Annotation of src/usr.bin/less/charset.c, Revision 1.1.1.4
1.1 etheisen 1: /*
1.1.1.4 ! shadchin 2: * Copyright (C) 1984-2012 Mark Nudelman
1.1 etheisen 3: *
1.1.1.2 millert 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.
1.1 etheisen 6: *
1.1.1.4 ! shadchin 7: * For more information, see the README file.
1.1 etheisen 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>
1.1.1.3 shadchin 20: #include <langinfo.h>
1.1 etheisen 21: #endif
22:
1.1.1.3 shadchin 23: #include "charset.h"
24:
1.1.1.2 millert 25: public int utf_mode = 0;
26:
1.1.1.4 ! shadchin 27: #if !SMALL
1.1 etheisen 28: /*
29: * Predefined character sets,
30: * selected by the LESSCHARSET environment variable.
31: */
32: struct charset {
33: char *name;
1.1.1.2 millert 34: int *p_flag;
1.1 etheisen 35: char *desc;
36: } charsets[] = {
1.1.1.3 shadchin 37: { "ascii", NULL, "8bcccbcc18b95.b" },
38: { "utf-8", &utf_mode, "8bcccbcc18b95.b126.bb" },
39: { "iso8859", NULL, "8bcccbcc18b95.33b." },
40: { "latin3", NULL, "8bcccbcc18b95.33b5.b8.b15.b4.b12.b18.b12.b." },
41: { "arabic", NULL, "8bcccbcc18b95.33b.3b.7b2.13b.3b.b26.5b19.b" },
42: { "greek", NULL, "8bcccbcc18b95.33b4.2b4.b3.b35.b44.b" },
43: { "greek2005", NULL, "8bcccbcc18b95.33b14.b35.b44.b" },
44: { "hebrew", NULL, "8bcccbcc18b95.33b.b29.32b28.2b2.b" },
45: { "koi8-r", NULL, "8bcccbcc18b95.b." },
46: { "KOI8-T", NULL, "8bcccbcc18b95.b8.b6.b8.b.b.5b7.3b4.b4.b3.b.b.3b." },
47: { "georgianps", NULL, "8bcccbcc18b95.3b11.4b12.2b." },
48: { "tcvn", NULL, "b..b...bcccbccbbb7.8b95.b48.5b." },
49: { "TIS-620", NULL, "8bcccbcc18b95.b.4b.11b7.8b." },
50: { "next", NULL, "8bcccbcc18b95.bb125.bb" },
51: { "dos", NULL, "8bcccbcc12bc5b95.b." },
52: { "windows-1251", NULL, "8bcccbcc12bc5b95.b24.b." },
53: { "windows-1252", NULL, "8bcccbcc12bc5b95.b.b11.b.2b12.b." },
54: { "windows-1255", NULL, "8bcccbcc12bc5b95.b.b8.b.5b9.b.4b." },
55: { "ebcdic", NULL, "5bc6bcc7bcc41b.9b7.9b5.b..8b6.10b6.b9.7b9.8b8.17b3.3b9.7b9.8b8.6b10.b.b.b." },
56: { "IBM-1047", NULL, "4cbcbc3b9cbccbccbb4c6bcc5b3cbbc4bc4bccbc191.b" },
1.1.1.2 millert 57: { NULL, NULL, NULL }
58: };
59:
1.1.1.3 shadchin 60: /*
61: * Support "locale charmap"/nl_langinfo(CODESET) values, as well as others.
62: */
1.1.1.2 millert 63: struct cs_alias {
64: char *name;
65: char *oname;
66: } cs_aliases[] = {
1.1.1.3 shadchin 67: { "UTF-8", "utf-8" },
68: { "ANSI_X3.4-1968", "ascii" },
69: { "US-ASCII", "ascii" },
70: { "latin1", "iso8859" },
71: { "ISO-8859-1", "iso8859" },
72: { "latin9", "iso8859" },
73: { "ISO-8859-15", "iso8859" },
74: { "latin2", "iso8859" },
75: { "ISO-8859-2", "iso8859" },
76: { "ISO-8859-3", "latin3" },
77: { "latin4", "iso8859" },
78: { "ISO-8859-4", "iso8859" },
79: { "cyrillic", "iso8859" },
80: { "ISO-8859-5", "iso8859" },
81: { "ISO-8859-6", "arabic" },
82: { "ISO-8859-7", "greek" },
83: { "IBM9005", "greek2005" },
84: { "ISO-8859-8", "hebrew" },
85: { "latin5", "iso8859" },
86: { "ISO-8859-9", "iso8859" },
87: { "latin6", "iso8859" },
88: { "ISO-8859-10", "iso8859" },
89: { "latin7", "iso8859" },
90: { "ISO-8859-13", "iso8859" },
91: { "latin8", "iso8859" },
92: { "ISO-8859-14", "iso8859" },
93: { "latin10", "iso8859" },
94: { "ISO-8859-16", "iso8859" },
95: { "IBM437", "dos" },
96: { "EBCDIC-US", "ebcdic" },
97: { "IBM1047", "IBM-1047" },
98: { "KOI8-R", "koi8-r" },
99: { "KOI8-U", "koi8-r" },
100: { "GEORGIAN-PS", "georgianps" },
101: { "TCVN5712-1", "tcvn" },
102: { "NEXTSTEP", "next" },
103: { "windows", "windows-1252" }, /* backward compatibility */
104: { "CP1251", "windows-1251" },
105: { "CP1252", "windows-1252" },
106: { "CP1255", "windows-1255" },
1.1.1.2 millert 107: { NULL, NULL }
1.1 etheisen 108: };
109:
110: #define IS_BINARY_CHAR 01
111: #define IS_CONTROL_CHAR 02
112:
113: static char chardef[256];
114: static char *binfmt = NULL;
1.1.1.3 shadchin 115: static char *utfbinfmt = NULL;
1.1 etheisen 116: public int binattr = AT_STANDOUT;
117:
118:
119: /*
120: * Define a charset, given a description string.
121: * The string consists of 256 letters,
122: * one for each character in the charset.
123: * If the string is shorter than 256 letters, missing letters
124: * are taken to be identical to the last one.
125: * A decimal number followed by a letter is taken to be a
126: * repetition of the letter.
127: *
128: * Each letter is one of:
129: * . normal character
130: * b binary character
131: * c control character
132: */
133: static void
134: ichardef(s)
135: char *s;
136: {
137: register char *cp;
138: register int n;
139: register char v;
140:
141: n = 0;
142: v = 0;
143: cp = chardef;
144: while (*s != '\0')
145: {
146: switch (*s++)
147: {
148: case '.':
149: v = 0;
150: break;
151: case 'c':
152: v = IS_CONTROL_CHAR;
153: break;
154: case 'b':
155: v = IS_BINARY_CHAR|IS_CONTROL_CHAR;
156: break;
157:
158: case '0': case '1': case '2': case '3': case '4':
159: case '5': case '6': case '7': case '8': case '9':
160: n = (10 * n) + (s[-1] - '0');
161: continue;
162:
163: default:
164: error("invalid chardef", NULL_PARG);
165: quit(QUIT_ERROR);
166: /*NOTREACHED*/
167: }
168:
169: do
170: {
171: if (cp >= chardef + sizeof(chardef))
172: {
173: error("chardef longer than 256", NULL_PARG);
174: quit(QUIT_ERROR);
175: /*NOTREACHED*/
176: }
177: *cp++ = v;
178: } while (--n > 0);
179: n = 0;
180: }
181:
182: while (cp < chardef + sizeof(chardef))
183: *cp++ = v;
184: }
185:
186: /*
187: * Define a charset, given a charset name.
188: * The valid charset names are listed in the "charsets" array.
189: */
190: static int
1.1.1.3 shadchin 191: icharset(name, no_error)
1.1 etheisen 192: register char *name;
1.1.1.3 shadchin 193: int no_error;
1.1 etheisen 194: {
195: register struct charset *p;
1.1.1.2 millert 196: register struct cs_alias *a;
1.1 etheisen 197:
198: if (name == NULL || *name == '\0')
199: return (0);
200:
1.1.1.2 millert 201: /* First see if the name is an alias. */
202: for (a = cs_aliases; a->name != NULL; a++)
203: {
204: if (strcmp(name, a->name) == 0)
205: {
206: name = a->oname;
207: break;
208: }
209: }
210:
1.1 etheisen 211: for (p = charsets; p->name != NULL; p++)
212: {
213: if (strcmp(name, p->name) == 0)
214: {
215: ichardef(p->desc);
1.1.1.2 millert 216: if (p->p_flag != NULL)
217: *(p->p_flag) = 1;
1.1 etheisen 218: return (1);
219: }
220: }
221:
1.1.1.3 shadchin 222: if (!no_error) {
223: error("invalid charset name", NULL_PARG);
224: quit(QUIT_ERROR);
225: }
1.1.1.2 millert 226: return (0);
1.1 etheisen 227: }
228:
229: #if HAVE_LOCALE
230: /*
231: * Define a charset, given a locale name.
