1 /* guniprop.c - Unicode character properties.
3 * Copyright (C) 1999 Tom Tromey
4 * Copyright (C) 2000 Red Hat, Inc.
6 * This library is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2 of the License, or (at your option) any later version.
11 * This library is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 * Lesser General Public License for more details.
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with this library; if not, write to the
18 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
19 * Boston, MA 02111-1307, USA.
29 #include "gunichartables.h"
30 #include "gmirroringtable.h"
31 #include "gunicodeprivate.h"
34 #define ATTR_TABLE(Page) (((Page) <= G_UNICODE_LAST_PAGE_PART1) \
35 ? attr_table_part1[Page] \
36 : attr_table_part2[(Page) - 0xe00])
38 #define ATTTABLE(Page, Char) \
39 ((ATTR_TABLE(Page) == G_UNICODE_MAX_TABLE_INDEX) ? 0 : (attr_data[ATTR_TABLE(Page)][Char]))
41 #define TTYPE_PART1(Page, Char) \
42 ((type_table_part1[Page] >= G_UNICODE_MAX_TABLE_INDEX) \
43 ? (type_table_part1[Page] - G_UNICODE_MAX_TABLE_INDEX) \
44 : (type_data[type_table_part1[Page]][Char]))
46 #define TTYPE_PART2(Page, Char) \
47 ((type_table_part2[Page] >= G_UNICODE_MAX_TABLE_INDEX) \
48 ? (type_table_part2[Page] - G_UNICODE_MAX_TABLE_INDEX) \
49 : (type_data[type_table_part2[Page]][Char]))
52 (((Char) <= G_UNICODE_LAST_CHAR_PART1) \
53 ? TTYPE_PART1 ((Char) >> 8, (Char) & 0xff) \
54 : (((Char) >= 0xe0000 && (Char) <= G_UNICODE_LAST_CHAR) \
55 ? TTYPE_PART2 (((Char) - 0xe0000) >> 8, (Char) & 0xff) \
56 : G_UNICODE_UNASSIGNED))
59 #define IS(Type, Class) (((guint)1 << (Type)) & (Class))
60 #define OR(Type, Rest) (((guint)1 << (Type)) | (Rest))
64 #define ISDIGIT(Type) IS ((Type), \
65 OR (G_UNICODE_DECIMAL_NUMBER, \
66 OR (G_UNICODE_LETTER_NUMBER, \
67 OR (G_UNICODE_OTHER_NUMBER, 0))))
69 #define ISALPHA(Type) IS ((Type), \
70 OR (G_UNICODE_LOWERCASE_LETTER, \
71 OR (G_UNICODE_UPPERCASE_LETTER, \
72 OR (G_UNICODE_TITLECASE_LETTER, \
73 OR (G_UNICODE_MODIFIER_LETTER, \
74 OR (G_UNICODE_OTHER_LETTER, 0))))))
76 #define ISALDIGIT(Type) IS ((Type), \
77 OR (G_UNICODE_DECIMAL_NUMBER, \
78 OR (G_UNICODE_LETTER_NUMBER, \
79 OR (G_UNICODE_OTHER_NUMBER, \
80 OR (G_UNICODE_LOWERCASE_LETTER, \
81 OR (G_UNICODE_UPPERCASE_LETTER, \
82 OR (G_UNICODE_TITLECASE_LETTER, \
83 OR (G_UNICODE_MODIFIER_LETTER, \
84 OR (G_UNICODE_OTHER_LETTER, 0)))))))))
86 #define ISMARK(Type) IS ((Type), \
87 OR (G_UNICODE_NON_SPACING_MARK, \
88 OR (G_UNICODE_COMBINING_MARK, \
89 OR (G_UNICODE_ENCLOSING_MARK, 0))))
93 * @c: a Unicode character
95 * Determines whether a character is alphanumeric.
96 * Given some UTF-8 text, obtain a character value
97 * with g_utf8_get_char().
99 * Return value: %TRUE if @c is an alphanumeric character
102 g_unichar_isalnum (gunichar c)
104 return ISALDIGIT (TYPE (c)) ? TRUE : FALSE;
109 * @c: a Unicode character
111 * Determines whether a character is alphabetic (i.e. a letter).
