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.
31 #include "gtestutils.h"
34 #include "gunichartables.h"
35 #include "gmirroringtable.h"
36 #include "gscripttable.h"
37 #include "gunicodeprivate.h"
42 #define ATTR_TABLE(Page) (((Page) <= G_UNICODE_LAST_PAGE_PART1) \
43 ? attr_table_part1[Page] \
44 : attr_table_part2[(Page) - 0xe00])
46 #define ATTTABLE(Page, Char) \
47 ((ATTR_TABLE(Page) == G_UNICODE_MAX_TABLE_INDEX) ? 0 : (attr_data[ATTR_TABLE(Page)][Char]))
49 #define TTYPE_PART1(Page, Char) \
50 ((type_table_part1[Page] >= G_UNICODE_MAX_TABLE_INDEX) \
51 ? (type_table_part1[Page] - G_UNICODE_MAX_TABLE_INDEX) \
52 : (type_data[type_table_part1[Page]][Char]))
54 #define TTYPE_PART2(Page, Char) \
55 ((type_table_part2[Page] >= G_UNICODE_MAX_TABLE_INDEX) \
56 ? (type_table_part2[Page] - G_UNICODE_MAX_TABLE_INDEX) \
57 : (type_data[type_table_part2[Page]][Char]))
60 (((Char) <= G_UNICODE_LAST_CHAR_PART1) \
61 ? TTYPE_PART1 ((Char) >> 8, (Char) & 0xff) \
62 : (((Char) >= 0xe0000 && (Char) <= G_UNICODE_LAST_CHAR) \
63 ? TTYPE_PART2 (((Char) - 0xe0000) >> 8, (Char) & 0xff) \
64 : G_UNICODE_UNASSIGNED))
67 #define IS(Type, Class) (((guint)1 << (Type)) & (Class))
68 #define OR(Type, Rest) (((guint)1 << (Type)) | (Rest))
72 #define ISALPHA(Type) IS ((Type), \
73 OR (G_UNICODE_LOWERCASE_LETTER, \
74 OR (G_UNICODE_UPPERCASE_LETTER, \
75 OR (G_UNICODE_TITLECASE_LETTER, \
76 OR (G_UNICODE_MODIFIER_LETTER, \
77 OR (G_UNICODE_OTHER_LETTER, 0))))))
79 #define ISALDIGIT(Type) IS ((Type), \
80 OR (G_UNICODE_DECIMAL_NUMBER, \
81 OR (G_UNICODE_LETTER_NUMBER, \
82 OR (G_UNICODE_OTHER_NUMBER, \
83 OR (G_UNICODE_LOWERCASE_LETTER, \
84 OR (G_UNICODE_UPPERCASE_LETTER, \
85 OR (G_UNICODE_TITLECASE_LETTER, \
86 OR (G_UNICODE_MODIFIER_LETTER, \
87 OR (G_UNICODE_OTHER_LETTER, 0)))))))))
89 #define ISMARK(Type) IS ((Type), \
90 OR (G_UNICODE_NON_SPACING_MARK, \
91 OR (G_UNICODE_SPACING_MARK, \
92 OR (G_UNICODE_ENCLOSING_MARK, 0))))
94 #define ISZEROWIDTHTYPE(Type) IS ((Type), \
95 OR (G_UNICODE_NON_SPACING_MARK, \
96 OR (G_UNICODE_ENCLOSING_MARK, \
97 OR (G_UNICODE_FORMAT, 0))))
101 * @c: a Unicode character
103 * Determines whether a character is alphanumeric.
104 * Given some UTF-8 text, obtain a character value
105 * with g_utf8_get_char().
107 * Return value: %TRUE if @c is an alphanumeric character
110 g_unichar_isalnum (gunichar c)
112 return ISALDIGIT (TYPE (c)) ? TRUE : FALSE;
117 * @c: a Unicode character
119 * Determines whether a character is alphabetic (i.e. a letter).
120 * Given some UTF-8 text, obtain a character value with
123 * Return value: %TRUE if @c is an alphabetic character
126 g_unichar_isalpha (gunichar c)
128 return ISALPHA (TYPE (c)) ? TRUE : FALSE;
134 * @c: a Unicode character
136 * Determines whether a character is a control character.
137 * Given some UTF-8 text, obtain a character value with
140 * Return value: %TRUE if @c is a control character
143 g_unichar_iscntrl (gunichar c)
145 return TYPE (c) == G_UNICODE_CONTROL;
150 * @c: a Unicode character
152 * Determines whether a character is numeric (i.e. a digit). This
153 * covers ASCII 0-9 and also digits in other languages/scripts. Given
154 * some UTF-8 text, obtain a character value with g_utf8_get_char().
156 * Return value: %TRUE if @c is a digit
159 g_unichar_isdigit (gunichar c)
161 return TYPE (c) == G_UNICODE_DECIMAL_NUMBER;
167 * @c: a Unicode character
169 * Determines whether a character is printable and not a space
170 * (returns %FALSE for control characters, format characters, and
171 * spaces). g_unichar_isprint() is similar, but returns %TRUE for
172 * spaces. Given some UTF-8 text, obtain a character value with
175 * Return value: %TRUE if @c is printable unless it's a space
178 g_unichar_isgraph (gunichar c)
181 OR (G_UNICODE_CONTROL,
182 OR (G_UNICODE_FORMAT,
183 OR (G_UNICODE_UNASSIGNED,
184 OR (G_UNICODE_SURROGATE,
185 OR (G_UNICODE_SPACE_SEPARATOR,
191 * @c: a Unicode character
193 * Determines whether a character is a lowercase letter.
194 * Given some UTF-8 text, obtain a character value with
197 * Return value: %TRUE if @c is a lowercase letter
200 g_unichar_islower (gunichar c)
202 return TYPE (c) == G_UNICODE_LOWERCASE_LETTER;
208 * @c: a Unicode character
210 * Determines whether a character is printable.
211 * Unlike g_unichar_isgraph(), returns %TRUE for spaces.
212 * Given some UTF-8 text, obtain a character value with
215 * Return value: %TRUE if @c is printable
218 g_unichar_isprint (gunichar c)
221 OR (G_UNICODE_CONTROL,
222 OR (G_UNICODE_FORMAT,
223 OR (G_UNICODE_UNASSIGNED,
224 OR (G_UNICODE_SURROGATE,
230 * @c: a Unicode character
232 * Determines whether a character is punctuation or a symbol.