232: */
233: static void
234: ilocale()
235: {
236: register int c;
237:
1.1.1.2 millert 238: for (c = 0; c < (int) sizeof(chardef); c++)
1.1 etheisen 239: {
240: if (isprint(c))
241: chardef[c] = 0;
242: else if (iscntrl(c))
243: chardef[c] = IS_CONTROL_CHAR;
244: else
245: chardef[c] = IS_BINARY_CHAR|IS_CONTROL_CHAR;
246: }
247: }
248: #endif
249:
250: /*
1.1.1.3 shadchin 251: * Define the printing format for control (or binary utf) chars.
1.1 etheisen 252: */
1.1.1.3 shadchin 253: static void
254: setbinfmt(s, fmtvarptr, default_fmt)
1.1 etheisen 255: char *s;
1.1.1.3 shadchin 256: char **fmtvarptr;
257: char *default_fmt;
1.1 etheisen 258: {
1.1.1.3 shadchin 259: if (s && utf_mode)
260: {
261: /* It would be too hard to account for width otherwise. */
262: char *t = s;
263: while (*t)
264: {
265: if (*t < ' ' || *t > '~')
266: {
267: s = default_fmt;
268: goto attr;
269: }
270: t++;
271: }
272: }
273:
274: /* %n is evil */
275: if (s == NULL || *s == '\0' ||
276: (*s == '*' && (s[1] == '\0' || s[2] == '\0' || strchr(s + 2, 'n'))) ||
277: (*s != '*' && strchr(s, 'n')))
278: s = default_fmt;
279:
1.1 etheisen 280: /*
281: * Select the attributes if it starts with "*".
282: */
1.1.1.3 shadchin 283: attr:
1.1 etheisen 284: if (*s == '*')
285: {
286: switch (s[1])
287: {
288: case 'd': binattr = AT_BOLD; break;
289: case 'k': binattr = AT_BLINK; break;
290: case 's': binattr = AT_STANDOUT; break;
291: case 'u': binattr = AT_UNDERLINE; break;
292: default: binattr = AT_NORMAL; break;
293: }
294: s += 2;
295: }
1.1.1.3 shadchin 296: *fmtvarptr = s;
1.1 etheisen 297: }
298:
299: /*
1.1.1.3 shadchin 300: *
1.1 etheisen 301: */
1.1.1.3 shadchin 302: static void
303: set_charset()
1.1 etheisen 304: {
1.1.1.3 shadchin 305: char *s;
1.1 etheisen 306:
307: /*
308: * See if environment variable LESSCHARSET is defined.
309: */
1.1.1.2 millert 310: s = lgetenv("LESSCHARSET");
1.1.1.3 shadchin 311: if (icharset(s, 0))
1.1 etheisen 312: return;
1.1.1.3 shadchin 313:
1.1 etheisen 314: /*
315: * LESSCHARSET is not defined: try LESSCHARDEF.
316: */
1.1.1.2 millert 317: s = lgetenv("LESSCHARDEF");
1.1 etheisen 318: if (s != NULL && *s != '\0')
319: {
320: ichardef(s);
321: return;
322: }
1.1.1.2 millert 323:
1.1.1.3 shadchin 324: #if HAVE_LOCALE
325: #ifdef CODESET
326: /*
327: * Try using the codeset name as the charset name.
328: */
329: s = nl_langinfo(CODESET);
330: if (icharset(s, 1))
331: return;
332: #endif
333: #endif
334:
1.1.1.2 millert 335: #if HAVE_STRSTR
336: /*
337: * Check whether LC_ALL, LC_CTYPE or LANG look like UTF-8 is used.
338: */
339: if ((s = lgetenv("LC_ALL")) != NULL ||
340: (s = lgetenv("LC_CTYPE")) != NULL ||
341: (s = lgetenv("LANG")) != NULL)
342: {
1.1.1.3 shadchin 343: if ( strstr(s, "UTF-8") != NULL || strstr(s, "utf-8") != NULL
344: || strstr(s, "UTF8") != NULL || strstr(s, "utf8") != NULL)
345: if (icharset("utf-8", 1))
1.1.1.2 millert 346: return;
347: }
348: #endif
349:
1.1 etheisen 350: #if HAVE_LOCALE
351: /*
1.1.1.3 shadchin 352: * Get character definitions from locale functions,
353: * rather than from predefined charset entry.
1.1 etheisen 354: */
355: ilocale();
1.1.1.2 millert 356: #if MSDOS_COMPILER
357: /*
358: * Default to "dos".
359: */
1.1.1.3 shadchin 360: (void) icharset("dos", 1);
1.1.1.2 millert 361: #else
1.1 etheisen 362: /*
1.1.1.2 millert 363: * Default to "latin1".
1.1 etheisen 364: */
1.1.1.3 shadchin 365: (void) icharset("latin1", 1);
1.1.1.2 millert 366: #endif
1.1 etheisen 367: #endif
368: }
369:
370: /*
1.1.1.3 shadchin 371: * Initialize charset data structures.
372: */
373: public void
374: init_charset()
375: {
376: char *s;
377:
378: #if HAVE_LOCALE
379: setlocale(LC_ALL, "");
380: #endif
381:
382: set_charset();
383:
384: s = lgetenv("LESSBINFMT");
385: setbinfmt(s, &binfmt, "*s<%02X>");
386:
387: s = lgetenv("LESSUTFBINFMT");
388: setbinfmt(s, &utfbinfmt, "<U+%04lX>");
389: }
390:
391: /*
1.1 etheisen 392: * Is a given character a "binary" character?
393: */
394: public int
395: binary_char(c)
1.1.1.3 shadchin 396: LWCHAR c;
1.1 etheisen 397: {
1.1.1.3 shadchin 398: if (utf_mode)
399: return (is_ubin_char(c));
1.1 etheisen 400: c &= 0377;
401: return (chardef[c] & IS_BINARY_CHAR);
402: }
403:
404: /*
405: * Is a given character a "control" character?
406: */
407: public int
408: control_char(c)
1.1.1.3 shadchin 409: LWCHAR c;
1.1 etheisen 410: {
411: c &= 0377;
412: return (chardef[c] & IS_CONTROL_CHAR);
413: }
414:
415: /*
416: * Return the printable form of a character.
417: * For example, in the "ascii" charset '\3' is printed as "^C".
418: */
419: public char *
420: prchar(c)
1.1.1.3 shadchin 421: LWCHAR c;
1.1 etheisen 422: {
1.1.1.3 shadchin 423: /* {{ This buffer can be overrun if LESSBINFMT is a long string. }} */
424: static char buf[32];
1.1 etheisen 425:
426: c &= 0377;
1.1.1.3 shadchin 427: if ((c < 128 || !utf_mode) && !control_char(c))
428: SNPRINTF1(buf, sizeof(buf), "%c", (int) c);
1.1 etheisen 429: else if (c == ESC)
1.1.1.4 ! shadchin 430: strlcpy(buf, "ESC", sizeof(buf));
1.1.1.2 millert 431: #if IS_EBCDIC_HOST
432: else if (!binary_char(c) && c < 64)
1.1.1.3 shadchin 433: SNPRINTF1(buf, sizeof(buf), "^%c",
1.1.1.2 millert 434: /*
435: * This array roughly inverts CONTROL() #defined in less.h,
436: * and should be kept in sync with CONTROL() and IBM-1047.
437: */
438: "@ABC.I.?...KLMNO"
439: "PQRS.JH.XY.."
440: "\\]^_"
441: "......W[.....EFG"
442: "..V....D....TU.Z"[c]);
443: #else
444: else if (c < 128 && !control_char(c ^ 0100))
1.1.1.3 shadchin 445: SNPRINTF1(buf, sizeof(buf), "^%c", (int) (c ^ 0100));
1.1.1.2 millert 446: #endif
1.1 etheisen 447: else
1.1.1.3 shadchin 448: SNPRINTF1(buf, sizeof(buf), binfmt, c);
1.1 etheisen 449: return (buf);
450: }
1.1.1.3 shadchin 451:
452: /*
453: * Return the printable form of a UTF-8 character.