112 * Given some UTF-8 text, obtain a character value with
115 * Return value: %TRUE if @c is an alphabetic character
118 g_unichar_isalpha (gunichar c)
120 return ISALPHA (TYPE (c)) ? TRUE : FALSE;
126 * @c: a Unicode character
128 * Determines whether a character is a control character.
129 * Given some UTF-8 text, obtain a character value with
132 * Return value: %TRUE if @c is a control character
135 g_unichar_iscntrl (gunichar c)
137 return TYPE (c) == G_UNICODE_CONTROL;
142 * @c: a Unicode character
144 * Determines whether a character is numeric (i.e. a digit). This
145 * covers ASCII 0-9 and also digits in other languages/scripts. Given
146 * some UTF-8 text, obtain a character value with g_utf8_get_char().
148 * Return value: %TRUE if @c is a digit
151 g_unichar_isdigit (gunichar c)
153 return TYPE (c) == G_UNICODE_DECIMAL_NUMBER;
159 * @c: a Unicode character
161 * Determines whether a character is printable and not a space
162 * (returns %FALSE for control characters, format characters, and
163 * spaces). g_unichar_isprint() is similar, but returns %TRUE for
164 * spaces. Given some UTF-8 text, obtain a character value with
167 * Return value: %TRUE if @c is printable unless it's a space
170 g_unichar_isgraph (gunichar c)
173 OR (G_UNICODE_CONTROL,
174 OR (G_UNICODE_FORMAT,
175 OR (G_UNICODE_UNASSIGNED,
176 OR (G_UNICODE_PRIVATE_USE,
177 OR (G_UNICODE_SURROGATE,
178 OR (G_UNICODE_SPACE_SEPARATOR,
184 * @c: a Unicode character
186 * Determines whether a character is a lowercase letter.
187 * Given some UTF-8 text, obtain a character value with
190 * Return value: %TRUE if @c is a lowercase letter
193 g_unichar_islower (gunichar c)
195 return TYPE (c) == G_UNICODE_LOWERCASE_LETTER;
201 * @c: a Unicode character
203 * Determines whether a character is printable.
204 * Unlike g_unichar_isgraph(), returns %TRUE for spaces.
205 * Given some UTF-8 text, obtain a character value with
208 * Return value: %TRUE if @c is printable
211 g_unichar_isprint (gunichar c)
214 OR (G_UNICODE_CONTROL,
215 OR (G_UNICODE_FORMAT,
216 OR (G_UNICODE_UNASSIGNED,
217 OR (G_UNICODE_PRIVATE_USE,
218 OR (G_UNICODE_SURROGATE,
224 * @c: a Unicode character
226 * Determines whether a character is punctuation or a symbol.
227 * Given some UTF-8 text, obtain a character value with
230 * Return value: %TRUE if @c is a punctuation or symbol character
233 g_unichar_ispunct (gunichar c)
236 OR (G_UNICODE_CONNECT_PUNCTUATION,
237 OR (G_UNICODE_DASH_PUNCTUATION,
238 OR (G_UNICODE_CLOSE_PUNCTUATION,
239 OR (G_UNICODE_FINAL_PUNCTUATION,
240 OR (G_UNICODE_INITIAL_PUNCTUATION,
241 OR (G_UNICODE_OTHER_PUNCTUATION,
242 OR (G_UNICODE_OPEN_PUNCTUATION,
243 OR (G_UNICODE_CURRENCY_SYMBOL,
244 OR (G_UNICODE_MODIFIER_SYMBOL,
245 OR (G_UNICODE_MATH_SYMBOL,
246 OR (G_UNICODE_OTHER_SYMBOL,
247 0)))))))))))) ? TRUE : FALSE;
252 * @c: a Unicode character
254 * Determines whether a character is a space, tab, or line separator
255 * (newline, carriage return, etc.). Given some UTF-8 text, obtain a
256 * character value with g_utf8_get_char().
258 * (Note: don't use this to do word breaking; you have to use
259 * Pango or equivalent to get word breaking right, the algorithm
260 * is fairly complex.)
262 * Return value: %TRUE if @c is a space character
265 g_unichar_isspace (gunichar c)
269 /* special-case these since Unicode thinks they are not spaces */
280 OR (G_UNICODE_SPACE_SEPARATOR,
281 OR (G_UNICODE_LINE_SEPARATOR,
282 OR (G_UNICODE_PARAGRAPH_SEPARATOR,
283 0)))) ? TRUE : FALSE;
291 * @c: a Unicode character
293 * Determines if a character is uppercase.
