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.
23 #include "gunichartables.h"
31 #define ATTTABLE(Page, Char) \
32 ((attr_table[Page] == 0) ? 0 : (attr_table[Page][Char]))
34 /* We cheat a bit and cast type values to (char *). We detect these
35 using the &0xff trick. */
36 #define TTYPE(Page, Char) \
37 (((GPOINTER_TO_INT(type_table[Page]) & 0xff) == GPOINTER_TO_INT(type_table[Page])) \
38 ? GPOINTER_TO_INT(type_table[Page]) \
39 : (type_table[Page][Char]))
41 #define TYPE(Char) (((Char) > (G_UNICODE_LAST_CHAR)) ? G_UNICODE_UNASSIGNED : TTYPE ((Char) >> 8, (Char) & 0xff))
43 #define ISDIGIT(Type) ((Type) == G_UNICODE_DECIMAL_NUMBER \
44 || (Type) == G_UNICODE_LETTER_NUMBER \
45 || (Type) == G_UNICODE_OTHER_NUMBER)
47 #define ISALPHA(Type) ((Type) == G_UNICODE_LOWERCASE_LETTER \
48 || (Type) == G_UNICODE_UPPERCASE_LETTER \
49 || (Type) == G_UNICODE_TITLECASE_LETTER \
50 || (Type) == G_UNICODE_MODIFIER_LETTER \
51 || (Type) == G_UNICODE_OTHER_LETTER)
53 #define ISMARK(Type) ((Type) == G_UNICODE_NON_SPACING_MARK || \
54 (Type) == G_UNICODE_COMBINING_MARK || \
55 (Type) == G_UNICODE_ENCLOSING_MARK)
60 * @c: a Unicode character
62 * Determines whether a character is alphanumeric.
63 * Given some UTF-8 text, obtain a character value
64 * with g_utf8_get_char().
66 * Return value: %TRUE if @c is an alphanumeric character
69 g_unichar_isalnum (gunichar c)
72 return ISDIGIT (t) || ISALPHA (t);
77 * @c: a Unicode character
79 * Determines whether a character is alphabetic (i.e. a letter).
80 * Given some UTF-8 text, obtain a character value with
83 * Return value: %TRUE if @c is an alphabetic character
86 g_unichar_isalpha (gunichar c)
95 * @c: a Unicode character
97 * Determines whether a character is a control character.
98 * Given some UTF-8 text, obtain a character value with
101 * Return value: %TRUE if @c is a control character
104 g_unichar_iscntrl (gunichar c)
106 return TYPE (c) == G_UNICODE_CONTROL;
111 * @c: a Unicode character
113 * Determines whether a character is numeric (i.e. a digit). This
114 * covers ASCII 0-9 and also digits in other languages/scripts. Given
115 * some UTF-8 text, obtain a character value with g_utf8_get_char().
117 * Return value: %TRUE if @c is a digit
120 g_unichar_isdigit (gunichar c)
122 return TYPE (c) == G_UNICODE_DECIMAL_NUMBER;
128 * @c: a Unicode character
130 * Determines whether a character is printable and not a space
131 * (returns %FALSE for control characters, format characters, and
132 * spaces). g_unichar_isprint() is similar, but returns %TRUE for
133 * spaces. Given some UTF-8 text, obtain a character value with
136 * Return value: %TRUE if @c is printable unless it's a space
139 g_unichar_isgraph (gunichar c)
142 return (t != G_UNICODE_CONTROL
143 && t != G_UNICODE_FORMAT
144 && t != G_UNICODE_UNASSIGNED
145 && t != G_UNICODE_PRIVATE_USE
146 && t != G_UNICODE_SURROGATE
147 && t != G_UNICODE_SPACE_SEPARATOR);
152 * @c: a Unicode character
154 * Determines whether a character is a lowercase letter.
155 * Given some UTF-8 text, obtain a character value with
158 * Return value: %TRUE if @c is a lowercase letter
161 g_unichar_islower (gunichar c)
163 return TYPE (c) == G_UNICODE_LOWERCASE_LETTER;
169 * @c: a Unicode character
171 * Determines whether a character is printable.