454: */
455: public char *
456: prutfchar(ch)
457: LWCHAR ch;
458: {
459: static char buf[32];
460:
461: if (ch == ESC)
1.1.1.4 ! shadchin 462: strlcpy(buf, "ESC", sizeof(buf));
1.1.1.3 shadchin 463: else if (ch < 128 && control_char(ch))
464: {
465: if (!control_char(ch ^ 0100))
466: SNPRINTF1(buf, sizeof(buf), "^%c", ((char) ch) ^ 0100);
467: else
468: SNPRINTF1(buf, sizeof(buf), binfmt, (char) ch);
469: } else if (is_ubin_char(ch))
470: SNPRINTF1(buf, sizeof(buf), utfbinfmt, ch);
471: else
472: {
473: int len;
474: if (ch >= 0x80000000)
475: {
476: len = 3;
477: ch = 0xFFFD;
478: } else
479: {
480: len = (ch < 0x80) ? 1
481: : (ch < 0x800) ? 2
482: : (ch < 0x10000) ? 3
483: : (ch < 0x200000) ? 4
484: : (ch < 0x4000000) ? 5
485: : 6;
486: }
487: buf[len] = '\0';
488: if (len == 1)
489: *buf = (char) ch;
490: else
491: {
492: *buf = ((1 << len) - 1) << (8 - len);
493: while (--len > 0)
494: {
495: buf[len] = (char) (0x80 | (ch & 0x3F));
496: ch >>= 6;
497: }
498: *buf |= ch;
499: }
500: }
501: return (buf);
502: }
503:
504: /*
505: * Get the length of a UTF-8 character in bytes.
506: */
507: public int
508: utf_len(ch)
509: char ch;
510: {
511: if ((ch & 0x80) == 0)
512: return 1;
513: if ((ch & 0xE0) == 0xC0)
514: return 2;
515: if ((ch & 0xF0) == 0xE0)
516: return 3;
517: if ((ch & 0xF8) == 0xF0)
518: return 4;
519: if ((ch & 0xFC) == 0xF8)
520: return 5;
521: if ((ch & 0xFE) == 0xFC)
522: return 6;
523: /* Invalid UTF-8 encoding. */
524: return 1;
525: }
526:
527: /*
528: * Is a UTF-8 character well-formed?
529: */
530: public int
531: is_utf8_well_formed(s)
532: unsigned char *s;
533: {
534: int i;
535: int len;
536:
537: if (IS_UTF8_INVALID(s[0]))
538: return (0);
539:
540: len = utf_len((char) s[0]);
541: if (len == 1)
542: return (1);
543: if (len == 2)
544: {
545: if (s[0] < 0xC2)
546: return (0);
547: } else
548: {
549: unsigned char mask;
550: mask = (~((1 << (8-len)) - 1)) & 0xFF;
551: if (s[0] == mask && (s[1] & mask) == 0x80)
552: return (0);
553: }
554:
555: for (i = 1; i < len; i++)
556: if (!IS_UTF8_TRAIL(s[i]))
557: return (0);
558: return (1);
559: }
560:
561: /*
562: * Get the value of a UTF-8 character.
563: */
564: public LWCHAR
565: get_wchar(p)
566: char *p;
567: {
568: switch (utf_len(p[0]))
569: {
570: case 1:
571: default:
572: /* 0xxxxxxx */
573: return (LWCHAR)
574: (p[0] & 0xFF);
575: case 2:
576: /* 110xxxxx 10xxxxxx */
577: return (LWCHAR) (
578: ((p[0] & 0x1F) << 6) |
579: (p[1] & 0x3F));
580: case 3:
581: /* 1110xxxx 10xxxxxx 10xxxxxx */
582: return (LWCHAR) (
583: ((p[0] & 0x0F) << 12) |
584: ((p[1] & 0x3F) << 6) |
585: (p[2] & 0x3F));
586: case 4:
587: /* 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
588: return (LWCHAR) (
589: ((p[0] & 0x07) << 18) |
590: ((p[1] & 0x3F) << 12) |
591: ((p[2] & 0x3F) << 6) |
592: (p[3] & 0x3F));
593: case 5:
594: /* 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx */
595: return (LWCHAR) (
596: ((p[0] & 0x03) << 24) |
597: ((p[1] & 0x3F) << 18) |
598: ((p[2] & 0x3F) << 12) |
599: ((p[3] & 0x3F) << 6) |
600: (p[4] & 0x3F));
601: case 6:
602: /* 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx */
603: return (LWCHAR) (
604: ((p[0] & 0x01) << 30) |
605: ((p[1] & 0x3F) << 24) |
606: ((p[2] & 0x3F) << 18) |
607: ((p[3] & 0x3F) << 12) |
608: ((p[4] & 0x3F) << 6) |
609: (p[5] & 0x3F));
610: }
611: }
612:
613: /*
614: * Store a character into a UTF-8 string.
615: */
616: public void
617: put_wchar(pp, ch)
618: char **pp;
619: LWCHAR ch;
620: {
621: if (!utf_mode || ch < 0x80)
622: {
623: /* 0xxxxxxx */
624: *(*pp)++ = (char) ch;
625: } else if (ch < 0x800)
626: {
627: /* 110xxxxx 10xxxxxx */
628: *(*pp)++ = (char) (0xC0 | ((ch >> 6) & 0x1F));
629: *(*pp)++ = (char) (0x80 | (ch & 0x3F));
630: } else if (ch < 0x10000)
631: {
632: /* 1110xxxx 10xxxxxx 10xxxxxx */
633: *(*pp)++ = (char) (0xE0 | ((ch >> 12) & 0x0F));
634: *(*pp)++ = (char) (0x80 | ((ch >> 6) & 0x3F));
635: *(*pp)++ = (char) (0x80 | (ch & 0x3F));
636: } else if (ch < 0x200000)
637: {
638: /* 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
639: *(*pp)++ = (char) (0xF0 | ((ch >> 18) & 0x07));
640: *(*pp)++ = (char) (0x80 | ((ch >> 12) & 0x3F));
641: *(*pp)++ = (char) (0x80 | ((ch >> 6) & 0x3F));
642: *(*pp)++ = (char) (0x80 | (ch & 0x3F));
643: } else if (ch < 0x4000000)
644: {
645: /* 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx */
646: *(*pp)++ = (char) (0xF0 | ((ch >> 24) & 0x03));
647: *(*pp)++ = (char) (0x80 | ((ch >> 18) & 0x3F));
648: *(*pp)++ = (char) (0x80 | ((ch >> 12) & 0x3F));
649: *(*pp)++ = (char) (0x80 | ((ch >> 6) & 0x3F));
650: *(*pp)++ = (char) (0x80 | (ch & 0x3F));
651: } else
652: {
653: /* 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx */
654: *(*pp)++ = (char) (0xF0 | ((ch >> 30) & 0x01));
655: *(*pp)++ = (char) (0x80 | ((ch >> 24) & 0x3F));
656: *(*pp)++ = (char) (0x80 | ((ch >> 18) & 0x3F));
657: *(*pp)++ = (char) (0x80 | ((ch >> 12) & 0x3F));
658: *(*pp)++ = (char) (0x80 | ((ch >> 6) & 0x3F));
659: *(*pp)++ = (char) (0x80 | (ch & 0x3F));
660: }
661: }
662:
663: /*
664: * Step forward or backward one character in a string.
