1088 { 0xAC00, 0xD7A3} /* W */, { 0xF900, 0xFAD9} /* W */, 1089 { 0xFE10, 0xFE19} /* W */, { 0xFE30, 0xFE6B} /* W */, 1090 { 0xFF01, 0xFF60} /* F */, { 0xFFE0, 0xFFE6} /* F */, 1091 { 0x20000, 0x2FFFD} /* W */, { 0x30000, 0x3FFFD} /* W */, 1092 }; 1093 1094 static int 1095 is_in_table(ch, table, tsize) 1096 LWCHAR ch; 1097 struct wchar_range table[]; 1098 int tsize; 1099 { 1100 int hi; 1101 int lo; 1102 1103 /* Binary search in the table. */ 1104 if (ch < table[0].first) 1105 return 0; 1106 lo = 0; 1107 hi = tsize - 1; 1108 while (lo <= hi) 1109 { 1110 int mid = (lo + hi) / 2; 1111 if (ch > table[mid].last) 1112 lo = mid + 1; 1113 else if (ch < table[mid].first) 1114 hi = mid - 1; 1115 else 1116 return 1; 1117 } 1118 return 0; 1119 } 1120 1121 /* 1122 * Is a character a UTF-8 composing character? 1123 * If a composing character follows any char, the two combine into one glyph. 1124 */ 1125 public int 1126 is_composing_char(ch) 1127 LWCHAR ch; 1128 { 1129 return is_in_table(ch, comp_table, (sizeof(comp_table) / sizeof(*comp_table))); 1130 } 1131 1132 /* 1133 * Should this UTF-8 character be treated as binary? 1134 */ 1135 public int 1136 is_ubin_char(ch) 1137 LWCHAR ch; 1138 { 1139 return is_in_table(ch, ubin_table, (sizeof(ubin_table) / sizeof(*ubin_table))); 1140 } 1141 1142 /* 1143 * Is this a double width UTF-8 character? 1144 */ 1145 public int 1146 is_wide_char(ch) 1147 LWCHAR ch; 1148 { 1149 return is_in_table(ch, wide_table, (sizeof(wide_table) / sizeof(*wide_table))); 1150 } 1151 1152 /* 1153 * Is a character a UTF-8 combining character? 1154 * A combining char acts like an ordinary char, but if it follows 1155 * a specific char (not any char), the two combine into one glyph. 1156 */ 1157 public int 1158 is_combining_char(ch1, ch2) 1159 LWCHAR ch1; 1160 LWCHAR ch2; 1161 { 1162 /* The table is small; use linear search. */ 1163 int i; 1164 for (i = 0; i < sizeof(comb_table)/sizeof(*comb_table); i++) 1165 { 1166 if (ch1 == comb_table[i].first && 1167 ch2 == comb_table[i].last) 1168 return 1; 1169 } 1170 return 0; 1171 } 1172 1173