233 * Given some UTF-8 text, obtain a character value with
236 * Return value: %TRUE if @c is a punctuation or symbol character
239 g_unichar_ispunct (gunichar c)
242 OR (G_UNICODE_CONNECT_PUNCTUATION,
243 OR (G_UNICODE_DASH_PUNCTUATION,
244 OR (G_UNICODE_CLOSE_PUNCTUATION,
245 OR (G_UNICODE_FINAL_PUNCTUATION,
246 OR (G_UNICODE_INITIAL_PUNCTUATION,
247 OR (G_UNICODE_OTHER_PUNCTUATION,
248 OR (G_UNICODE_OPEN_PUNCTUATION,
249 OR (G_UNICODE_CURRENCY_SYMBOL,
250 OR (G_UNICODE_MODIFIER_SYMBOL,
251 OR (G_UNICODE_MATH_SYMBOL,
252 OR (G_UNICODE_OTHER_SYMBOL,
253 0)))))))))))) ? TRUE : FALSE;
258 * @c: a Unicode character
260 * Determines whether a character is a space, tab, or line separator
261 * (newline, carriage return, etc.). Given some UTF-8 text, obtain a
262 * character value with g_utf8_get_char().
264 * (Note: don't use this to do word breaking; you have to use
265 * Pango or equivalent to get word breaking right, the algorithm
266 * is fairly complex.)
268 * Return value: %TRUE if @c is a space character
271 g_unichar_isspace (gunichar c)
275 /* special-case these since Unicode thinks they are not spaces */
286 OR (G_UNICODE_SPACE_SEPARATOR,
287 OR (G_UNICODE_LINE_SEPARATOR,
288 OR (G_UNICODE_PARAGRAPH_SEPARATOR,
289 0)))) ? TRUE : FALSE;
297 * @c: a Unicode character
299 * Determines whether a character is a mark (non-spacing mark,
300 * combining mark, or enclosing mark in Unicode speak).
301 * Given some UTF-8 text, obtain a character value
302 * with g_utf8_get_char().
304 * Note: in most cases where isalpha characters are allowed,
305 * ismark characters should be allowed to as they are essential
306 * for writing most European languages as well as many non-Latin
309 * Return value: %TRUE if @c is a mark character
314 g_unichar_ismark (gunichar c)
316 return ISMARK (TYPE (c));
321 * @c: a Unicode character
323 * Determines if a character is uppercase.
325 * Return value: %TRUE if @c is an uppercase character
328 g_unichar_isupper (gunichar c)
330 return TYPE (c) == G_UNICODE_UPPERCASE_LETTER;
335 * @c: a Unicode character
337 * Determines if a character is titlecase. Some characters in
338 * Unicode which are composites, such as the DZ digraph
339 * have three case variants instead of just two. The titlecase
340 * form is used at the beginning of a word where only the
341 * first letter is capitalized. The titlecase form of the DZ
342 * digraph is U+01F2 LATIN CAPITAL LETTTER D WITH SMALL LETTER Z.
344 * Return value: %TRUE if the character is titlecase
347 g_unichar_istitle (gunichar c)
350 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
351 if (title_table[i][0] == c)
357 * g_unichar_isxdigit:
358 * @c: a Unicode character.
360 * Determines if a character is a hexidecimal digit.
362 * Return value: %TRUE if the character is a hexadecimal digit
365 g_unichar_isxdigit (gunichar c)
367 return ((c >= 'a' && c <= 'f')
368 || (c >= 'A' && c <= 'F')
369 || (TYPE (c) == G_UNICODE_DECIMAL_NUMBER));
373 * g_unichar_isdefined:
374 * @c: a Unicode character
376 * Determines if a given character is assigned in the Unicode
379 * Return value: %TRUE if the character has an assigned value
382 g_unichar_isdefined (gunichar c)
385 OR (G_UNICODE_UNASSIGNED,
386 OR (G_UNICODE_SURROGATE,
391 * g_unichar_iszerowidth:
392 * @c: a Unicode character
394 * Determines if a given character typically takes zero width when rendered.
395 * The return value is %TRUE for all non-spacing and enclosing marks
396 * (e.g., combining accents), format characters, zero-width
397 * space, but not U+00AD SOFT HYPHEN.
399 * A typical use of this function is with one of g_unichar_iswide() or
400 * g_unichar_iswide_cjk() to determine the number of cells a string occupies
401 * when displayed on a grid display (terminals). However, note that not all
402 * terminals support zero-width rendering of zero-width marks.
404 * Return value: %TRUE if the character has zero width
409 g_unichar_iszerowidth (gunichar c)
411 if (G_UNLIKELY (c == 0x00AD))
414 if (G_UNLIKELY (ISZEROWIDTHTYPE (TYPE (c))))
417 if (G_UNLIKELY ((c >= 0x1160 && c < 0x1200) ||
430 interval_compare (const void *key, const void *elt)
432 gunichar c = GPOINTER_TO_UINT (key);
433 struct Interval *interval = (struct Interval *)elt;
435 if (c < interval->start)
437 if (c > interval->end)
447 * The tables for g_unichar_iswide() and g_unichar_iswide_cjk() are
448 * generated from the Unicode Character Database's file
449 * extracted/DerivedEastAsianWidth.txt using the gen-iswide-table.py
452 * ./gen-iswide-table.py < path/to/ucd/extracted/DerivedEastAsianWidth.txt | fmt
454 * Last update for Unicode 6.0.
459 * @c: a Unicode character
461 * Determines if a character is typically rendered in a double-width
464 * Return value: %TRUE if the character is wide
467 g_unichar_iswide (gunichar c)
469 /* See NOTE earlier for how to update this table. */
470 static const struct Interval wide[] = {
471 {0x1100, 0x115F}, {0x11A3, 0x11A7}, {0x11FA, 0x11FF}, {0x2329, 0x232A},
472 {0x2E80, 0x2E99}, {0x2E9B, 0x2EF3}, {0x2F00, 0x2FD5}, {0x2FF0, 0x2FFB},
473 {0x3000, 0x303E}, {0x3041, 0x3096}, {0x3099, 0x30FF}, {0x3105, 0x312D},
474 {0x3131, 0x318E}, {0x3190, 0x31BA}, {0x31C0, 0x31E3}, {0x31F0, 0x321E},
475 {0x3220, 0x3247}, {0x3250, 0x32FE}, {0x3300, 0x4DBF}, {0x4E00, 0xA48C},
476 {0xA490, 0xA4C6}, {0xA960, 0xA97C}, {0xAC00, 0xD7A3}, {0xD7B0, 0xD7C6},
477 {0xD7CB, 0xD7FB}, {0xF900, 0xFAFF}, {0xFE10, 0xFE19}, {0xFE30, 0xFE52},
478 {0xFE54, 0xFE66}, {0xFE68, 0xFE6B}, {0xFF01, 0xFF60}, {0xFFE0, 0xFFE6},
479 {0x1B000, 0x1B001}, {0x1F200, 0x1F202}, {0x1F210, 0x1F23A}, {0x1F240,
480 0x1F248}, {0x1F250, 0x1F251}, {0x20000, 0x2FFFD}, {0x30000, 0x3FFFD}
483 if (bsearch (GUINT_TO_POINTER (c), wide, G_N_ELEMENTS (wide), sizeof wide[0],
492 * g_unichar_iswide_cjk:
493 * @c: a Unicode character
495 * Determines if a character is typically rendered in a double-width
496 * cell under legacy East Asian locales. If a character is wide according to
497 * g_unichar_iswide(), then it is also reported wide with this function, but
498 * the converse is not necessarily true. See the
499 * <ulink url="http://www.unicode.org/reports/tr11/">Unicode Standard
500 * Annex #11</ulink> for details.