172 * Unlike g_unichar_isgraph(), returns %TRUE for spaces.
173 * Given some UTF-8 text, obtain a character value with
176 * Return value: %TRUE if @c is printable
179 g_unichar_isprint (gunichar c)
182 return (t != G_UNICODE_CONTROL
183 && t != G_UNICODE_FORMAT
184 && t != G_UNICODE_UNASSIGNED
185 && t != G_UNICODE_PRIVATE_USE
186 && t != G_UNICODE_SURROGATE);
191 * @c: a Unicode character
193 * Determines whether a character is punctuation.
194 * Given some UTF-8 text, obtain a character value with
197 * Return value: %TRUE if @c is a punctuation character
200 g_unichar_ispunct (gunichar c)
203 return (t == G_UNICODE_CONNECT_PUNCTUATION || t == G_UNICODE_DASH_PUNCTUATION
204 || t == G_UNICODE_CLOSE_PUNCTUATION || t == G_UNICODE_FINAL_PUNCTUATION
205 || t == G_UNICODE_INITIAL_PUNCTUATION || t == G_UNICODE_OTHER_PUNCTUATION
206 || t == G_UNICODE_OPEN_PUNCTUATION);
211 * @c: a Unicode character
213 * Determines whether a character is a space, tab, or line separator
214 * (newline, carriage return, etc.). Given some UTF-8 text, obtain a
215 * character value with g_utf8_get_char().
217 * (Note: don't use this to do word breaking; you have to use
218 * Pango or equivalent to get word breaking right, the algorithm
219 * is fairly complex.)
221 * Return value: %TRUE if @c is a punctuation character
224 g_unichar_isspace (gunichar c)
228 /* special-case these since Unicode thinks they are not spaces */
233 case '\v': /* vertical tab - as if anyone has ever used this... */
240 return (t == G_UNICODE_SPACE_SEPARATOR || t == G_UNICODE_LINE_SEPARATOR
241 || t == G_UNICODE_PARAGRAPH_SEPARATOR);
249 * @c: a unicode character
251 * Determines if a character is uppercase.
253 * Return value: %TRUE if @c is an uppercase character.
256 g_unichar_isupper (gunichar c)
258 return TYPE (c) == G_UNICODE_UPPERCASE_LETTER;
263 * @c: a unicode character
265 * Determines if a character is titlecase. Some characters in
266 * Unicode which are composites, such as the DZ digraph
267 * have three case variants instead of just two. The titlecase
268 * form is used at the beginning of a word where only the
269 * first letter is capitalized. The titlecase form of the DZ
270 * digraph is U+01F2 LATIN CAPITAL LETTTER D WITH SMALL LETTER Z
272 * Return value: %TRUE if the character is titlecase.
275 g_unichar_istitle (gunichar c)
278 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
279 if (title_table[i][0] == c)
285 * g_unichar_isxdigit:
286 * @c: a unicode character.
288 * Determines if a characters is a hexidecimal digit
290 * Return value: %TRUE if the character is a hexadecimal digit.
293 g_unichar_isxdigit (gunichar c)
296 return ((c >= 'a' && c <= 'f')
297 || (c >= 'A' && c <= 'F')
302 * g_unichar_isdefined:
303 * @c: a unicode character
305 * Determines if a given character is assigned in the Unicode
308 * Return value: %TRUE if the character has an assigned value.
311 g_unichar_isdefined (gunichar c)
314 return t != G_UNICODE_UNASSIGNED;
319 * @c: a unicode character
321 * Determines if a character is typically rendered in a double-width
324 * Return value: %TRUE if the character is wide.