295 * Return value: %TRUE if @c is an uppercase character
298 g_unichar_isupper (gunichar c)
300 return TYPE (c) == G_UNICODE_UPPERCASE_LETTER;
305 * @c: a Unicode character
307 * Determines if a character is titlecase. Some characters in
308 * Unicode which are composites, such as the DZ digraph
309 * have three case variants instead of just two. The titlecase
310 * form is used at the beginning of a word where only the
311 * first letter is capitalized. The titlecase form of the DZ
312 * digraph is U+01F2 LATIN CAPITAL LETTTER D WITH SMALL LETTER Z.
314 * Return value: %TRUE if the character is titlecase
317 g_unichar_istitle (gunichar c)
320 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
321 if (title_table[i][0] == c)
327 * g_unichar_isxdigit:
328 * @c: a Unicode character.
330 * Determines if a character is a hexidecimal digit.
332 * Return value: %TRUE if the character is a hexadecimal digit
335 g_unichar_isxdigit (gunichar c)
337 return ((c >= 'a' && c <= 'f')
338 || (c >= 'A' && c <= 'F')
339 || ISDIGIT (TYPE (c)));
343 * g_unichar_isdefined:
344 * @c: a Unicode character
346 * Determines if a given character is assigned in the Unicode
349 * Return value: %TRUE if the character has an assigned value
352 g_unichar_isdefined (gunichar c)
354 return TYPE (c) != G_UNICODE_UNASSIGNED;
359 * @c: a Unicode character
361 * Determines if a character is typically rendered in a double-width
364 * Return value: %TRUE if the character is wide
366 /* This function stolen from Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>. */
368 g_unichar_iswide (gunichar c)
373 return (c <= 0x115f /* Hangul Jamo init. consonants */
374 || c == 0x2329 || c == 0x232a /* angle brackets */
375 || (c >= 0x2e80 && c <= 0xa4cf && (c < 0x302a || c > 0x302f)
376 && c != 0x303f && c != 0x3099 && c!= 0x309a) /* CJK ... Yi */
377 || (c >= 0xac00 && c <= 0xd7a3) /* Hangul Syllables */
378 || (c >= 0xf900 && c <= 0xfaff) /* CJK Compatibility Ideographs */
379 || (c >= 0xfe30 && c <= 0xfe6f) /* CJK Compatibility Forms */
380 || (c >= 0xff00 && c <= 0xff60) /* Fullwidth Forms */
381 || (c >= 0xffe0 && c <= 0xffe6) /* Fullwidth Forms */
382 || (c >= 0x20000 && c <= 0x2fffd) /* CJK extra stuff */
383 || (c >= 0x30000 && c <= 0x3fffd));
393 interval_compare (const void *key, const void *elt)
395 gunichar c = GPOINTER_TO_UINT (key);
396 struct Interval *interval = elt;
398 if (c < interval->start)
400 if (c > interval->end)
407 * g_unichar_iswide_cjk:
408 * @c: a Unicode character
410 * Determines if a character is typically rendered in a double-width
411 * cell under legacy East Asian locales. If a character is wide according to
412 * g_unichar_iswide(), then it is also reported wide with this function, but
413 * the converse is not necessarily true. See the
414 * <ulink url="http://www.unicode.org/reports/tr11/">Unicode Standard
415 * Annex #11</ulink> for details:
417 * Return value: %TRUE if the character is wide in legacy East Asian locales