502 * If a character passes the g_unichar_iswide() test then it will also pass
503 * this test, but not the other way around. Note that some characters may
504 * pas both this test and g_unichar_iszerowidth().
506 * Return value: %TRUE if the character is wide in legacy East Asian locales
511 g_unichar_iswide_cjk (gunichar c)
513 /* See NOTE earlier for how to update this table. */
514 static const struct Interval ambiguous[] = {
515 {0x00A1, 0x00A1}, {0x00A4, 0x00A4}, {0x00A7, 0x00A8}, {0x00AA, 0x00AA},
516 {0x00AD, 0x00AE}, {0x00B0, 0x00B4}, {0x00B6, 0x00BA}, {0x00BC, 0x00BF},
517 {0x00C6, 0x00C6}, {0x00D0, 0x00D0}, {0x00D7, 0x00D8}, {0x00DE, 0x00E1},
518 {0x00E6, 0x00E6}, {0x00E8, 0x00EA}, {0x00EC, 0x00ED}, {0x00F0, 0x00F0},
519 {0x00F2, 0x00F3}, {0x00F7, 0x00FA}, {0x00FC, 0x00FC}, {0x00FE, 0x00FE},
520 {0x0101, 0x0101}, {0x0111, 0x0111}, {0x0113, 0x0113}, {0x011B, 0x011B},
521 {0x0126, 0x0127}, {0x012B, 0x012B}, {0x0131, 0x0133}, {0x0138, 0x0138},
522 {0x013F, 0x0142}, {0x0144, 0x0144}, {0x0148, 0x014B}, {0x014D, 0x014D},
523 {0x0152, 0x0153}, {0x0166, 0x0167}, {0x016B, 0x016B}, {0x01CE, 0x01CE},
524 {0x01D0, 0x01D0}, {0x01D2, 0x01D2}, {0x01D4, 0x01D4}, {0x01D6, 0x01D6},
525 {0x01D8, 0x01D8}, {0x01DA, 0x01DA}, {0x01DC, 0x01DC}, {0x0251, 0x0251},
526 {0x0261, 0x0261}, {0x02C4, 0x02C4}, {0x02C7, 0x02C7}, {0x02C9, 0x02CB},
527 {0x02CD, 0x02CD}, {0x02D0, 0x02D0}, {0x02D8, 0x02DB}, {0x02DD, 0x02DD},
528 {0x02DF, 0x02DF}, {0x0300, 0x036F}, {0x0391, 0x03A1}, {0x03A3, 0x03A9},
529 {0x03B1, 0x03C1}, {0x03C3, 0x03C9}, {0x0401, 0x0401}, {0x0410, 0x044F},
530 {0x0451, 0x0451}, {0x2010, 0x2010}, {0x2013, 0x2016}, {0x2018, 0x2019},
531 {0x201C, 0x201D}, {0x2020, 0x2022}, {0x2024, 0x2027}, {0x2030, 0x2030},
532 {0x2032, 0x2033}, {0x2035, 0x2035}, {0x203B, 0x203B}, {0x203E, 0x203E},
533 {0x2074, 0x2074}, {0x207F, 0x207F}, {0x2081, 0x2084}, {0x20AC, 0x20AC},
534 {0x2103, 0x2103}, {0x2105, 0x2105}, {0x2109, 0x2109}, {0x2113, 0x2113},
535 {0x2116, 0x2116}, {0x2121, 0x2122}, {0x2126, 0x2126}, {0x212B, 0x212B},
536 {0x2153, 0x2154}, {0x215B, 0x215E}, {0x2160, 0x216B}, {0x2170, 0x2179},
537 {0x2189, 0x2189}, {0x2190, 0x2199}, {0x21B8, 0x21B9}, {0x21D2, 0x21D2},
538 {0x21D4, 0x21D4}, {0x21E7, 0x21E7}, {0x2200, 0x2200}, {0x2202, 0x2203},
539 {0x2207, 0x2208}, {0x220B, 0x220B}, {0x220F, 0x220F}, {0x2211, 0x2211},
540 {0x2215, 0x2215}, {0x221A, 0x221A}, {0x221D, 0x2220}, {0x2223, 0x2223},
541 {0x2225, 0x2225}, {0x2227, 0x222C}, {0x222E, 0x222E}, {0x2234, 0x2237},
542 {0x223C, 0x223D}, {0x2248, 0x2248}, {0x224C, 0x224C}, {0x2252, 0x2252},
543 {0x2260, 0x2261}, {0x2264, 0x2267}, {0x226A, 0x226B}, {0x226E, 0x226F},
544 {0x2282, 0x2283}, {0x2286, 0x2287}, {0x2295, 0x2295}, {0x2299, 0x2299},
545 {0x22A5, 0x22A5}, {0x22BF, 0x22BF}, {0x2312, 0x2312}, {0x2460, 0x24E9},
546 {0x24EB, 0x254B}, {0x2550, 0x2573}, {0x2580, 0x258F}, {0x2592, 0x2595},
547 {0x25A0, 0x25A1}, {0x25A3, 0x25A9}, {0x25B2, 0x25B3}, {0x25B6, 0x25B7},
548 {0x25BC, 0x25BD}, {0x25C0, 0x25C1}, {0x25C6, 0x25C8}, {0x25CB, 0x25CB},
549 {0x25CE, 0x25D1}, {0x25E2, 0x25E5}, {0x25EF, 0x25EF}, {0x2605, 0x2606},
550 {0x2609, 0x2609}, {0x260E, 0x260F}, {0x2614, 0x2615}, {0x261C, 0x261C},
551 {0x261E, 0x261E}, {0x2640, 0x2640}, {0x2642, 0x2642}, {0x2660, 0x2661},
552 {0x2663, 0x2665}, {0x2667, 0x266A}, {0x266C, 0x266D}, {0x266F, 0x266F},
553 {0x269E, 0x269F}, {0x26BE, 0x26BF}, {0x26C4, 0x26CD}, {0x26CF, 0x26E1},
554 {0x26E3, 0x26E3}, {0x26E8, 0x26FF}, {0x273D, 0x273D}, {0x2757, 0x2757},
555 {0x2776, 0x277F}, {0x2B55, 0x2B59}, {0x3248, 0x324F}, {0xE000, 0xF8FF},
556 {0xFE00, 0xFE0F}, {0xFFFD, 0xFFFD}, {0x1F100, 0x1F10A}, {0x1F110,
557 0x1F12D}, {0x1F130, 0x1F169}, {0x1F170, 0x1F19A}, {0xE0100, 0xE01EF},
558 {0xF0000, 0xFFFFD}, {0x100000, 0x10FFFD}
561 if (g_unichar_iswide (c))
564 if (bsearch (GUINT_TO_POINTER (c), ambiguous, G_N_ELEMENTS (ambiguous), sizeof ambiguous[0],
574 * @c: a Unicode character
576 * Converts a character to uppercase.