326 /* This function stolen from Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>. */
328 g_unichar_iswide (gunichar c)
333 return ((c >= 0x1100 && c <= 0x115f) /* Hangul Jamo */
334 || (c >= 0x2e80 && c <= 0xa4cf && (c & ~0x0011) != 0x300a &&
335 c != 0x303f) /* CJK ... Yi */
336 || (c >= 0xac00 && c <= 0xd7a3) /* Hangul Syllables */
337 || (c >= 0xf900 && c <= 0xfaff) /* CJK Compatibility Ideographs */
338 || (c >= 0xfe30 && c <= 0xfe6f) /* CJK Compatibility Forms */
339 || (c >= 0xff00 && c <= 0xff5f) /* Fullwidth Forms */
340 || (c >= 0xffe0 && c <= 0xffe6));
345 * @c: a unicode character
347 * Convert a character to uppercase.
349 * Return value: the result of converting @c to uppercase.
350 * If @c is not an lowercase or titlecase character,
351 * @c is returned unchanged.
354 g_unichar_toupper (gunichar c)
357 if (t == G_UNICODE_LOWERCASE_LETTER)
359 gunichar val = ATTTABLE (c >> 8, c & 0xff);
360 if (val >= 0xd800 && val < 0xdc00)
362 guchar *p = special_case_table[val - 0xd800];
363 return p[0] * 256 + p[1];
368 else if (t == G_UNICODE_TITLECASE_LETTER)
371 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
373 if (title_table[i][0] == c)
374 return title_table[i][1];
382 * @c: a unicode character.
384 * Convert a character to lower case
386 * Return value: the result of converting @c to lower case.
387 * If @c is not an upperlower or titlecase character,
388 * @c is returned unchanged.
391 g_unichar_tolower (gunichar c)
394 if (t == G_UNICODE_UPPERCASE_LETTER)
396 gunichar val = ATTTABLE (c >> 8, c & 0xff);
397 if (val >= 0xd800 && val < 0xdc00)
399 guchar *p = special_case_table[val - 0xd800];
400 return p[0] * 256 + p[1];
405 else if (t == G_UNICODE_TITLECASE_LETTER)
408 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
410 if (title_table[i][0] == c)
411 return title_table[i][2];
419 * @c: a unicode character
421 * Convert a character to the titlecase
423 * Return value: the result of converting @c to titlecase.
424 * If @c is not an uppercase or lowercase character,
425 * @c is returned unchanged.
428 g_unichar_totitle (gunichar c)
431 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
433 if (title_table[i][0] == c || title_table[i][1] == c
434 || title_table[i][2] == c)
435 return title_table[i][0];
437 return (TYPE (c) == G_UNICODE_LOWERCASE_LETTER
438 ? ATTTABLE (c >> 8, c & 0xff)
443 * g_unichar_digit_value:
444 * @c: a unicode character
446 * Determines the numeric value of a character as a decimal
449 * Return value: If @c is a decimal digit (according to
450 * `g_unichar_isdigit'), its numeric value. Otherwise, -1.
453 g_unichar_digit_value (gunichar c)
455 if (TYPE (c) == G_UNICODE_DECIMAL_NUMBER)
456 return ATTTABLE (c >> 8, c & 0xff);
461 * g_unichar_xdigit_value:
462 * @c: a unicode character
464 * Determines the numeric value of a character as a hexidecimal
467 * Return value: If @c is a hex digit (according to
468 * `g_unichar_isxdigit'), its numeric value. Otherwise, -1.
471 g_unichar_xdigit_value (gunichar c)
473 if (c >= 'A' && c <= 'F')
475 if (c >= 'a' && c <= 'f')
477 if (TYPE (c) == G_UNICODE_DECIMAL_NUMBER)
478 return ATTTABLE (c >> 8, c & 0xff);
484 * @c: a unicode character
486 * Classifies a unicode character by type.
488 * Return value: the type of the character.