419 /* This function stolen from Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>. */
421 g_unichar_iswide_cjk (gunichar c)
423 /* sorted list of non-overlapping intervals of East Asian Ambiguous
424 * characters, generated by "uniset +WIDTH-A -cat=Me -cat=Mn -cat=Cf c" */
425 static const struct Interval ambiguous[] = {
426 { 0x00A1, 0x00A1 }, { 0x00A4, 0x00A4 }, { 0x00A7, 0x00A8 },
427 { 0x00AA, 0x00AA }, { 0x00AE, 0x00AE }, { 0x00B0, 0x00B4 },
428 { 0x00B6, 0x00BA }, { 0x00BC, 0x00BF }, { 0x00C6, 0x00C6 },
429 { 0x00D0, 0x00D0 }, { 0x00D7, 0x00D8 }, { 0x00DE, 0x00E1 },
430 { 0x00E6, 0x00E6 }, { 0x00E8, 0x00EA }, { 0x00EC, 0x00ED },
431 { 0x00F0, 0x00F0 }, { 0x00F2, 0x00F3 }, { 0x00F7, 0x00FA },
432 { 0x00FC, 0x00FC }, { 0x00FE, 0x00FE }, { 0x0101, 0x0101 },
433 { 0x0111, 0x0111 }, { 0x0113, 0x0113 }, { 0x011B, 0x011B },
434 { 0x0126, 0x0127 }, { 0x012B, 0x012B }, { 0x0131, 0x0133 },
435 { 0x0138, 0x0138 }, { 0x013F, 0x0142 }, { 0x0144, 0x0144 },
436 { 0x0148, 0x014B }, { 0x014D, 0x014D }, { 0x0152, 0x0153 },
437 { 0x0166, 0x0167 }, { 0x016B, 0x016B }, { 0x01CE, 0x01CE },
438 { 0x01D0, 0x01D0 }, { 0x01D2, 0x01D2 }, { 0x01D4, 0x01D4 },
439 { 0x01D6, 0x01D6 }, { 0x01D8, 0x01D8 }, { 0x01DA, 0x01DA },
440 { 0x01DC, 0x01DC }, { 0x0251, 0x0251 }, { 0x0261, 0x0261 },
441 { 0x02C4, 0x02C4 }, { 0x02C7, 0x02C7 }, { 0x02C9, 0x02CB },
442 { 0x02CD, 0x02CD }, { 0x02D0, 0x02D0 }, { 0x02D8, 0x02DB },
443 { 0x02DD, 0x02DD }, { 0x02DF, 0x02DF }, { 0x0391, 0x03A1 },
444 { 0x03A3, 0x03A9 }, { 0x03B1, 0x03C1 }, { 0x03C3, 0x03C9 },
445 { 0x0401, 0x0401 }, { 0x0410, 0x044F }, { 0x0451, 0x0451 },
446 { 0x2010, 0x2010 }, { 0x2013, 0x2016 }, { 0x2018, 0x2019 },
447 { 0x201C, 0x201D }, { 0x2020, 0x2022 }, { 0x2024, 0x2027 },
448 { 0x2030, 0x2030 }, { 0x2032, 0x2033 }, { 0x2035, 0x2035 },
449 { 0x203B, 0x203B }, { 0x203E, 0x203E }, { 0x2074, 0x2074 },
450 { 0x207F, 0x207F }, { 0x2081, 0x2084 }, { 0x20AC, 0x20AC },
451 { 0x2103, 0x2103 }, { 0x2105, 0x2105 }, { 0x2109, 0x2109 },
452 { 0x2113, 0x2113 }, { 0x2116, 0x2116 }, { 0x2121, 0x2122 },
453 { 0x2126, 0x2126 }, { 0x212B, 0x212B }, { 0x2153, 0x2154 },
454 { 0x215B, 0x215E }, { 0x2160, 0x216B }, { 0x2170, 0x2179 },
455 { 0x2190, 0x2199 }, { 0x21B8, 0x21B9 }, { 0x21D2, 0x21D2 },
456 { 0x21D4, 0x21D4 }, { 0x21E7, 0x21E7 }, { 0x2200, 0x2200 },
457 { 0x2202, 0x2203 }, { 0x2207, 0x2208 }, { 0x220B, 0x220B },
458 { 0x220F, 0x220F }, { 0x2211, 0x2211 }, { 0x2215, 0x2215 },
459 { 0x221A, 0x221A }, { 0x221D, 0x2220 }, { 0x2223, 0x2223 },
460 { 0x2225, 0x2225 }, { 0x2227, 0x222C }, { 0x222E, 0x222E },
461 { 0x2234, 0x2237 }, { 0x223C, 0x223D }, { 0x2248, 0x2248 },
462 { 0x224C, 0x224C }, { 0x2252, 0x2252 }, { 0x2260, 0x2261 },
463 { 0x2264, 0x2267 }, { 0x226A, 0x226B }, { 0x226E, 0x226F },
464 { 0x2282, 0x2283 }, { 0x2286, 0x2287 }, { 0x2295, 0x2295 },
465 { 0x2299, 0x2299 }, { 0x22A5, 0x22A5 }, { 0x22BF, 0x22BF },
466 { 0x2312, 0x2312 }, { 0x2460, 0x24E9 }, { 0x24EB, 0x254B },