578 * Return value: the result of converting @c to uppercase.
579 * If @c is not an lowercase or titlecase character,
580 * or has no upper case equivalent @c is returned unchanged.
583 g_unichar_toupper (gunichar c)
586 if (t == G_UNICODE_LOWERCASE_LETTER)
588 gunichar val = ATTTABLE (c >> 8, c & 0xff);
589 if (val >= 0x1000000)
591 const gchar *p = special_case_table + val - 0x1000000;
592 val = g_utf8_get_char (p);
594 /* Some lowercase letters, e.g., U+000AA, FEMININE ORDINAL INDICATOR,
595 * do not have an uppercase equivalent, in which case val will be
598 return val ? val : c;
600 else if (t == G_UNICODE_TITLECASE_LETTER)
603 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
605 if (title_table[i][0] == c)
606 return title_table[i][1] ? title_table[i][1] : c;
614 * @c: a Unicode character.
616 * Converts a character to lower case.
618 * Return value: the result of converting @c to lower case.
619 * If @c is not an upperlower or titlecase character,
620 * or has no lowercase equivalent @c is returned unchanged.
623 g_unichar_tolower (gunichar c)
626 if (t == G_UNICODE_UPPERCASE_LETTER)
628 gunichar val = ATTTABLE (c >> 8, c & 0xff);
629 if (val >= 0x1000000)
631 const gchar *p = special_case_table + val - 0x1000000;
632 return g_utf8_get_char (p);
636 /* Not all uppercase letters are guaranteed to have a lowercase
637 * equivalent. If this is the case, val will be zero. */
638 return val ? val : c;
641 else if (t == G_UNICODE_TITLECASE_LETTER)
644 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
646 if (title_table[i][0] == c)
647 return title_table[i][2];
655 * @c: a Unicode character
657 * Converts a character to the titlecase.
659 * Return value: the result of converting @c to titlecase.
660 * If @c is not an uppercase or lowercase character,
661 * @c is returned unchanged.
664 g_unichar_totitle (gunichar c)
667 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
669 if (title_table[i][0] == c || title_table[i][1] == c
670 || title_table[i][2] == c)
671 return title_table[i][0];
674 if (TYPE (c) == G_UNICODE_LOWERCASE_LETTER)
675 return g_unichar_toupper (c);
681 * g_unichar_digit_value:
682 * @c: a Unicode character
684 * Determines the numeric value of a character as a decimal
687 * Return value: If @c is a decimal digit (according to
688 * g_unichar_isdigit()), its numeric value. Otherwise, -1.
691 g_unichar_digit_value (gunichar c)
693 if (TYPE (c) == G_UNICODE_DECIMAL_NUMBER)
694 return ATTTABLE (c >> 8, c & 0xff);
699 * g_unichar_xdigit_value:
700 * @c: a Unicode character
702 * Determines the numeric value of a character as a hexidecimal
705 * Return value: If @c is a hex digit (according to
706 * g_unichar_isxdigit()), its numeric value. Otherwise, -1.
709 g_unichar_xdigit_value (gunichar c)
711 if (c >= 'A' && c <= 'F')
713 if (c >= 'a' && c <= 'f')
715 if (TYPE (c) == G_UNICODE_DECIMAL_NUMBER)
716 return ATTTABLE (c >> 8, c & 0xff);
722 * @c: a Unicode character
724 * Classifies a Unicode character by type.
726 * Return value: the type of the character.
729 g_unichar_type (gunichar c)
735 * Case mapping functions
745 get_locale_type (void)
748 char *tem = g_win32_getlocale ();
755 const char *locale = setlocale (LC_CTYPE, NULL);
758 return LOCALE_NORMAL;
764 if (locale[1] == 'z')
765 return LOCALE_TURKIC;
768 if (locale[1] == 't')
769 return LOCALE_LITHUANIAN;
772 if (locale[1] == 'r')
773 return LOCALE_TURKIC;
777 return LOCALE_NORMAL;
781 output_marks (const char **p_inout,
785 const char *p = *p_inout;
790 gunichar c = g_utf8_get_char (p);
792 if (ISMARK (TYPE (c)))
794 if (!remove_dot || c != 0x307 /* COMBINING DOT ABOVE */)
795 len += g_unichar_to_utf8 (c, out_buffer ? out_buffer + len : NULL);
796 p = g_utf8_next_char (p);
807 output_special_case (gchar *out_buffer,
812 const gchar *p = special_case_table + offset;
815 if (type != G_UNICODE_TITLECASE_LETTER)
816 p = g_utf8_next_char (p);
823 memcpy (out_buffer, p, len);
829 real_toupper (const gchar *str,
832 LocaleType locale_type)
834 const gchar *p = str;
835 const char *last = NULL;
837 gboolean last_was_i = FALSE;
839 while ((max_len < 0 || p < str + max_len) && *p)
841 gunichar c = g_utf8_get_char (p);
846 p = g_utf8_next_char (p);
848 if (locale_type == LOCALE_LITHUANIAN)
856 /* Nasty, need to remove any dot above. Though
857 * I think only E WITH DOT ABOVE occurs in practice
858 * which could simplify this considerably.