665: */
666: public LWCHAR
667: step_char(pp, dir, limit)
668: char **pp;
669: signed int dir;
670: char *limit;
671: {
672: LWCHAR ch;
673: int len;
674: char *p = *pp;
675:
676: if (!utf_mode)
677: {
678: /* It's easy if chars are one byte. */
679: if (dir > 0)
680: ch = (LWCHAR) ((p < limit) ? *p++ : 0);
681: else
682: ch = (LWCHAR) ((p > limit) ? *--p : 0);
683: } else if (dir > 0)
684: {
685: len = utf_len(*p);
686: if (p + len > limit)
687: {
688: ch = 0;
689: p = limit;
690: } else
691: {
692: ch = get_wchar(p);
693: p += len;
694: }
695: } else
696: {
697: while (p > limit && IS_UTF8_TRAIL(p[-1]))
698: p--;
699: if (p > limit)
700: ch = get_wchar(--p);
701: else
702: ch = 0;
703: }
704: *pp = p;
705: return ch;
706: }
707:
708: /*
709: * Unicode characters data
710: */
711: struct wchar_range { LWCHAR first, last; };
712:
713: /*
714: * Characters with general category values
715: * Mn: Mark, Nonspacing
716: * Me: Mark, Enclosing
717: * Last synched with
718: * <http://www.unicode.org/Public/5.0.0/ucd/UnicodeData-5.0.0d7.txt>
719: * dated 2005-11-30T00:58:48Z
720: */
721: static struct wchar_range comp_table[] = {
722: { 0x0300, 0x036F} /* Mn */, { 0x0483, 0x0486} /* Mn */,
723: { 0x0488, 0x0489} /* Me */,
724: { 0x0591, 0x05BD} /* Mn */, { 0x05BF, 0x05BF} /* Mn */,
725: { 0x05C1, 0x05C2} /* Mn */, { 0x05C4, 0x05C5} /* Mn */,
726: { 0x05C7, 0x05C7} /* Mn */, { 0x0610, 0x0615} /* Mn */,
727: { 0x064B, 0x065E} /* Mn */, { 0x0670, 0x0670} /* Mn */,
728: { 0x06D6, 0x06DC} /* Mn */,
729: { 0x06DE, 0x06DE} /* Me */,
730: { 0x06DF, 0x06E4} /* Mn */, { 0x06E7, 0x06E8} /* Mn */,
731: { 0x06EA, 0x06ED} /* Mn */, { 0x0711, 0x0711} /* Mn */,
732: { 0x0730, 0x074A} /* Mn */, { 0x07A6, 0x07B0} /* Mn */,
733: { 0x07EB, 0x07F3} /* Mn */, { 0x0901, 0x0902} /* Mn */,
734: { 0x093C, 0x093C} /* Mn */, { 0x0941, 0x0948} /* Mn */,
735: { 0x094D, 0x094D} /* Mn */, { 0x0951, 0x0954} /* Mn */,
736: { 0x0962, 0x0963} /* Mn */, { 0x0981, 0x0981} /* Mn */,
737: { 0x09BC, 0x09BC} /* Mn */, { 0x09C1, 0x09C4} /* Mn */,
738: { 0x09CD, 0x09CD} /* Mn */, { 0x09E2, 0x09E3} /* Mn */,
739: { 0x0A01, 0x0A02} /* Mn */, { 0x0A3C, 0x0A3C} /* Mn */,
740: { 0x0A41, 0x0A42} /* Mn */, { 0x0A47, 0x0A48} /* Mn */,
741: { 0x0A4B, 0x0A4D} /* Mn */, { 0x0A70, 0x0A71} /* Mn */,
742: { 0x0A81, 0x0A82} /* Mn */, { 0x0ABC, 0x0ABC} /* Mn */,
743: { 0x0AC1, 0x0AC5} /* Mn */, { 0x0AC7, 0x0AC8} /* Mn */,
744: { 0x0ACD, 0x0ACD} /* Mn */, { 0x0AE2, 0x0AE3} /* Mn */,
745: { 0x0B01, 0x0B01} /* Mn */, { 0x0B3C, 0x0B3C} /* Mn */,
746: { 0x0B3F, 0x0B3F} /* Mn */, { 0x0B41, 0x0B43} /* Mn */,
747: { 0x0B4D, 0x0B4D} /* Mn */, { 0x0B56, 0x0B56} /* Mn */,
748: { 0x0B82, 0x0B82} /* Mn */, { 0x0BC0, 0x0BC0} /* Mn */,
749: { 0x0BCD, 0x0BCD} /* Mn */, { 0x0C3E, 0x0C40} /* Mn */,
750: { 0x0C46, 0x0C48} /* Mn */, { 0x0C4A, 0x0C4D} /* Mn */,
751: { 0x0C55, 0x0C56} /* Mn */, { 0x0CBC, 0x0CBC} /* Mn */,
752: { 0x0CBF, 0x0CBF} /* Mn */, { 0x0CC6, 0x0CC6} /* Mn */,
753: { 0x0CCC, 0x0CCD} /* Mn */, { 0x0CE2, 0x0CE3} /* Mn */,
754: { 0x0D41, 0x0D43} /* Mn */, { 0x0D4D, 0x0D4D} /* Mn */,
755: { 0x0DCA, 0x0DCA} /* Mn */, { 0x0DD2, 0x0DD4} /* Mn */,
756: { 0x0DD6, 0x0DD6} /* Mn */, { 0x0E31, 0x0E31} /* Mn */,
757: { 0x0E34, 0x0E3A} /* Mn */, { 0x0E47, 0x0E4E} /* Mn */,
758: { 0x0EB1, 0x0EB1} /* Mn */, { 0x0EB4, 0x0EB9} /* Mn */,
759: { 0x0EBB, 0x0EBC} /* Mn */, { 0x0EC8, 0x0ECD} /* Mn */,
760: { 0x0F18, 0x0F19} /* Mn */, { 0x0F35, 0x0F35} /* Mn */,
761: { 0x0F37, 0x0F37} /* Mn */, { 0x0F39, 0x0F39} /* Mn */,
762: { 0x0F71, 0x0F7E} /* Mn */, { 0x0F80, 0x0F84} /* Mn */,
763: { 0x0F86, 0x0F87} /* Mn */, { 0x0F90, 0x0F97} /* Mn */,
764: { 0x0F99, 0x0FBC} /* Mn */, { 0x0FC6, 0x0FC6} /* Mn */,
765: { 0x102D, 0x1030} /* Mn */, { 0x1032, 0x1032} /* Mn */,
766: { 0x1036, 0x1037} /* Mn */, { 0x1039, 0x1039} /* Mn */,
767: { 0x1058, 0x1059} /* Mn */, { 0x135F, 0x135F} /* Mn */,
768: { 0x1712, 0x1714} /* Mn */, { 0x1732, 0x1734} /* Mn */,
769: { 0x1752, 0x1753} /* Mn */, { 0x1772, 0x1773} /* Mn */,
770: { 0x17B7, 0x17BD} /* Mn */, { 0x17C6, 0x17C6} /* Mn */,
771: { 0x17C9, 0x17D3} /* Mn */, { 0x17DD, 0x17DD} /* Mn */,
772: { 0x180B, 0x180D} /* Mn */, { 0x18A9, 0x18A9} /* Mn */,
773: { 0x1920, 0x1922} /* Mn */, { 0x1927, 0x1928} /* Mn */,
774: { 0x1932, 0x1932} /* Mn */, { 0x1939, 0x193B} /* Mn */,
775: { 0x1A17, 0x1A18} /* Mn */, { 0x1B00, 0x1B03} /* Mn */,
776: { 0x1B34, 0x1B34} /* Mn */, { 0x1B36, 0x1B3A} /* Mn */,
777: { 0x1B3C, 0x1B3C} /* Mn */, { 0x1B42, 0x1B42} /* Mn */,
778: { 0x1B6B, 0x1B73} /* Mn */, { 0x1DC0, 0x1DCA} /* Mn */,
779: { 0x1DFE, 0x1DFF} /* Mn */, { 0x20D0, 0x20DC} /* Mn */,
780: { 0x20DD, 0x20E0} /* Me */,
781: { 0x20E1, 0x20E1} /* Mn */,
782: { 0x20E2, 0x20E4} /* Me */,
783: { 0x20E5, 0x20EF} /* Mn */, { 0x302A, 0x302F} /* Mn */,
784: { 0x3099, 0x309A} /* Mn */, { 0xA806, 0xA806} /* Mn */,
785: { 0xA80B, 0xA80B} /* Mn */, { 0xA825, 0xA826} /* Mn */,
786: { 0xFB1E, 0xFB1E} /* Mn */, { 0xFE00, 0xFE0F} /* Mn */,
787: { 0xFE20, 0xFE23} /* Mn */, { 0x10A01, 0x10A03} /* Mn */,
788: { 0x10A05, 0x10A06} /* Mn */, { 0x10A0C, 0x10A0F} /* Mn */,
789: { 0x10A38, 0x10A3A} /* Mn */, { 0x10A3F, 0x10A3F} /* Mn */,
790: { 0x1D167, 0x1D169} /* Mn */, { 0x1D17B, 0x1D182} /* Mn */,
791: { 0x1D185, 0x1D18B} /* Mn */, { 0x1D1AA, 0x1D1AD} /* Mn */,
792: { 0x1D242, 0x1D244} /* Mn */, { 0xE0100, 0xE01EF} /* Mn */,
793: };
794:
795: /*
796: * Special pairs, not ranges.