467 { 0x2550, 0x2573 }, { 0x2580, 0x258F }, { 0x2592, 0x2595 },
468 { 0x25A0, 0x25A1 }, { 0x25A3, 0x25A9 }, { 0x25B2, 0x25B3 },
469 { 0x25B6, 0x25B7 }, { 0x25BC, 0x25BD }, { 0x25C0, 0x25C1 },
470 { 0x25C6, 0x25C8 }, { 0x25CB, 0x25CB }, { 0x25CE, 0x25D1 },
471 { 0x25E2, 0x25E5 }, { 0x25EF, 0x25EF }, { 0x2605, 0x2606 },
472 { 0x2609, 0x2609 }, { 0x260E, 0x260F }, { 0x2614, 0x2615 },
473 { 0x261C, 0x261C }, { 0x261E, 0x261E }, { 0x2640, 0x2640 },
474 { 0x2642, 0x2642 }, { 0x2660, 0x2661 }, { 0x2663, 0x2665 },
475 { 0x2667, 0x266A }, { 0x266C, 0x266D }, { 0x266F, 0x266F },
476 { 0x273D, 0x273D }, { 0x2776, 0x277F }, { 0xE000, 0xF8FF },
477 { 0xFFFD, 0xFFFD }, { 0xF0000, 0xFFFFD }, { 0x100000, 0x10FFFD }
480 if (g_unichar_iswide (c))
483 if (bsearch (GUINT_TO_POINTER (c), ambiguous, G_N_ELEMENTS (ambiguous), sizeof ambiguous[0],
493 * @c: a Unicode character
495 * Converts a character to uppercase.
497 * Return value: the result of converting @c to uppercase.
498 * If @c is not an lowercase or titlecase character,
499 * or has no upper case equivalent @c is returned unchanged.
502 g_unichar_toupper (gunichar c)
505 if (t == G_UNICODE_LOWERCASE_LETTER)
507 gunichar val = ATTTABLE (c >> 8, c & 0xff);
508 if (val >= 0x1000000)
510 const gchar *p = special_case_table + val - 0x1000000;
511 return g_utf8_get_char (p);
514 return val ? val : c;
516 else if (t == G_UNICODE_TITLECASE_LETTER)
519 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
521 if (title_table[i][0] == c)
522 return title_table[i][1];
530 * @c: a Unicode character.
532 * Converts a character to lower case.
534 * Return value: the result of converting @c to lower case.
535 * If @c is not an upperlower or titlecase character,
536 * or has no lowercase equivalent @c is returned unchanged.
539 g_unichar_tolower (gunichar c)
542 if (t == G_UNICODE_UPPERCASE_LETTER)
544 gunichar val = ATTTABLE (c >> 8, c & 0xff);
545 if (val >= 0x1000000)
547 const gchar *p = special_case_table + val - 0x1000000;
548 return g_utf8_get_char (p);
551 return val ? val : c;
553 else if (t == G_UNICODE_TITLECASE_LETTER)
556 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
558 if (title_table[i][0] == c)
559 return title_table[i][2];
567 * @c: a Unicode character
569 * Converts a character to the titlecase.
571 * Return value: the result of converting @c to titlecase.
572 * If @c is not an uppercase or lowercase character,
573 * @c is returned unchanged.
576 g_unichar_totitle (gunichar c)
579 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
581 if (title_table[i][0] == c || title_table[i][1] == c
582 || title_table[i][2] == c)
583 return title_table[i][0];
585 return (TYPE (c) == G_UNICODE_LOWERCASE_LETTER
586 ? ATTTABLE (c >> 8, c & 0xff)
591 * g_unichar_digit_value:
592 * @c: a Unicode character
594 * Determines the numeric value of a character as a decimal
597 * Return value: If @c is a decimal digit (according to
598 * g_unichar_isdigit()), its numeric value. Otherwise, -1.