861 gunichar decomp[G_UNICHAR_MAX_DECOMPOSITION_LENGTH];
863 decomp_len = g_unichar_fully_decompose (c, FALSE, decomp, G_N_ELEMENTS (decomp));
864 for (i=0; i < decomp_len; i++)
866 if (decomp[i] != 0x307 /* COMBINING DOT ABOVE */)
867 len += g_unichar_to_utf8 (g_unichar_toupper (decomp[i]), out_buffer ? out_buffer + len : NULL);
870 len += output_marks (&p, out_buffer ? out_buffer + len : NULL, TRUE);
880 if (locale_type == LOCALE_TURKIC && c == 'i')
882 /* i => LATIN CAPITAL LETTER I WITH DOT ABOVE */
883 len += g_unichar_to_utf8 (0x130, out_buffer ? out_buffer + len : NULL);
885 else if (c == 0x0345) /* COMBINING GREEK YPOGEGRAMMENI */
887 /* Nasty, need to move it after other combining marks .. this would go away if
888 * we normalized first.
890 len += output_marks (&p, out_buffer ? out_buffer + len : NULL, FALSE);
892 /* And output as GREEK CAPITAL LETTER IOTA */
893 len += g_unichar_to_utf8 (0x399, out_buffer ? out_buffer + len : NULL);
896 OR (G_UNICODE_LOWERCASE_LETTER,
897 OR (G_UNICODE_TITLECASE_LETTER,
900 val = ATTTABLE (c >> 8, c & 0xff);
902 if (val >= 0x1000000)
904 len += output_special_case (out_buffer ? out_buffer + len : NULL, val - 0x1000000, t,
905 t == G_UNICODE_LOWERCASE_LETTER ? 0 : 1);
909 if (t == G_UNICODE_TITLECASE_LETTER)
912 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
914 if (title_table[i][0] == c)
916 val = title_table[i][1];
922 /* Some lowercase letters, e.g., U+000AA, FEMININE ORDINAL INDICATOR,
923 * do not have an uppercase equivalent, in which case val will be
925 len += g_unichar_to_utf8 (val ? val : c, out_buffer ? out_buffer + len : NULL);
930 gsize char_len = g_utf8_skip[*(guchar *)last];
933 memcpy (out_buffer + len, last, char_len);
945 * @str: a UTF-8 encoded string
946 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
948 * Converts all Unicode characters in the string that have a case
949 * to uppercase. The exact manner that this is done depends
950 * on the current locale, and may result in the number of
951 * characters in the string increasing. (For instance, the
952 * German ess-zet will be changed to SS.)
954 * Return value: a newly allocated string, with all characters
955 * converted to uppercase.
958 g_utf8_strup (const gchar *str,
962 LocaleType locale_type;
965 g_return_val_if_fail (str != NULL, NULL);
967 locale_type = get_locale_type ();
970 * We use a two pass approach to keep memory management simple
972 result_len = real_toupper (str, len, NULL, locale_type);
973 result = g_malloc (result_len + 1);
974 real_toupper (str, len, result, locale_type);
975 result[result_len] = '\0';
980 /* traverses the string checking for characters with combining class == 230
981 * until a base character is found */
983 has_more_above (const gchar *str)
985 const gchar *p = str;
986 gint combining_class;
990 combining_class = g_unichar_combining_class (g_utf8_get_char (p));
991 if (combining_class == 230)
993 else if (combining_class == 0)
996 p = g_utf8_next_char (p);
1003 real_tolower (const gchar *str,
1006 LocaleType locale_type)
1008 const gchar *p = str;
1009 const char *last = NULL;
1012 while ((max_len < 0 || p < str + max_len) && *p)
1014 gunichar c = g_utf8_get_char (p);
1019 p = g_utf8_next_char (p);
1021 if (locale_type == LOCALE_TURKIC && c == 'I')
1023 if (g_utf8_get_char (p) == 0x0307)
1025 /* I + COMBINING DOT ABOVE => i (U+0069) */
1026 len += g_unichar_to_utf8 (0x0069, out_buffer ? out_buffer + len : NULL);
1027 p = g_utf8_next_char (p);
1031 /* I => LATIN SMALL LETTER DOTLESS I */
1032 len += g_unichar_to_utf8 (0x131, out_buffer ? out_buffer + len : NULL);
1035 /* Introduce an explicit dot above when lowercasing capital I's and J's
1036 * whenever there are more accents above. [SpecialCasing.txt] */
1037 else if (locale_type == LOCALE_LITHUANIAN &&
1038 (c == 0x00cc || c == 0x00cd || c == 0x0128))
1040 len += g_unichar_to_utf8 (0x0069, out_buffer ? out_buffer + len : NULL);
1041 len += g_unichar_to_utf8 (0x0307, out_buffer ? out_buffer + len : NULL);
1046 len += g_unichar_to_utf8 (0x0300, out_buffer ? out_buffer + len : NULL);
1049 len += g_unichar_to_utf8 (0x0301, out_buffer ? out_buffer + len : NULL);
1052 len += g_unichar_to_utf8 (0x0303, out_buffer ? out_buffer + len : NULL);
1056 else if (locale_type == LOCALE_LITHUANIAN &&
1057 (c == 'I' || c == 'J' || c == 0x012e) &&
1060 len += g_unichar_to_utf8 (g_unichar_tolower (c), out_buffer ? out_buffer + len : NULL);
1061 len += g_unichar_to_utf8 (0x0307, out_buffer ? out_buffer + len : NULL);
1063 else if (c == 0x03A3) /* GREEK CAPITAL LETTER SIGMA */
1065 if ((max_len < 0 || p < str + max_len) && *p)
1067 gunichar next_c = g_utf8_get_char (p);
1068 int next_type = TYPE(next_c);
1070 /* SIGMA mapps differently depending on whether it is
1071 * final or not. The following simplified test would
1072 * fail in the case of combining marks following the
1073 * sigma, but I don't think that occurs in real text.
1074 * The test here matches that in ICU.