797: */
798: static struct wchar_range comb_table[] = {
799: {0x0644,0x0622}, {0x0644,0x0623}, {0x0644,0x0625}, {0x0644,0x0627},
800: };
801:
802: /*
803: * Characters with general category values
804: * Cc: Other, Control
805: * Cf: Other, Format
806: * Cs: Other, Surrogate
807: * Co: Other, Private Use
808: * Cn: Other, Not Assigned
809: * Zl: Separator, Line
810: * Zp: Separator, Paragraph
811: * Last synched with
812: * <http://www.unicode.org/Public/5.0.0/ucd/UnicodeData-5.0.0d7.txt>
813: * dated 2005-11-30T00:58:48Z
814: */
815: static struct wchar_range ubin_table[] = {
816: { 0x0000, 0x0007} /* Cc */,
817: { 0x000B, 0x000C} /* Cc */,
818: { 0x000E, 0x001A} /* Cc */,
819: { 0x001C, 0x001F} /* Cc */,
820: { 0x007F, 0x009F} /* Cc */,
821: #if 0
822: { 0x00AD, 0x00AD} /* Cf */,
823: #endif
824: { 0x0370, 0x0373} /* Cn */, { 0x0376, 0x0379} /* Cn */,
825: { 0x037F, 0x0383} /* Cn */, { 0x038B, 0x038B} /* Cn */,
826: { 0x038D, 0x038D} /* Cn */, { 0x03A2, 0x03A2} /* Cn */,
827: { 0x03CF, 0x03CF} /* Cn */, { 0x0487, 0x0487} /* Cn */,
828: { 0x0514, 0x0530} /* Cn */, { 0x0557, 0x0558} /* Cn */,
829: { 0x0560, 0x0560} /* Cn */, { 0x0588, 0x0588} /* Cn */,
830: { 0x058B, 0x0590} /* Cn */, { 0x05C8, 0x05CF} /* Cn */,
831: { 0x05EB, 0x05EF} /* Cn */, { 0x05F5, 0x05FF} /* Cn */,
832: #if 0
833: { 0x0600, 0x0603} /* Cf */,
834: #endif
835: { 0x0604, 0x060A} /* Cn */, { 0x0616, 0x061A} /* Cn */,
836: { 0x061C, 0x061D} /* Cn */, { 0x0620, 0x0620} /* Cn */,
837: { 0x063B, 0x063F} /* Cn */, { 0x065F, 0x065F} /* Cn */,
838: #if 0
839: { 0x06DD, 0x06DD} /* Cf */,
840: #endif
841: { 0x070E, 0x070E} /* Cn */,
842: #if 0
843: { 0x070F, 0x070F} /* Cf */,
844: #endif
845: { 0x074B, 0x074C} /* Cn */, { 0x076E, 0x077F} /* Cn */,
846: { 0x07B2, 0x07BF} /* Cn */, { 0x07FB, 0x0900} /* Cn */,
847: { 0x093A, 0x093B} /* Cn */, { 0x094E, 0x094F} /* Cn */,
848: { 0x0955, 0x0957} /* Cn */, { 0x0971, 0x097A} /* Cn */,
849: { 0x0980, 0x0980} /* Cn */, { 0x0984, 0x0984} /* Cn */,
850: { 0x098D, 0x098E} /* Cn */, { 0x0991, 0x0992} /* Cn */,
851: { 0x09A9, 0x09A9} /* Cn */, { 0x09B1, 0x09B1} /* Cn */,
852: { 0x09B3, 0x09B5} /* Cn */, { 0x09BA, 0x09BB} /* Cn */,
853: { 0x09C5, 0x09C6} /* Cn */, { 0x09C9, 0x09CA} /* Cn */,
854: { 0x09CF, 0x09D6} /* Cn */, { 0x09D8, 0x09DB} /* Cn */,
855: { 0x09DE, 0x09DE} /* Cn */, { 0x09E4, 0x09E5} /* Cn */,
856: { 0x09FB, 0x0A00} /* Cn */, { 0x0A04, 0x0A04} /* Cn */,
857: { 0x0A0B, 0x0A0E} /* Cn */, { 0x0A11, 0x0A12} /* Cn */,
858: { 0x0A29, 0x0A29} /* Cn */, { 0x0A31, 0x0A31} /* Cn */,
859: { 0x0A34, 0x0A34} /* Cn */, { 0x0A37, 0x0A37} /* Cn */,
860: { 0x0A3A, 0x0A3B} /* Cn */, { 0x0A3D, 0x0A3D} /* Cn */,
861: { 0x0A43, 0x0A46} /* Cn */, { 0x0A49, 0x0A4A} /* Cn */,
862: { 0x0A4E, 0x0A58} /* Cn */, { 0x0A5D, 0x0A5D} /* Cn */,
863: { 0x0A5F, 0x0A65} /* Cn */, { 0x0A75, 0x0A80} /* Cn */,
864: { 0x0A84, 0x0A84} /* Cn */, { 0x0A8E, 0x0A8E} /* Cn */,
865: { 0x0A92, 0x0A92} /* Cn */, { 0x0AA9, 0x0AA9} /* Cn */,
866: { 0x0AB1, 0x0AB1} /* Cn */, { 0x0AB4, 0x0AB4} /* Cn */,
867: { 0x0ABA, 0x0ABB} /* Cn */, { 0x0AC6, 0x0AC6} /* Cn */,
868: { 0x0ACA, 0x0ACA} /* Cn */, { 0x0ACE, 0x0ACF} /* Cn */,
869: { 0x0AD1, 0x0ADF} /* Cn */, { 0x0AE4, 0x0AE5} /* Cn */,
870: { 0x0AF0, 0x0AF0} /* Cn */, { 0x0AF2, 0x0B00} /* Cn */,
871: { 0x0B04, 0x0B04} /* Cn */, { 0x0B0D, 0x0B0E} /* Cn */,
872: { 0x0B11, 0x0B12} /* Cn */, { 0x0B29, 0x0B29} /* Cn */,
873: { 0x0B31, 0x0B31} /* Cn */, { 0x0B34, 0x0B34} /* Cn */,
874: { 0x0B3A, 0x0B3B} /* Cn */, { 0x0B44, 0x0B46} /* Cn */,
875: { 0x0B49, 0x0B4A} /* Cn */, { 0x0B4E, 0x0B55} /* Cn */,
876: { 0x0B58, 0x0B5B} /* Cn */, { 0x0B5E, 0x0B5E} /* Cn */,
877: { 0x0B62, 0x0B65} /* Cn */, { 0x0B72, 0x0B81} /* Cn */,
878: { 0x0B84, 0x0B84} /* Cn */, { 0x0B8B, 0x0B8D} /* Cn */,
879: { 0x0B91, 0x0B91} /* Cn */, { 0x0B96, 0x0B98} /* Cn */,
880: { 0x0B9B, 0x0B9B} /* Cn */, { 0x0B9D, 0x0B9D} /* Cn */,
881: { 0x0BA0, 0x0BA2} /* Cn */, { 0x0BA5, 0x0BA7} /* Cn */,
882: { 0x0BAB, 0x0BAD} /* Cn */, { 0x0BBA, 0x0BBD} /* Cn */,
883: { 0x0BC3, 0x0BC5} /* Cn */, { 0x0BC9, 0x0BC9} /* Cn */,
884: { 0x0BCE, 0x0BD6} /* Cn */, { 0x0BD8, 0x0BE5} /* Cn */,
885: { 0x0BFB, 0x0C00} /* Cn */, { 0x0C04, 0x0C04} /* Cn */,
886: { 0x0C0D, 0x0C0D} /* Cn */, { 0x0C11, 0x0C11} /* Cn */,
887: { 0x0C29, 0x0C29} /* Cn */, { 0x0C34, 0x0C34} /* Cn */,
888: { 0x0C3A, 0x0C3D} /* Cn */, { 0x0C45, 0x0C45} /* Cn */,
889: { 0x0C49, 0x0C49} /* Cn */, { 0x0C4E, 0x0C54} /* Cn */,
890: { 0x0C57, 0x0C5F} /* Cn */, { 0x0C62, 0x0C65} /* Cn */,
891: { 0x0C70, 0x0C81} /* Cn */, { 0x0C84, 0x0C84} /* Cn */,
892: { 0x0C8D, 0x0C8D} /* Cn */, { 0x0C91, 0x0C91} /* Cn */,
893: { 0x0CA9, 0x0CA9} /* Cn */, { 0x0CB4, 0x0CB4} /* Cn */,
894: { 0x0CBA, 0x0CBB} /* Cn */, { 0x0CC5, 0x0CC5} /* Cn */,
895: { 0x0CC9, 0x0CC9} /* Cn */, { 0x0CCE, 0x0CD4} /* Cn */,
896: { 0x0CD7, 0x0CDD} /* Cn */, { 0x0CDF, 0x0CDF} /* Cn */,
897: { 0x0CE4, 0x0CE5} /* Cn */, { 0x0CF0, 0x0CF0} /* Cn */,
898: { 0x0CF3, 0x0D01} /* Cn */, { 0x0D04, 0x0D04} /* Cn */,
899: { 0x0D0D, 0x0D0D} /* Cn */, { 0x0D11, 0x0D11} /* Cn */,