601 g_unichar_digit_value (gunichar c)
603 if (TYPE (c) == G_UNICODE_DECIMAL_NUMBER)
604 return ATTTABLE (c >> 8, c & 0xff);
609 * g_unichar_xdigit_value:
610 * @c: a Unicode character
612 * Determines the numeric value of a character as a hexidecimal
615 * Return value: If @c is a hex digit (according to
616 * g_unichar_isxdigit()), its numeric value. Otherwise, -1.
619 g_unichar_xdigit_value (gunichar c)
621 if (c >= 'A' && c <= 'F')
623 if (c >= 'a' && c <= 'f')
625 if (TYPE (c) == G_UNICODE_DECIMAL_NUMBER)
626 return ATTTABLE (c >> 8, c & 0xff);
632 * @c: a Unicode character
634 * Classifies a Unicode character by type.
636 * Return value: the type of the character.
639 g_unichar_type (gunichar c)
645 * Case mapping functions
655 get_locale_type (void)
658 char *tem = g_win32_getlocale ();
665 const char *locale = setlocale (LC_CTYPE, NULL);
671 if (locale[1] == 'z')
672 return LOCALE_TURKIC;
675 if (locale[1] == 't')
676 return LOCALE_LITHUANIAN;
679 if (locale[1] == 'r')
680 return LOCALE_TURKIC;
684 return LOCALE_NORMAL;
688 output_marks (const char **p_inout,
692 const char *p = *p_inout;
697 gunichar c = g_utf8_get_char (p);
699 if (ISMARK (TYPE (c)))
701 if (!remove_dot || c != 0x307 /* COMBINING DOT ABOVE */)
702 len += g_unichar_to_utf8 (c, out_buffer ? out_buffer + len : NULL);
703 p = g_utf8_next_char (p);
714 output_special_case (gchar *out_buffer,
719 const gchar *p = special_case_table + offset;
722 if (type != G_UNICODE_TITLECASE_LETTER)
723 p = g_utf8_next_char (p);
730 memcpy (out_buffer, p, len);
736 real_toupper (const gchar *str,
739 LocaleType locale_type)
741 const gchar *p = str;
742 const char *last = NULL;
744 gboolean last_was_i = FALSE;
746 while ((max_len < 0 || p < str + max_len) && *p)
748 gunichar c = g_utf8_get_char (p);
753 p = g_utf8_next_char (p);
755 if (locale_type == LOCALE_LITHUANIAN)
763 /* Nasty, need to remove any dot above. Though
764 * I think only E WITH DOT ABOVE occurs in practice
765 * which could simplify this considerably.
770 decomp = g_unicode_canonical_decomposition (c, &decomp_len);
771 for (i=0; i < decomp_len; i++)
773 if (decomp[i] != 0x307 /* COMBINING DOT ABOVE */)
774 len += g_unichar_to_utf8 (g_unichar_toupper (decomp[i]), out_buffer ? out_buffer + len : NULL);
778 len += output_marks (&p, out_buffer ? out_buffer + len : NULL, TRUE);
788 if (locale_type == LOCALE_TURKIC && c == 'i')
790 /* i => LATIN CAPITAL LETTER I WITH DOT ABOVE */
791 len += g_unichar_to_utf8 (0x130, out_buffer ? out_buffer + len : NULL);
793 else if (c == 0x0345) /* COMBINING GREEK YPOGEGRAMMENI */
795 /* Nasty, need to move it after other combining marks .. this would go away if
796 * we normalized first.
798 len += output_marks (&p, out_buffer ? out_buffer + len : NULL, FALSE);
800 /* And output as GREEK CAPITAL LETTER IOTA */
801 len += g_unichar_to_utf8 (0x399, out_buffer ? out_buffer + len : NULL);
804 OR (G_UNICODE_LOWERCASE_LETTER,
805 OR (G_UNICODE_TITLECASE_LETTER,
808 val = ATTTABLE (c >> 8, c & 0xff);
810 if (val >= 0x1000000)
812 len += output_special_case (out_buffer ? out_buffer + len : NULL, val - 0x1000000, t,
813 t == G_UNICODE_LOWERCASE_LETTER ? 0 : 1);
817 if (t == G_UNICODE_TITLECASE_LETTER)
820 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
822 if (title_table[i][0] == c)
823 val = title_table[i][1];
827 len += g_unichar_to_utf8 (val, out_buffer ? out_buffer + len : NULL);
832 gsize char_len = g_utf8_skip[*(guchar *)last];
835 memcpy (out_buffer + len, last, char_len);
847 * @str: a UTF-8 encoded string
848 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
850 * Converts all Unicode characters in the string that have a case
851 * to uppercase. The exact manner that this is done depends
852 * on the current locale, and may result in the number of
853 * characters in the string increasing. (For instance, the
854 * German ess-zet will be changed to SS.)