1076 if (ISALPHA (next_type)) /* Lu,Ll,Lt,Lm,Lo */
1077 val = 0x3c3; /* GREEK SMALL SIGMA */
1079 val = 0x3c2; /* GREEK SMALL FINAL SIGMA */
1082 val = 0x3c2; /* GREEK SMALL FINAL SIGMA */
1084 len += g_unichar_to_utf8 (val, out_buffer ? out_buffer + len : NULL);
1087 OR (G_UNICODE_UPPERCASE_LETTER,
1088 OR (G_UNICODE_TITLECASE_LETTER,
1091 val = ATTTABLE (c >> 8, c & 0xff);
1093 if (val >= 0x1000000)
1095 len += output_special_case (out_buffer ? out_buffer + len : NULL, val - 0x1000000, t, 0);
1099 if (t == G_UNICODE_TITLECASE_LETTER)
1102 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
1104 if (title_table[i][0] == c)
1106 val = title_table[i][2];
1112 /* Not all uppercase letters are guaranteed to have a lowercase
1113 * equivalent. If this is the case, val will be zero. */
1114 len += g_unichar_to_utf8 (val ? val : c, out_buffer ? out_buffer + len : NULL);
1119 gsize char_len = g_utf8_skip[*(guchar *)last];
1122 memcpy (out_buffer + len, last, char_len);
1134 * @str: a UTF-8 encoded string
1135 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
1137 * Converts all Unicode characters in the string that have a case
1138 * to lowercase. The exact manner that this is done depends
1139 * on the current locale, and may result in the number of
1140 * characters in the string changing.
1142 * Return value: a newly allocated string, with all characters
1143 * converted to lowercase.
1146 g_utf8_strdown (const gchar *str,
1150 LocaleType locale_type;
1153 g_return_val_if_fail (str != NULL, NULL);
1155 locale_type = get_locale_type ();
1158 * We use a two pass approach to keep memory management simple
1160 result_len = real_tolower (str, len, NULL, locale_type);
1161 result = g_malloc (result_len + 1);
1162 real_tolower (str, len, result, locale_type);
1163 result[result_len] = '\0';
1170 * @str: a UTF-8 encoded string
1171 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
1173 * Converts a string into a form that is independent of case. The
1174 * result will not correspond to any particular case, but can be
1175 * compared for equality or ordered with the results of calling
1176 * g_utf8_casefold() on other strings.
1178 * Note that calling g_utf8_casefold() followed by g_utf8_collate() is
1179 * only an approximation to the correct linguistic case insensitive
1180 * ordering, though it is a fairly good one. Getting this exactly
1181 * right would require a more sophisticated collation function that
1182 * takes case sensitivity into account. GLib does not currently
1183 * provide such a function.
1185 * Return value: a newly allocated string, that is a
1186 * case independent form of @str.
1189 g_utf8_casefold (const gchar *str,
1195 g_return_val_if_fail (str != NULL, NULL);
1197 result = g_string_new (NULL);
1199 while ((len < 0 || p < str + len) && *p)
1201 gunichar ch = g_utf8_get_char (p);
1204 int end = G_N_ELEMENTS (casefold_table);
1206 if (ch >= casefold_table[start].ch &&
1207 ch <= casefold_table[end - 1].ch)
1211 int half = (start + end) / 2;
1212 if (ch == casefold_table[half].ch)
1214 g_string_append (result, casefold_table[half].data);
1217 else if (half == start)
1219 else if (ch > casefold_table[half].ch)
1226 g_string_append_unichar (result, g_unichar_tolower (ch));
1229 p = g_utf8_next_char (p);
1232 return g_string_free (result, FALSE);
1236 * g_unichar_get_mirror_char:
1237 * @ch: a Unicode character
1238 * @mirrored_ch: location to store the mirrored character
1240 * In Unicode, some characters are <firstterm>mirrored</firstterm>. This
1241 * means that their images are mirrored horizontally in text that is laid
1242 * out from right to left. For instance, "(" would become its mirror image,
1243 * ")", in right-to-left text.
1245 * If @ch has the Unicode mirrored property and there is another unicode
1246 * character that typically has a glyph that is the mirror image of @ch's
1247 * glyph and @mirrored_ch is set, it puts that character in the address
1248 * pointed to by @mirrored_ch. Otherwise the original character is put.
1250 * Return value: %TRUE if @ch has a mirrored character, %FALSE otherwise
1255 g_unichar_get_mirror_char (gunichar ch,
1256 gunichar *mirrored_ch)
1261 mirrored = GLIB_GET_MIRRORING(ch);
1263 found = ch != mirrored;
1265 *mirrored_ch = mirrored;
1271 #define G_SCRIPT_TABLE_MIDPOINT (G_N_ELEMENTS (g_script_table) / 2)
1273 static inline GUnicodeScript
1274 g_unichar_get_script_bsearch (gunichar ch)
1277 int upper = G_N_ELEMENTS (g_script_table) - 1;
1278 static int saved_mid = G_SCRIPT_TABLE_MIDPOINT;
1279 int mid = saved_mid;
1284 if (ch < g_script_table[mid].start)
1286 else if (ch >= g_script_table[mid].start + g_script_table[mid].chars)
1289 return g_script_table[saved_mid = mid].script;
1291 mid = (lower + upper) / 2;
1293 while (lower <= upper);
1295 return G_UNICODE_SCRIPT_UNKNOWN;
1299 * g_unichar_get_script:
1300 * @ch: a Unicode character
1302 * Looks up the #GUnicodeScript for a particular character (as defined
1303 * by Unicode Standard Annex \#24). No check is made for @ch being a
1304 * valid Unicode character; if you pass in invalid character, the
1305 * result is undefined.
1307 * This function is equivalent to pango_script_for_unichar() and the
1308 * two are interchangeable.
1310 * Return value: the #GUnicodeScript for the character.