900: { 0x0D29, 0x0D29} /* Cn */, { 0x0D3A, 0x0D3D} /* Cn */,
901: { 0x0D44, 0x0D45} /* Cn */, { 0x0D49, 0x0D49} /* Cn */,
902: { 0x0D4E, 0x0D56} /* Cn */, { 0x0D58, 0x0D5F} /* Cn */,
903: { 0x0D62, 0x0D65} /* Cn */, { 0x0D70, 0x0D81} /* Cn */,
904: { 0x0D84, 0x0D84} /* Cn */, { 0x0D97, 0x0D99} /* Cn */,
905: { 0x0DB2, 0x0DB2} /* Cn */, { 0x0DBC, 0x0DBC} /* Cn */,
906: { 0x0DBE, 0x0DBF} /* Cn */, { 0x0DC7, 0x0DC9} /* Cn */,
907: { 0x0DCB, 0x0DCE} /* Cn */, { 0x0DD5, 0x0DD5} /* Cn */,
908: { 0x0DD7, 0x0DD7} /* Cn */, { 0x0DE0, 0x0DF1} /* Cn */,
909: { 0x0DF5, 0x0E00} /* Cn */, { 0x0E3B, 0x0E3E} /* Cn */,
910: { 0x0E5C, 0x0E80} /* Cn */, { 0x0E83, 0x0E83} /* Cn */,
911: { 0x0E85, 0x0E86} /* Cn */, { 0x0E89, 0x0E89} /* Cn */,
912: { 0x0E8B, 0x0E8C} /* Cn */, { 0x0E8E, 0x0E93} /* Cn */,
913: { 0x0E98, 0x0E98} /* Cn */, { 0x0EA0, 0x0EA0} /* Cn */,
914: { 0x0EA4, 0x0EA4} /* Cn */, { 0x0EA6, 0x0EA6} /* Cn */,
915: { 0x0EA8, 0x0EA9} /* Cn */, { 0x0EAC, 0x0EAC} /* Cn */,
916: { 0x0EBA, 0x0EBA} /* Cn */, { 0x0EBE, 0x0EBF} /* Cn */,
917: { 0x0EC5, 0x0EC5} /* Cn */, { 0x0EC7, 0x0EC7} /* Cn */,
918: { 0x0ECE, 0x0ECF} /* Cn */, { 0x0EDA, 0x0EDB} /* Cn */,
919: { 0x0EDE, 0x0EFF} /* Cn */, { 0x0F48, 0x0F48} /* Cn */,
920: { 0x0F6B, 0x0F70} /* Cn */, { 0x0F8C, 0x0F8F} /* Cn */,
921: { 0x0F98, 0x0F98} /* Cn */, { 0x0FBD, 0x0FBD} /* Cn */,
922: { 0x0FCD, 0x0FCE} /* Cn */, { 0x0FD2, 0x0FFF} /* Cn */,
923: { 0x1022, 0x1022} /* Cn */, { 0x1028, 0x1028} /* Cn */,
924: { 0x102B, 0x102B} /* Cn */, { 0x1033, 0x1035} /* Cn */,
925: { 0x103A, 0x103F} /* Cn */, { 0x105A, 0x109F} /* Cn */,
926: { 0x10C6, 0x10CF} /* Cn */, { 0x10FD, 0x10FF} /* Cn */,
927: { 0x115A, 0x115E} /* Cn */, { 0x11A3, 0x11A7} /* Cn */,
928: { 0x11FA, 0x11FF} /* Cn */, { 0x1249, 0x1249} /* Cn */,
929: { 0x124E, 0x124F} /* Cn */, { 0x1257, 0x1257} /* Cn */,
930: { 0x1259, 0x1259} /* Cn */, { 0x125E, 0x125F} /* Cn */,
931: { 0x1289, 0x1289} /* Cn */, { 0x128E, 0x128F} /* Cn */,
932: { 0x12B1, 0x12B1} /* Cn */, { 0x12B6, 0x12B7} /* Cn */,
933: { 0x12BF, 0x12BF} /* Cn */, { 0x12C1, 0x12C1} /* Cn */,
934: { 0x12C6, 0x12C7} /* Cn */, { 0x12D7, 0x12D7} /* Cn */,
935: { 0x1311, 0x1311} /* Cn */, { 0x1316, 0x1317} /* Cn */,
936: { 0x135B, 0x135E} /* Cn */, { 0x137D, 0x137F} /* Cn */,
937: { 0x139A, 0x139F} /* Cn */, { 0x13F5, 0x1400} /* Cn */,
938: { 0x1677, 0x167F} /* Cn */, { 0x169D, 0x169F} /* Cn */,
939: { 0x16F1, 0x16FF} /* Cn */, { 0x170D, 0x170D} /* Cn */,
940: { 0x1715, 0x171F} /* Cn */, { 0x1737, 0x173F} /* Cn */,
941: { 0x1754, 0x175F} /* Cn */, { 0x176D, 0x176D} /* Cn */,
942: { 0x1771, 0x1771} /* Cn */, { 0x1774, 0x177F} /* Cn */,
943: #if 0
944: { 0x17B4, 0x17B5} /* Cf */,
945: #endif
946: { 0x17DE, 0x17DF} /* Cn */, { 0x17EA, 0x17EF} /* Cn */,
947: { 0x17FA, 0x17FF} /* Cn */, { 0x180F, 0x180F} /* Cn */,
948: { 0x181A, 0x181F} /* Cn */, { 0x1878, 0x187F} /* Cn */,
949: { 0x18AA, 0x18FF} /* Cn */, { 0x191D, 0x191F} /* Cn */,
950: { 0x192C, 0x192F} /* Cn */, { 0x193C, 0x193F} /* Cn */,
951: { 0x1941, 0x1943} /* Cn */, { 0x196E, 0x196F} /* Cn */,
952: { 0x1975, 0x197F} /* Cn */, { 0x19AA, 0x19AF} /* Cn */,
953: { 0x19CA, 0x19CF} /* Cn */, { 0x19DA, 0x19DD} /* Cn */,
954: { 0x1A1C, 0x1A1D} /* Cn */, { 0x1A20, 0x1AFF} /* Cn */,
955: { 0x1B4C, 0x1B4F} /* Cn */, { 0x1B7D, 0x1CFF} /* Cn */,
956: { 0x1DCB, 0x1DFD} /* Cn */, { 0x1E9C, 0x1E9F} /* Cn */,
957: { 0x1EFA, 0x1EFF} /* Cn */, { 0x1F16, 0x1F17} /* Cn */,
958: { 0x1F1E, 0x1F1F} /* Cn */, { 0x1F46, 0x1F47} /* Cn */,
959: { 0x1F4E, 0x1F4F} /* Cn */, { 0x1F58, 0x1F58} /* Cn */,
960: { 0x1F5A, 0x1F5A} /* Cn */, { 0x1F5C, 0x1F5C} /* Cn */,
961: { 0x1F5E, 0x1F5E} /* Cn */, { 0x1F7E, 0x1F7F} /* Cn */,
962: { 0x1FB5, 0x1FB5} /* Cn */, { 0x1FC5, 0x1FC5} /* Cn */,
963: { 0x1FD4, 0x1FD5} /* Cn */, { 0x1FDC, 0x1FDC} /* Cn */,
964: { 0x1FF0, 0x1FF1} /* Cn */, { 0x1FF5, 0x1FF5} /* Cn */,
965: { 0x1FFF, 0x1FFF} /* Cn */,
966: { 0x200B, 0x200F} /* Cf */,
967: { 0x2028, 0x2028} /* Zl */,
968: { 0x2029, 0x2029} /* Zp */,
969: { 0x202A, 0x202E} /* Cf */,
970: { 0x2060, 0x2063} /* Cf */,
971: { 0x2064, 0x2069} /* Cn */,
972: { 0x206A, 0x206F} /* Cf */,
973: { 0x2072, 0x2073} /* Cn */, { 0x208F, 0x208F} /* Cn */,
974: { 0x2095, 0x209F} /* Cn */, { 0x20B6, 0x20CF} /* Cn */,
975: { 0x20F0, 0x20FF} /* Cn */, { 0x214F, 0x2152} /* Cn */,
976: { 0x2185, 0x218F} /* Cn */, { 0x23E8, 0x23FF} /* Cn */,
977: { 0x2427, 0x243F} /* Cn */, { 0x244B, 0x245F} /* Cn */,
978: { 0x269D, 0x269F} /* Cn */, { 0x26B3, 0x2700} /* Cn */,
979: { 0x2705, 0x2705} /* Cn */, { 0x270A, 0x270B} /* Cn */,
980: { 0x2728, 0x2728} /* Cn */, { 0x274C, 0x274C} /* Cn */,
981: { 0x274E, 0x274E} /* Cn */, { 0x2753, 0x2755} /* Cn */,
982: { 0x2757, 0x2757} /* Cn */, { 0x275F, 0x2760} /* Cn */,
983: { 0x2795, 0x2797} /* Cn */, { 0x27B0, 0x27B0} /* Cn */,
984: { 0x27BF, 0x27BF} /* Cn */, { 0x27CB, 0x27CF} /* Cn */,
985: { 0x27EC, 0x27EF} /* Cn */, { 0x2B1B, 0x2B1F} /* Cn */,
986: { 0x2B24, 0x2BFF} /* Cn */, { 0x2C2F, 0x2C2F} /* Cn */,
987: { 0x2C5F, 0x2C5F} /* Cn */, { 0x2C6D, 0x2C73} /* Cn */,
988: { 0x2C78, 0x2C7F} /* Cn */, { 0x2CEB, 0x2CF8} /* Cn */,