856 * Return value: a newly allocated string, with all characters
857 * converted to uppercase.
860 g_utf8_strup (const gchar *str,
864 LocaleType locale_type;
867 g_return_val_if_fail (str != NULL, NULL);
869 locale_type = get_locale_type ();
872 * We use a two pass approach to keep memory management simple
874 result_len = real_toupper (str, len, NULL, locale_type);
875 result = g_malloc (result_len + 1);
876 real_toupper (str, len, result, locale_type);
877 result[result_len] = '\0';
882 /* traverses the string checking for characters with combining class == 230
883 * until a base character is found */
885 has_more_above (const gchar *str)
887 const gchar *p = str;
888 gint combining_class;
892 combining_class = _g_unichar_combining_class (g_utf8_get_char (p));
893 if (combining_class == 230)
895 else if (combining_class == 0)
898 p = g_utf8_next_char (p);
905 real_tolower (const gchar *str,
908 LocaleType locale_type)
910 const gchar *p = str;
911 const char *last = NULL;
914 while ((max_len < 0 || p < str + max_len) && *p)
916 gunichar c = g_utf8_get_char (p);
921 p = g_utf8_next_char (p);
923 if (locale_type == LOCALE_TURKIC && c == 'I')
925 if (g_utf8_get_char (p) == 0x0307)
927 /* I + COMBINING DOT ABOVE => i (U+0069) */
928 len += g_unichar_to_utf8 (0x0069, out_buffer ? out_buffer + len : NULL);
929 p = g_utf8_next_char (p);
933 /* I => LATIN SMALL LETTER DOTLESS I */
934 len += g_unichar_to_utf8 (0x131, out_buffer ? out_buffer + len : NULL);
937 /* Introduce an explicit dot above when lowercasing capital I's and J's
938 * whenever there are more accents above. [SpecialCasing.txt] */
939 else if (locale_type == LOCALE_LITHUANIAN &&
940 (c == 0x00cc || c == 0x00cd || c == 0x0128))
942 len += g_unichar_to_utf8 (0x0069, out_buffer ? out_buffer + len : NULL);
943 len += g_unichar_to_utf8 (0x0307, out_buffer ? out_buffer + len : NULL);
948 len += g_unichar_to_utf8 (0x0300, out_buffer ? out_buffer + len : NULL);
951 len += g_unichar_to_utf8 (0x0301, out_buffer ? out_buffer + len : NULL);
954 len += g_unichar_to_utf8 (0x0303, out_buffer ? out_buffer + len : NULL);
958 else if (locale_type == LOCALE_LITHUANIAN &&
959 (c == 'I' || c == 'J' || c == 0x012e) &&
962 len += g_unichar_to_utf8 (g_unichar_tolower (c), out_buffer ? out_buffer + len : NULL);
963 len += g_unichar_to_utf8 (0x0307, out_buffer ? out_buffer + len : NULL);
965 else if (c == 0x03A3) /* GREEK CAPITAL LETTER SIGMA */
967 if ((max_len < 0 || p < str + max_len) && *p)
969 gunichar next_c = g_utf8_get_char (p);
970 int next_type = TYPE(next_c);
972 /* SIGMA mapps differently depending on whether it is
973 * final or not. The following simplified test would
974 * fail in the case of combining marks following the
975 * sigma, but I don't think that occurs in real text.
976 * The test here matches that in ICU.