1315 g_unichar_get_script (gunichar ch)
1317 if (ch < G_EASY_SCRIPTS_RANGE)
1318 return g_script_easy_table[ch];
1320 return g_unichar_get_script_bsearch (ch);
1324 /* http://unicode.org/iso15924/ */
1325 static const guint32 iso15924_tags[] =
1327 #define PACK(a,b,c,d) ((guint32)((((guint8)(a))<<24)|(((guint8)(b))<<16)|(((guint8)(c))<<8)|((guint8)(d))))
1329 PACK ('Z','y','y','y'), /* G_UNICODE_SCRIPT_COMMON */
1330 PACK ('Z','i','n','h'), /* G_UNICODE_SCRIPT_INHERITED */
1331 PACK ('A','r','a','b'), /* G_UNICODE_SCRIPT_ARABIC */
1332 PACK ('A','r','m','n'), /* G_UNICODE_SCRIPT_ARMENIAN */
1333 PACK ('B','e','n','g'), /* G_UNICODE_SCRIPT_BENGALI */
1334 PACK ('B','o','p','o'), /* G_UNICODE_SCRIPT_BOPOMOFO */
1335 PACK ('C','h','e','r'), /* G_UNICODE_SCRIPT_CHEROKEE */
1336 PACK ('C','o','p','t'), /* G_UNICODE_SCRIPT_COPTIC */
1337 PACK ('C','y','r','l'), /* G_UNICODE_SCRIPT_CYRILLIC */
1338 PACK ('D','s','r','t'), /* G_UNICODE_SCRIPT_DESERET */
1339 PACK ('D','e','v','a'), /* G_UNICODE_SCRIPT_DEVANAGARI */
1340 PACK ('E','t','h','i'), /* G_UNICODE_SCRIPT_ETHIOPIC */
1341 PACK ('G','e','o','r'), /* G_UNICODE_SCRIPT_GEORGIAN */
1342 PACK ('G','o','t','h'), /* G_UNICODE_SCRIPT_GOTHIC */
1343 PACK ('G','r','e','k'), /* G_UNICODE_SCRIPT_GREEK */
1344 PACK ('G','u','j','r'), /* G_UNICODE_SCRIPT_GUJARATI */
1345 PACK ('G','u','r','u'), /* G_UNICODE_SCRIPT_GURMUKHI */
1346 PACK ('H','a','n','i'), /* G_UNICODE_SCRIPT_HAN */
1347 PACK ('H','a','n','g'), /* G_UNICODE_SCRIPT_HANGUL */
1348 PACK ('H','e','b','r'), /* G_UNICODE_SCRIPT_HEBREW */
1349 PACK ('H','i','r','a'), /* G_UNICODE_SCRIPT_HIRAGANA */
1350 PACK ('K','n','d','a'), /* G_UNICODE_SCRIPT_KANNADA */
1351 PACK ('K','a','n','a'), /* G_UNICODE_SCRIPT_KATAKANA */
1352 PACK ('K','h','m','r'), /* G_UNICODE_SCRIPT_KHMER */
1353 PACK ('L','a','o','o'), /* G_UNICODE_SCRIPT_LAO */
1354 PACK ('L','a','t','n'), /* G_UNICODE_SCRIPT_LATIN */
1355 PACK ('M','l','y','m'), /* G_UNICODE_SCRIPT_MALAYALAM */
1356 PACK ('M','o','n','g'), /* G_UNICODE_SCRIPT_MONGOLIAN */
1357 PACK ('M','y','m','r'), /* G_UNICODE_SCRIPT_MYANMAR */
1358 PACK ('O','g','a','m'), /* G_UNICODE_SCRIPT_OGHAM */
1359 PACK ('I','t','a','l'), /* G_UNICODE_SCRIPT_OLD_ITALIC */
1360 PACK ('O','r','y','a'), /* G_UNICODE_SCRIPT_ORIYA */
1361 PACK ('R','u','n','r'), /* G_UNICODE_SCRIPT_RUNIC */
1362 PACK ('S','i','n','h'), /* G_UNICODE_SCRIPT_SINHALA */
1363 PACK ('S','y','r','c'), /* G_UNICODE_SCRIPT_SYRIAC */
1364 PACK ('T','a','m','l'), /* G_UNICODE_SCRIPT_TAMIL */
1365 PACK ('T','e','l','u'), /* G_UNICODE_SCRIPT_TELUGU */
1366 PACK ('T','h','a','a'), /* G_UNICODE_SCRIPT_THAANA */
1367 PACK ('T','h','a','i'), /* G_UNICODE_SCRIPT_THAI */
1368 PACK ('T','i','b','t'), /* G_UNICODE_SCRIPT_TIBETAN */
1369 PACK ('C','a','n','s'), /* G_UNICODE_SCRIPT_CANADIAN_ABORIGINAL */
1370 PACK ('Y','i','i','i'), /* G_UNICODE_SCRIPT_YI */
1371 PACK ('T','g','l','g'), /* G_UNICODE_SCRIPT_TAGALOG */
1372 PACK ('H','a','n','o'), /* G_UNICODE_SCRIPT_HANUNOO */
1373 PACK ('B','u','h','d'), /* G_UNICODE_SCRIPT_BUHID */
1374 PACK ('T','a','g','b'), /* G_UNICODE_SCRIPT_TAGBANWA */
1376 /* Unicode-4.0 additions */
1377 PACK ('B','r','a','i'), /* G_UNICODE_SCRIPT_BRAILLE */
1378 PACK ('C','p','r','t'), /* G_UNICODE_SCRIPT_CYPRIOT */
1379 PACK ('L','i','m','b'), /* G_UNICODE_SCRIPT_LIMBU */
1380 PACK ('O','s','m','a'), /* G_UNICODE_SCRIPT_OSMANYA */
1381 PACK ('S','h','a','w'), /* G_UNICODE_SCRIPT_SHAVIAN */
1382 PACK ('L','i','n','b'), /* G_UNICODE_SCRIPT_LINEAR_B */
1383 PACK ('T','a','l','e'), /* G_UNICODE_SCRIPT_TAI_LE */
1384 PACK ('U','g','a','r'), /* G_UNICODE_SCRIPT_UGARITIC */
1386 /* Unicode-4.1 additions */
1387 PACK ('T','a','l','u'), /* G_UNICODE_SCRIPT_NEW_TAI_LUE */
1388 PACK ('B','u','g','i'), /* G_UNICODE_SCRIPT_BUGINESE */
1389 PACK ('G','l','a','g'), /* G_UNICODE_SCRIPT_GLAGOLITIC */
1390 PACK ('T','f','n','g'), /* G_UNICODE_SCRIPT_TIFINAGH */
1391 PACK ('S','y','l','o'), /* G_UNICODE_SCRIPT_SYLOTI_NAGRI */
1392 PACK ('X','p','e','o'), /* G_UNICODE_SCRIPT_OLD_PERSIAN */
1393 PACK ('K','h','a','r'), /* G_UNICODE_SCRIPT_KHAROSHTHI */
1395 /* Unicode-5.0 additions */
1396 PACK ('Z','z','z','z'), /* G_UNICODE_SCRIPT_UNKNOWN */
1397 PACK ('B','a','l','i'), /* G_UNICODE_SCRIPT_BALINESE */
1398 PACK ('X','s','u','x'), /* G_UNICODE_SCRIPT_CUNEIFORM */
1399 PACK ('P','h','n','x'), /* G_UNICODE_SCRIPT_PHOENICIAN */
1400 PACK ('P','h','a','g'), /* G_UNICODE_SCRIPT_PHAGS_PA */
1401 PACK ('N','k','o','o'), /* G_UNICODE_SCRIPT_NKO */