989: { 0x2D26, 0x2D2F} /* Cn */, { 0x2D66, 0x2D6E} /* Cn */,
990: { 0x2D70, 0x2D7F} /* Cn */, { 0x2D97, 0x2D9F} /* Cn */,
991: { 0x2DA7, 0x2DA7} /* Cn */, { 0x2DAF, 0x2DAF} /* Cn */,
992: { 0x2DB7, 0x2DB7} /* Cn */, { 0x2DBF, 0x2DBF} /* Cn */,
993: { 0x2DC7, 0x2DC7} /* Cn */, { 0x2DCF, 0x2DCF} /* Cn */,
994: { 0x2DD7, 0x2DD7} /* Cn */, { 0x2DDF, 0x2DFF} /* Cn */,
995: { 0x2E18, 0x2E1B} /* Cn */, { 0x2E1E, 0x2E7F} /* Cn */,
996: { 0x2E9A, 0x2E9A} /* Cn */, { 0x2EF4, 0x2EFF} /* Cn */,
997: { 0x2FD6, 0x2FEF} /* Cn */, { 0x2FFC, 0x2FFF} /* Cn */,
998: { 0x3040, 0x3040} /* Cn */, { 0x3097, 0x3098} /* Cn */,
999: { 0x3100, 0x3104} /* Cn */, { 0x312D, 0x3130} /* Cn */,
1000: { 0x318F, 0x318F} /* Cn */, { 0x31B8, 0x31BF} /* Cn */,
1001: { 0x31D0, 0x31EF} /* Cn */, { 0x321F, 0x321F} /* Cn */,
1002: { 0x3244, 0x324F} /* Cn */, { 0x32FF, 0x32FF} /* Cn */,
1003: { 0x4DB6, 0x4DBF} /* Cn */, { 0x9FBC, 0x9FFF} /* Cn */,
1004: { 0xA48D, 0xA48F} /* Cn */, { 0xA4C7, 0xA6FF} /* Cn */,
1005: { 0xA71B, 0xA71F} /* Cn */, { 0xA722, 0xA7FF} /* Cn */,
1006: { 0xA82C, 0xA83F} /* Cn */, { 0xA878, 0xABFF} /* Cn */,
1007: { 0xD7A4, 0xD7FF} /* Cn */,
1008: { 0xD800, 0xDFFF} /* Cs */,
1009: { 0xE000, 0xF8FF} /* Co */,
1010: { 0xFA2E, 0xFA2F} /* Cn */, { 0xFA6B, 0xFA6F} /* Cn */,
1011: { 0xFADA, 0xFAFF} /* Cn */, { 0xFB07, 0xFB12} /* Cn */,
1012: { 0xFB18, 0xFB1C} /* Cn */, { 0xFB37, 0xFB37} /* Cn */,
1013: { 0xFB3D, 0xFB3D} /* Cn */, { 0xFB3F, 0xFB3F} /* Cn */,
1014: { 0xFB42, 0xFB42} /* Cn */, { 0xFB45, 0xFB45} /* Cn */,
1015: { 0xFBB2, 0xFBD2} /* Cn */, { 0xFD40, 0xFD4F} /* Cn */,
1016: { 0xFD90, 0xFD91} /* Cn */, { 0xFDC8, 0xFDEF} /* Cn */,
1017: { 0xFDFE, 0xFDFF} /* Cn */, { 0xFE1A, 0xFE1F} /* Cn */,
1018: { 0xFE24, 0xFE2F} /* Cn */, { 0xFE53, 0xFE53} /* Cn */,
1019: { 0xFE67, 0xFE67} /* Cn */, { 0xFE6C, 0xFE6F} /* Cn */,
1020: { 0xFE75, 0xFE75} /* Cn */, { 0xFEFD, 0xFEFE} /* Cn */,
1021: { 0xFEFF, 0xFEFF} /* Cf */,
1022: { 0xFF00, 0xFF00} /* Cn */, { 0xFFBF, 0xFFC1} /* Cn */,
1023: { 0xFFC8, 0xFFC9} /* Cn */, { 0xFFD0, 0xFFD1} /* Cn */,
1024: { 0xFFD8, 0xFFD9} /* Cn */, { 0xFFDD, 0xFFDF} /* Cn */,
1025: { 0xFFE7, 0xFFE7} /* Cn */, { 0xFFEF, 0xFFF8} /* Cn */,
1026: { 0xFFF9, 0xFFFB} /* Cf */,
1027: { 0xFFFE, 0xFFFF} /* Cn */, { 0x1000C, 0x1000C} /* Cn */,
1028: { 0x10027, 0x10027} /* Cn */, { 0x1003B, 0x1003B} /* Cn */,
1029: { 0x1003E, 0x1003E} /* Cn */, { 0x1004E, 0x1004F} /* Cn */,
1030: { 0x1005E, 0x1007F} /* Cn */, { 0x100FB, 0x100FF} /* Cn */,
1031: { 0x10103, 0x10106} /* Cn */, { 0x10134, 0x10136} /* Cn */,
1032: { 0x1018B, 0x102FF} /* Cn */, { 0x1031F, 0x1031F} /* Cn */,
1033: { 0x10324, 0x1032F} /* Cn */, { 0x1034B, 0x1037F} /* Cn */,
1034: { 0x1039E, 0x1039E} /* Cn */, { 0x103C4, 0x103C7} /* Cn */,
1035: { 0x103D6, 0x103FF} /* Cn */,
1036: { 0x1049E, 0x1049F} /* Cn */, { 0x104AA, 0x107FF} /* Cn */,
1037: { 0x10806, 0x10807} /* Cn */, { 0x10809, 0x10809} /* Cn */,
1038: { 0x10836, 0x10836} /* Cn */, { 0x10839, 0x1083B} /* Cn */,
1039: { 0x1083D, 0x1083E} /* Cn */, { 0x10840, 0x108FF} /* Cn */,
1040: { 0x1091A, 0x1091E} /* Cn */, { 0x10920, 0x109FF} /* Cn */,
1041: { 0x10A04, 0x10A04} /* Cn */, { 0x10A07, 0x10A0B} /* Cn */,
1042: { 0x10A14, 0x10A14} /* Cn */, { 0x10A18, 0x10A18} /* Cn */,
1043: { 0x10A34, 0x10A37} /* Cn */, { 0x10A3B, 0x10A3E} /* Cn */,
1044: { 0x10A48, 0x10A4F} /* Cn */, { 0x10A59, 0x11FFF} /* Cn */,
1045: { 0x1236F, 0x123FF} /* Cn */, { 0x12463, 0x1246F} /* Cn */,
1046: { 0x12474, 0x1CFFF} /* Cn */, { 0x1D0F6, 0x1D0FF} /* Cn */,
1047: { 0x1D127, 0x1D129} /* Cn */,
1048: { 0x1D173, 0x1D17A} /* Cf */,
1049: { 0x1D1DE, 0x1D1FF} /* Cn */, { 0x1D246, 0x1D2FF} /* Cn */,
1050: { 0x1D357, 0x1D35F} /* Cn */, { 0x1D372, 0x1D3FF} /* Cn */,
1051: { 0x1D455, 0x1D455} /* Cn */, { 0x1D49D, 0x1D49D} /* Cn */,
1052: { 0x1D4A0, 0x1D4A1} /* Cn */, { 0x1D4A3, 0x1D4A4} /* Cn */,
1053: { 0x1D4A7, 0x1D4A8} /* Cn */, { 0x1D4AD, 0x1D4AD} /* Cn */,
1054: { 0x1D4BA, 0x1D4BA} /* Cn */, { 0x1D4BC, 0x1D4BC} /* Cn */,
1055: { 0x1D4C4, 0x1D4C4} /* Cn */, { 0x1D506, 0x1D506} /* Cn */,
1056: { 0x1D50B, 0x1D50C} /* Cn */, { 0x1D515, 0x1D515} /* Cn */,
1057: { 0x1D51D, 0x1D51D} /* Cn */, { 0x1D53A, 0x1D53A} /* Cn */,
1058: { 0x1D53F, 0x1D53F} /* Cn */, { 0x1D545, 0x1D545} /* Cn */,
1059: { 0x1D547, 0x1D549} /* Cn */, { 0x1D551, 0x1D551} /* Cn */,
1060: { 0x1D6A6, 0x1D6A7} /* Cn */, { 0x1D7CC, 0x1D7CD} /* Cn */,
1061: { 0x1D800, 0x1FFFF} /* Cn */, { 0x2A6D7, 0x2F7FF} /* Cn */,
1062: { 0x2FA1E, 0xE0000} /* Cn */,
1063: { 0xE0001, 0xE0001} /* Cf */,
1064: { 0xE0002, 0xE001F} /* Cn */,
1065: { 0xE0020, 0xE007F} /* Cf */,
1066: { 0xE0080, 0xE00FF} /* Cn */, { 0xE01F0, 0xEFFFF} /* Cn */,
1067: { 0xF0000, 0xFFFFD} /* Co */,
1068: { 0xFFFFE, 0xFFFFF} /* Cn */,
1069: {0x100000,0x10FFFD} /* Co */,
1070: {0x10FFFE,0x10FFFF} /* Cn */,
1071: {0x110000,0x7FFFFFFF} /* ISO 10646?? */
1072: };
1073:
1074: /*
1075: * Double width characters
1076: * W: East Asian Wide
1077: * F: East Asian Full-width
1078: * Unassigned code points may be included when they allow ranges to be merged.