491 g_unichar_type (gunichar c)
497 * Case mapping functions
507 get_locale_type (void)
509 const char *locale = setlocale (LC_CTYPE, NULL);
514 if (locale[1] == 'z')
515 return LOCALE_TURKIC;
518 if (locale[1] == 't')
519 return LOCALE_LITHUANIAN;
522 if (locale[1] == 'r')
523 return LOCALE_TURKIC;
527 return LOCALE_NORMAL;
531 output_marks (const char **p_inout,
536 const char *p = *p_inout;
540 gunichar c = g_utf8_get_char (p);
545 if (!remove_dot || c != 0x307 /* COMBINING DOT ABOVE */)
546 len += g_unichar_to_utf8 (c, out_buffer ? out_buffer + len : NULL);
547 p = g_utf8_next_char (p);
558 output_special_case (gchar *out_buffer,
564 guchar *p = special_case_table[index];
566 if (type != G_UNICODE_TITLECASE_LETTER)
567 p += 2; /* +2 to skip over "best single match" */
578 gunichar ch = p[0] * 256 + p[1];
582 len += g_unichar_to_utf8 (ch, out_buffer ? out_buffer + len : NULL);
590 real_toupper (const gchar *str,
593 LocaleType locale_type)
595 const gchar *p = str;
596 const char *last = NULL;
598 gboolean last_was_i = FALSE;
600 while ((max_len < 0 || p < str + max_len) && *p)
602 gunichar c = g_utf8_get_char (p);
607 p = g_utf8_next_char (p);
609 if (locale_type == LOCALE_LITHUANIAN)
617 /* Nasty, need to remove any dot above. Though
618 * I think only E WITH DOT ABOVE occurs in practice
619 * which could simplify this considerably.
624 decomp = g_unicode_canonical_decomposition (c, &decomp_len);
625 for (i=0; i < decomp_len; i++)
627 if (decomp[i] != 0x307 /* COMBINING DOT ABOVE */)
628 len += g_unichar_to_utf8 (g_unichar_toupper (decomp[i]), out_buffer ? out_buffer + len : NULL);
632 len = output_marks (&p, out_buffer, len, TRUE);
642 if (locale_type == LOCALE_TURKIC && c == 'i')
644 /* i => LATIN CAPITAL LETTER I WITH DOT ABOVE */
645 len += g_unichar_to_utf8 (0x130, out_buffer ? out_buffer + len : NULL);
647 else if (c == 0x0345) /* COMBINING GREEK YPOGEGRAMMENI */
649 /* Nasty, need to move it after other combining marks .. this would go away if
650 * we normalized first.
652 len = output_marks (&p, out_buffer, len, FALSE);
654 /* And output as GREEK CAPITAL LETTER IOTA */
655 len += g_unichar_to_utf8 (0x399, out_buffer ? out_buffer + len : NULL);
657 else if (t == G_UNICODE_LOWERCASE_LETTER || t == G_UNICODE_TITLECASE_LETTER)
659 val = ATTTABLE (c >> 8, c & 0xff);
661 if (val >= 0xd800 && val < 0xdc00)
663 len += output_special_case (out_buffer, len, val - 0xd800, t,
664 t == G_UNICODE_LOWERCASE_LETTER ? 0 : 1);
668 if (t == G_UNICODE_TITLECASE_LETTER)
671 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
673 if (title_table[i][0] == c)
674 val = title_table[i][1];
678 len += g_unichar_to_utf8 (val, out_buffer ? out_buffer + len : NULL);
683 gsize char_len = g_utf8_skip[*(guchar *)last];
686 memcpy (out_buffer + len, last, char_len);
698 * @str: a UTF-8 encoded string
699 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
701 * Converts all Unicode characters in the string that have a case
702 * to uppercase. The exact manner that this is done depends
703 * on the current locale, and may result in the number of
704 * characters in the string increasing. (For instance, the
705 * German ess-zet will be changed to SS.)
707 * Return value: a newly allocated string, with all characters
708 * converted to uppercase.