1403 /* Unicode-5.1 additions */
1404 PACK ('K','a','l','i'), /* G_UNICODE_SCRIPT_KAYAH_LI */
1405 PACK ('L','e','p','c'), /* G_UNICODE_SCRIPT_LEPCHA */
1406 PACK ('R','j','n','g'), /* G_UNICODE_SCRIPT_REJANG */
1407 PACK ('S','u','n','d'), /* G_UNICODE_SCRIPT_SUNDANESE */
1408 PACK ('S','a','u','r'), /* G_UNICODE_SCRIPT_SAURASHTRA */
1409 PACK ('C','h','a','m'), /* G_UNICODE_SCRIPT_CHAM */
1410 PACK ('O','l','c','k'), /* G_UNICODE_SCRIPT_OL_CHIKI */
1411 PACK ('V','a','i','i'), /* G_UNICODE_SCRIPT_VAI */
1412 PACK ('C','a','r','i'), /* G_UNICODE_SCRIPT_CARIAN */
1413 PACK ('L','y','c','i'), /* G_UNICODE_SCRIPT_LYCIAN */
1414 PACK ('L','y','d','i'), /* G_UNICODE_SCRIPT_LYDIAN */
1416 /* Unicode-5.2 additions */
1417 PACK ('A','v','s','t'), /* G_UNICODE_SCRIPT_AVESTAN */
1418 PACK ('B','a','m','u'), /* G_UNICODE_SCRIPT_BAMUM */
1419 PACK ('E','g','y','p'), /* G_UNICODE_SCRIPT_EGYPTIAN_HIEROGLYPHS */
1420 PACK ('A','r','m','i'), /* G_UNICODE_SCRIPT_IMPERIAL_ARAMAIC */
1421 PACK ('P','h','l','i'), /* G_UNICODE_SCRIPT_INSCRIPTIONAL_PAHLAVI */
1422 PACK ('P','r','t','i'), /* G_UNICODE_SCRIPT_INSCRIPTIONAL_PARTHIAN */
1423 PACK ('J','a','v','a'), /* G_UNICODE_SCRIPT_JAVANESE */
1424 PACK ('K','t','h','i'), /* G_UNICODE_SCRIPT_KAITHI */
1425 PACK ('L','i','s','u'), /* G_UNICODE_SCRIPT_LISU */
1426 PACK ('M','t','e','i'), /* G_UNICODE_SCRIPT_MEETEI_MAYEK */
1427 PACK ('S','a','r','b'), /* G_UNICODE_SCRIPT_OLD_SOUTH_ARABIAN */
1428 PACK ('O','r','k','h'), /* G_UNICODE_SCRIPT_OLD_TURKIC */
1429 PACK ('S','a','m','r'), /* G_UNICODE_SCRIPT_SAMARITAN */
1430 PACK ('L','a','n','a'), /* G_UNICODE_SCRIPT_TAI_THAM */
1431 PACK ('T','a','v','t'), /* G_UNICODE_SCRIPT_TAI_VIET */
1433 /* Unicode-6.0 additions */
1434 PACK ('B','a','t','k'), /* G_UNICODE_SCRIPT_BATAK */
1435 PACK ('B','r','a','h'), /* G_UNICODE_SCRIPT_BRAHMI */
1436 PACK ('M','a','n','d'), /* G_UNICODE_SCRIPT_MANDAIC */
1438 /* Unicode-6.1 additions */
1439 PACK ('C','a','k','m'), /* G_UNICODE_SCRIPT_CHAKMA */
1440 PACK ('M','e','r','c'), /* G_UNICODE_SCRIPT_MEROITIC_CURSIVE */
1441 PACK ('M','e','r','o'), /* G_UNICODE_SCRIPT_MEROITIC_HIEROGLYPHS */
1442 PACK ('P','l','r','d'), /* G_UNICODE_SCRIPT_MIAO */
1443 PACK ('S','h','r','d'), /* G_UNICODE_SCRIPT_SHARADA */
1444 PACK ('S','o','r','a'), /* G_UNICODE_SCRIPT_SORA_SOMPENG */
1445 PACK ('T','a','k','r'), /* G_UNICODE_SCRIPT_TAKRI */
1451 * g_unicode_script_to_iso15924:
1452 * @script: a Unicode script
1454 * Looks up the ISO 15924 code for @script. ISO 15924 assigns four-letter
1455 * codes to scripts. For example, the code for Arabic is 'Arab'. The
1456 * four letter codes are encoded as a @guint32 by this function in a
1457 * big-endian fashion. That is, the code returned for Arabic is
1458 * 0x41726162 (0x41 is ASCII code for 'A', 0x72 is ASCII code for 'r', etc).
1460 * See <ulink url="http://unicode.org/iso15924/codelists.html">Codes for the
1461 * representation of names of scripts</ulink> for details.
1463 * Return value: the ISO 15924 code for @script, encoded as an integer,
1464 * of zero if @script is %G_UNICODE_SCRIPT_INVALID_CODE or
1465 * ISO 15924 code 'Zzzz' (script code for UNKNOWN) if @script is not understood.
1470 g_unicode_script_to_iso15924 (GUnicodeScript script)
1472 if (G_UNLIKELY (script == G_UNICODE_SCRIPT_INVALID_CODE))
1475 if (G_UNLIKELY (script < 0 || script >= (int) G_N_ELEMENTS (iso15924_tags)))
1478 return iso15924_tags[script];
1482 * g_unicode_script_from_iso15924:
1483 * @iso15924: a Unicode script
1485 * Looks up the Unicode script for @iso15924. ISO 15924 assigns four-letter
1486 * codes to scripts. For example, the code for Arabic is 'Arab'.
1487 * This function accepts four letter codes encoded as a @guint32 in a
1488 * big-endian fashion. That is, the code expected for Arabic is
1489 * 0x41726162 (0x41 is ASCII code for 'A', 0x72 is ASCII code for 'r', etc).
1491 * See <ulink url="http://unicode.org/iso15924/codelists.html">Codes for the
1492 * representation of names of scripts</ulink> for details.
1494 * Return value: the Unicode script for @iso15924, or
1495 * of %G_UNICODE_SCRIPT_INVALID_CODE if @iso15924 is zero and
1496 * %G_UNICODE_SCRIPT_UNKNOWN if @iso15924 is unknown.
1501 g_unicode_script_from_iso15924 (guint32 iso15924)
1506 return G_UNICODE_SCRIPT_INVALID_CODE;
1508 for (i = 0; i < G_N_ELEMENTS (iso15924_tags); i++)
1509 if (iso15924_tags[i] == iso15924)
1510 return (GUnicodeScript) i;
1512 return G_UNICODE_SCRIPT_UNKNOWN;