1079: * Last synched with
1080: * <http://www.unicode.org/Public/5.0.0/ucd/EastAsianWidth-5.0.0d2.txt>
1081: * dated 2005-11-08T01:32:56Z
1082: */
1083: static struct wchar_range wide_table[] = {
1084: { 0x1100, 0x115F} /* W */, { 0x2329, 0x232A} /* W */,
1085: { 0x2E80, 0x2FFB} /* W */,
1086: { 0x3000, 0x3000} /* F */,
1087: { 0x3001, 0x303E} /* W */, { 0x3041, 0x4DB5} /* W */,
1088: { 0x4E00, 0x9FBB} /* W */, { 0xA000, 0xA4C6} /* W */,
1089: { 0xAC00, 0xD7A3} /* W */, { 0xF900, 0xFAD9} /* W */,
1090: { 0xFE10, 0xFE19} /* W */, { 0xFE30, 0xFE6B} /* W */,
1091: { 0xFF01, 0xFF60} /* F */, { 0xFFE0, 0xFFE6} /* F */,
1092: { 0x20000, 0x2FFFD} /* W */, { 0x30000, 0x3FFFD} /* W */,
1093: };
1094:
1095: static int
1096: is_in_table(ch, table, tsize)
1097: LWCHAR ch;
1098: struct wchar_range table[];
1099: int tsize;
1100: {
1101: int hi;
1102: int lo;
1103:
1104: /* Binary search in the table. */
1105: if (ch < table[0].first)
1106: return 0;
1107: lo = 0;
1108: hi = tsize - 1;
1109: while (lo <= hi)
1110: {
1111: int mid = (lo + hi) / 2;
1112: if (ch > table[mid].last)
1113: lo = mid + 1;
1114: else if (ch < table[mid].first)
1115: hi = mid - 1;
1116: else
1117: return 1;
1118: }
1119: return 0;
1120: }
1121:
1122: /*
1123: * Is a character a UTF-8 composing character?
1124: * If a composing character follows any char, the two combine into one glyph.
1125: */
1126: public int
1127: is_composing_char(ch)
1128: LWCHAR ch;
1129: {
1130: return is_in_table(ch, comp_table, (sizeof(comp_table) / sizeof(*comp_table)));
1131: }
1132:
1133: /*
1134: * Should this UTF-8 character be treated as binary?
1135: */
1136: public int
1137: is_ubin_char(ch)
1138: LWCHAR ch;
1139: {
1140: return is_in_table(ch, ubin_table, (sizeof(ubin_table) / sizeof(*ubin_table)));
1141: }
1142:
1143: /*
1144: * Is this a double width UTF-8 character?
1145: */
1146: public int
1147: is_wide_char(ch)
1148: LWCHAR ch;
1149: {
1150: return is_in_table(ch, wide_table, (sizeof(wide_table) / sizeof(*wide_table)));
1151: }
1152:
1153: /*
1154: * Is a character a UTF-8 combining character?
1155: * A combining char acts like an ordinary char, but if it follows
1156: * a specific char (not any char), the two combine into one glyph.
1157: */
1158: public int
1159: is_combining_char(ch1, ch2)
1160: LWCHAR ch1;
1161: LWCHAR ch2;
1162: {
1163: /* The table is small; use linear search. */
1164: int i;
1165: for (i = 0; i < sizeof(comb_table)/sizeof(*comb_table); i++)
1166: {
1167: if (ch1 == comb_table[i].first &&
1168: ch2 == comb_table[i].last)
1169: return 1;
1170: }
1171: return 0;
1172: }
1173:
1.1.1.4 ! shadchin 1174: #else /* !SMALL */
! 1175:
! 1176: public int binattr = AT_STANDOUT;
! 1177:
! 1178: public void
! 1179: init_charset()
! 1180: {
! 1181: return;
! 1182: }
! 1183:
! 1184: /*
! 1185: * Is a given character a "binary" character?
! 1186: */
! 1187: public int
! 1188: binary_char(c)
! 1189: LWCHAR c;
! 1190: {
! 1191: return (!isprint(c) && !isspace(c));
! 1192: }
! 1193:
! 1194: /*
! 1195: * Is a given character a "control" character?
! 1196: */
! 1197: public int
! 1198: control_char(c)
! 1199: LWCHAR c;
! 1200: {
! 1201: return (iscntrl(c));
! 1202: }
! 1203:
! 1204: /*
! 1205: * Return the printable form of a character.
! 1206: * For example, in the "ascii" charset '\3' is printed as "^C".
! 1207: */
! 1208: public char *
! 1209: prchar(c)
! 1210: LWCHAR c;
! 1211: {
! 1212: static char buf[8];
! 1213:
! 1214: c &= 0377;
! 1215: if (!iscntrl(c))
! 1216: snprintf(buf, sizeof(buf), "%c", c);
! 1217: else if (c == ESC)
! 1218: strlcpy(buf, "ESC", sizeof(buf));
! 1219: else if (c < 128 && !iscntrl(c ^ 0100))
! 1220: snprintf(buf, sizeof(buf), "^%c", c ^ 0100);
! 1221: else
! 1222: snprintf(buf, sizeof(buf), "*s<%X>", c);
! 1223: return (buf);
! 1224: }
! 1225:
! 1226: /*
! 1227: * Step forward or backward one character in a string.
! 1228: */
! 1229: public LWCHAR
! 1230: step_char(pp, dir, limit)
! 1231: char **pp;
! 1232: signed int dir;
! 1233: char *limit;
! 1234: {
! 1235: LWCHAR ch;
! 1236: int len;
! 1237: char *p = *pp;
! 1238:
! 1239: /* It's easy if chars are one byte. */
! 1240: if (dir > 0)
! 1241: ch = (LWCHAR) ((p < limit) ? *p++ : 0);
! 1242: else
! 1243: ch = (LWCHAR) ((p > limit) ? *--p : 0);
! 1244:
! 1245: *pp = p;
! 1246: return ch;
! 1247: }
! 1248:
! 1249: #endif /* !SMALL */
![[BACK]](/icons/back.gif){kind=icon}
Return to charset.c CVS log ![[TXT]](/icons/text.gif){kind=icon}
![[DIR]](/icons/dir.gif){kind=icon}
Up to [local] / src / usr.bin / less