978 if (ISALPHA (next_type)) /* Lu,Ll,Lt,Lm,Lo */
979 val = 0x3c3; /* GREEK SMALL SIGMA */
981 val = 0x3c2; /* GREEK SMALL FINAL SIGMA */
984 val = 0x3c2; /* GREEK SMALL FINAL SIGMA */
986 len += g_unichar_to_utf8 (val, out_buffer ? out_buffer + len : NULL);
989 OR (G_UNICODE_UPPERCASE_LETTER,
990 OR (G_UNICODE_TITLECASE_LETTER,
993 val = ATTTABLE (c >> 8, c & 0xff);
995 if (val >= 0x1000000)
997 len += output_special_case (out_buffer ? out_buffer + len : NULL, val - 0x1000000, t, 0);
1001 if (t == G_UNICODE_TITLECASE_LETTER)
1004 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
1006 if (title_table[i][0] == c)
1007 val = title_table[i][2];
1011 len += g_unichar_to_utf8 (val, out_buffer ? out_buffer + len : NULL);
1016 gsize char_len = g_utf8_skip[*(guchar *)last];
1019 memcpy (out_buffer + len, last, char_len);
1031 * @str: a UTF-8 encoded string
1032 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
1034 * Converts all Unicode characters in the string that have a case
1035 * to lowercase. The exact manner that this is done depends
1036 * on the current locale, and may result in the number of
1037 * characters in the string changing.
1039 * Return value: a newly allocated string, with all characters
1040 * converted to lowercase.
1043 g_utf8_strdown (const gchar *str,
1047 LocaleType locale_type;
1050 g_return_val_if_fail (str != NULL, NULL);
1052 locale_type = get_locale_type ();
1055 * We use a two pass approach to keep memory management simple
1057 result_len = real_tolower (str, len, NULL, locale_type);
1058 result = g_malloc (result_len + 1);
1059 real_tolower (str, len, result, locale_type);
1060 result[result_len] = '\0';
1067 * @str: a UTF-8 encoded string
1068 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
1070 * Converts a string into a form that is independent of case. The
1071 * result will not correspond to any particular case, but can be
1072 * compared for equality or ordered with the results of calling
1073 * g_utf8_casefold() on other strings.
1075 * Note that calling g_utf8_casefold() followed by g_utf8_collate() is
1076 * only an approximation to the correct linguistic case insensitive
1077 * ordering, though it is a fairly good one. Getting this exactly
1078 * right would require a more sophisticated collation function that
1079 * takes case sensitivity into account. GLib does not currently
1080 * provide such a function.
1082 * Return value: a newly allocated string, that is a
1083 * case independent form of @str.
1086 g_utf8_casefold (const gchar *str,
1092 g_return_val_if_fail (str != NULL, NULL);
1094 result = g_string_new (NULL);
1096 while ((len < 0 || p < str + len) && *p)
1098 gunichar ch = g_utf8_get_char (p);
1101 int end = G_N_ELEMENTS (casefold_table);
1103 if (ch >= casefold_table[start].ch &&
1104 ch <= casefold_table[end - 1].ch)
1108 int half = (start + end) / 2;
1109 if (ch == casefold_table[half].ch)
1111 g_string_append (result, casefold_table[half].data);
1114 else if (half == start)
1116 else if (ch > casefold_table[half].ch)
1123 g_string_append_unichar (result, g_unichar_tolower (ch));
1126 p = g_utf8_next_char (p);
1129 return g_string_free (result, FALSE);
1133 * g_unichar_get_mirror_char:
1134 * @ch: a Unicode character
1135 * @mirrored_ch: location to store the mirrored character
1137 * In Unicode, some characters are <firstterm>mirrored</firstterm>. This
1138 * means that their images are mirrored horizontally in text that is laid
1139 * out from right to left. For instance, "(" would become its mirror image,
1140 * ")", in right-to-left text.
1142 * If @ch has the Unicode mirrored property and there is another unicode
1143 * character that typically has a glyph that is the mirror image of @ch's
1144 * glyph and @mirrored_ch is set, it puts that character in the address
1145 * pointed to by @mirrored_ch. Otherwise the original character is put.
1147 * Return value: %TRUE if @ch has a mirrored character, %FALSE otherwise
1152 g_unichar_get_mirror_char (gunichar ch,
1153 gunichar *mirrored_ch)
1158 mirrored = GLIB_GET_MIRRORING(ch);
1160 found = ch != mirrored;
1162 *mirrored_ch = mirrored;
1168 #define __G_UNIPROP_C__
1169 #include "galiasdef.c"