1 /* gunicollate.c - Collation
3 * Copyright 2001 Red Hat, Inc.
5 * The Gnome Library is free software; you can redistribute it and/or
6 * modify it under the terms of the GNU Lesser General Public License as
7 * published by the Free Software Foundation; either version 2 of the
8 * License, or (at your option) any later version.
10 * The Gnome Library is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 * Lesser General Public License for more details.
15 * You should have received a copy of the GNU Lesser General Public
16 * License along with the Gnome Library; see the file COPYING.LIB. If not,
17 * write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
18 * Boston, MA 02111-1307, USA.
25 #ifdef __STDC_ISO_10646__
31 extern gunichar *_g_utf8_normalize_wc (const gchar *str,
37 * @str1: a UTF-8 encoded string
38 * @str2: a UTF-8 encoded string
40 * Compares two strings for ordering using the linguistically
41 * correct rules for the current locale. When sorting a large
42 * number of strings, it will be significantly faster to
43 * obtain collation keys with g_utf8_collate_key() and
44 * compare the keys with <function>strcmp()</function> when
45 * sorting instead of sorting the original strings.
47 * Return value: -1 if @str1 compares before @str2, 0 if they
48 * compare equal, 1 if @str1 compares after @str2.
51 g_utf8_collate (const gchar *str1,
56 #ifdef __STDC_ISO_10646__
61 g_return_val_if_fail (str1 != NULL, 0);
62 g_return_val_if_fail (str2 != NULL, 0);
64 str1_norm = _g_utf8_normalize_wc (str1, -1, G_NORMALIZE_ALL_COMPOSE);
65 str2_norm = _g_utf8_normalize_wc (str2, -1, G_NORMALIZE_ALL_COMPOSE);
67 result = wcscoll ((wchar_t *)str1_norm, (wchar_t *)str2_norm);
72 #else /* !__STDC_ISO_10646__ */
78 g_return_val_if_fail (str1 != NULL, 0);
79 g_return_val_if_fail (str2 != NULL, 0);
81 str1_norm = g_utf8_normalize (str1, -1, G_NORMALIZE_ALL_COMPOSE);
82 str2_norm = g_utf8_normalize (str2, -1, G_NORMALIZE_ALL_COMPOSE);
84 if (g_get_charset (&charset))
86 result = strcoll (str1_norm, str2_norm);
90 gchar *str1_locale = g_convert (str1_norm, -1, "UTF-8", charset, NULL, NULL, NULL);
91 gchar *str2_locale = g_convert (str2_norm, -1, "UTF-8", charset, NULL, NULL, NULL);
93 if (str1_locale && str2_locale)
94 result = strcoll (str1_locale, str2_locale);
100 result = strcmp (str1_norm, str2_norm);
102 g_free (str1_locale);
103 g_free (str2_locale);
109 #endif /* __STDC_ISO_10646__ */
114 #ifdef __STDC_ISO_10646__
115 /* We need UTF-8 encoding of numbers to encode the weights if
116 * we are using wcsxfrm. However, we aren't encoding Unicode
117 * characters, so we can't simply use g_unichar_to_utf8.
119 * The following routine is taken (with modification) from GNU
120 * libc's strxfrm routine:
122 * Copyright (C) 1995-1999,2000,2001 Free Software Foundation, Inc.
123 * Written by Ulrich Drepper <drepper@cygnus.com>, 1995.
126 utf8_encode (char *buf, wchar_t val)
140 for (step = 2; step < 6; ++step)
141 if ((val & (~(guint32)0 << (5 * step + 1))) == 0)
147 *buf = (unsigned char) (~0xff >> step);
151 buf[step] = 0x80 | (val & 0x3f);
161 #endif /* __STDC_ISO_10646__ */
164 * g_utf8_collate_key:
165 * @str: a UTF-8 encoded string.
166 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
168 * Converts a string into a collation key that can be compared
169 * with other collation keys using <function>strcmp()</function>.
170 * The results of comparing the collation keys of two strings
171 * with <function>strcmp()</function> will always be the same as
172 * comparing the two original keys with g_utf8_collate().
174 * Return value: a newly allocated string. This string should
175 * be freed with g_free() when you are done with it.
178 g_utf8_collate_key (const gchar *str,
184 #ifdef __STDC_ISO_10646__
189 size_t result_len = 0;
191 g_return_val_if_fail (str != NULL, NULL);
193 str_norm = _g_utf8_normalize_wc (str, len, G_NORMALIZE_ALL_COMPOSE);
195 setlocale (LC_COLLATE, "");
197 xfrm_len = wcsxfrm (NULL, (wchar_t *)str_norm, 0);
198 result_wc = g_new (wchar_t, xfrm_len + 1);
199 wcsxfrm (result_wc, (wchar_t *)str_norm, xfrm_len + 1);
201 for (i=0; i < xfrm_len; i++)
202 result_len += utf8_encode (NULL, result_wc[i]);
204 result = g_malloc (result_len + 1);
206 for (i=0; i < xfrm_len; i++)
207 result_len += utf8_encode (result + result_len, result_wc[i]);
209 result[result_len] = '\0';
215 #else /* !__STDC_ISO_10646__ */
217 const gchar *charset;
220 g_return_val_if_fail (str != NULL, NULL);
222 str_norm = g_utf8_normalize (str, len, G_NORMALIZE_ALL_COMPOSE);
224 if (g_get_charset (&charset))
226 xfrm_len = strxfrm (NULL, str_norm, 0);
227 result = g_malloc (xfrm_len + 1);
228 strxfrm (result, str_norm, xfrm_len + 1);
232 gchar *str_locale = g_convert (str_norm, -1, "UTF-8", charset, NULL, NULL, NULL);
236 xfrm_len = strxfrm (NULL, str_locale, 0);
237 result = g_malloc (xfrm_len + 2);
239 strxfrm (result + 1, str_locale, xfrm_len + 1);
245 xfrm_len = strlen (str_norm);
246 result = g_malloc (xfrm_len + 2);
248 memcpy (result + 1, str_norm, xfrm_len);
249 result[xfrm_len+1] = '\0';
254 #endif /* __STDC_ISO_10646__ */