711 g_utf8_strup (const gchar *str,
715 LocaleType locale_type;
718 g_return_val_if_fail (str != NULL, NULL);
720 locale_type = get_locale_type ();
723 * We use a two pass approach to keep memory management simple
725 result_len = real_toupper (str, len, NULL, locale_type);
726 result = g_malloc (result_len + 1);
727 real_toupper (str, len, result, locale_type);
728 result[result_len] = '\0';
734 real_tolower (const gchar *str,
737 LocaleType locale_type)
739 const gchar *p = str;
740 const char *last = NULL;
743 while ((max_len < 0 || p < str + max_len) && *p)
745 gunichar c = g_utf8_get_char (p);
750 p = g_utf8_next_char (p);
752 if (locale_type == LOCALE_TURKIC && c == 'I')
754 /* I => LATIN SMALL LETTER DOTLESS I */
755 len += g_unichar_to_utf8 (0x131, out_buffer ? out_buffer + len : NULL);
757 else if (c == 0x03A3) /* GREEK CAPITAL LETTER SIGMA */
759 gunichar next_c = g_utf8_get_char (p);
760 int next_t = TYPE(next_c);
762 /* SIGMA mapps differently depending on whether it is
763 * final or not. The following simplified test would
764 * fail in the case of combining marks following the
765 * sigma, but I don't think that occurs in real text.
766 * The test here matches that in ICU.
768 if (ISALPHA(next_t)) /* Lu,Ll,Lt,Lm,Lo */
769 val = 0x3c3; /* GREEK SMALL SIGMA */
771 val = 0x3c2; /* GREEK SMALL FINAL SIGMA */
773 len += g_unichar_to_utf8 (val, out_buffer ? out_buffer + len : NULL);
775 else if (t == G_UNICODE_UPPERCASE_LETTER || t == G_UNICODE_TITLECASE_LETTER)
777 val = ATTTABLE (c >> 8, c & 0xff);
779 if (val >= 0xd800 && val < 0xdc00)
781 len += output_special_case (out_buffer, len, val - 0xd800, t, 0);
785 if (t == G_UNICODE_TITLECASE_LETTER)
788 for (i = 0; i < G_N_ELEMENTS (title_table); ++i)
790 if (title_table[i][0] == c)
791 val = title_table[i][2];
795 len += g_unichar_to_utf8 (val, out_buffer ? out_buffer + len : NULL);
800 gsize char_len = g_utf8_skip[*(guchar *)last];
803 memcpy (out_buffer + len, last, char_len);
815 * @str: a UTF-8 encoded string
816 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
818 * Converts all Unicode characters in the string that have a case
819 * to lowercase. The exact manner that this is done depends
820 * on the current locale, and may result in the number of
821 * characters in the string changing.
823 * Return value: a newly allocated string, with all characters
824 * converted to lowercase.
827 g_utf8_strdown (const gchar *str,
831 LocaleType locale_type;
834 g_return_val_if_fail (str != NULL, NULL);
836 locale_type = get_locale_type ();
839 * We use a two pass approach to keep memory management simple
841 result_len = real_tolower (str, len, NULL, locale_type);
842 result = g_malloc (result_len + 1);
843 real_tolower (str, len, result, locale_type);
844 result[result_len] = '\0';
851 * @str: a UTF-8 encoded string
852 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
854 * Converts a string into a form that is independent of case. The
855 * result will not correspond to any particular case, but can be
856 * compared for equality or ordered with the results of calling
857 * g_utf8_casefold() on other strings.
859 * Note that calling g_utf8_casefold() followed by g_utf8_collate() is
860 * only an approximation to the correct linguistic case insensitive
861 * ordering, though it is a fairly good one. Getting this exactly
862 * right would require a more sophisticated collation function that
863 * takes case sensitivity into account. GLib does not currently
864 * provide such a function.
866 * Return value: a newly allocated string, that is a
867 * case independent form of @str.
870 g_utf8_casefold (const gchar *str,
873 GString *result = g_string_new (NULL);
877 while ((len < 0 || p < str + len) && *p)
879 gunichar ch = g_utf8_get_char (p);
882 int end = G_N_ELEMENTS (casefold_table);
884 if (ch >= casefold_table[start].ch &&
885 ch <= casefold_table[end - 1].ch)
889 int half = (start + end) / 2;
890 if (ch == casefold_table[half].ch)
892 g_string_append (result, casefold_table[half].data);
895 else if (half == start)
897 else if (ch > casefold_table[half].ch)
904 g_string_append_unichar (result, g_unichar_tolower (ch));
907 p = g_utf8_next_char (p);
910 return g_string_free (result, FALSE);