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.
23 #ifdef __STDC_ISO_10646__
29 extern gunichar *_g_utf8_normalize_wc (const gchar *str,
35 * @str1: a UTF-8 encoded string
36 * @str2: a UTF-8 encoded string
38 * Compares two strings for ordering using the linguistically
39 * correct rules for the current locale. When sorting a large
40 * number of strings, it will be significantly faster to
41 * obtain collation keys with g_utf8_collate_key() and
42 * compare the keys with <function>strcmp()</function> when
43 * sorting instead of sorting the original strings.
45 * Return value: -1 if @str1 compares before @str2, 0 if they
46 * compare equal, 1 if @str1 compares after @str2.
49 g_utf8_collate (const gchar *str1,
54 #ifdef __STDC_ISO_10646__
56 gunichar *str1_norm = _g_utf8_normalize_wc (str1, -1, G_NORMALIZE_ALL_COMPOSE);
57 gunichar *str2_norm = _g_utf8_normalize_wc (str2, -1, G_NORMALIZE_ALL_COMPOSE);
59 result = wcscoll ((wchar_t *)str1_norm, (wchar_t *)str2_norm);
64 #else /* !__STDC_ISO_10646__ */
67 gchar *str1_norm = g_utf8_normalize (str1, -1, G_NORMALIZE_ALL_COMPOSE);
68 gchar *str2_norm = g_utf8_normalize (str2, -1, G_NORMALIZE_ALL_COMPOSE);
70 if (g_get_charset (&charset))
72 result = strcoll (str1_norm, str2_norm);
76 gchar *str1_locale = g_convert (str1_norm, -1, "UTF-8", charset, NULL, NULL, NULL);
77 gchar *str2_locale = g_convert (str2_norm, -1, "UTF-8", charset, NULL, NULL, NULL);
79 if (str1_locale && str2_locale)
80 result = strcoll (str1_locale, str2_locale);
86 result = strcmp (str1_norm, str2_norm);
95 #endif /* __STDC_ISO_10646__ */
100 #ifdef __STDC_ISO_10646__
101 /* We need UTF-8 encoding of numbers to encode the weights if
102 * we are using wcsxfrm. However, we aren't encoding Unicode
103 * characters, so we can't simply use g_unichar_to_utf8.
105 * The following routine is taken (with modification) from GNU
106 * libc's strxfrm routine:
108 * Copyright (C) 1995-1999,2000,2001 Free Software Foundation, Inc.
109 * Written by Ulrich Drepper <drepper@cygnus.com>, 1995.
112 utf8_encode (char *buf, wchar_t val)
126 for (step = 2; step < 6; ++step)
127 if ((val & (~(guint32)0 << (5 * step + 1))) == 0)
133 *buf = (unsigned char) (~0xff >> step);
137 buf[step] = 0x80 | (val & 0x3f);
147 #endif /* __STDC_ISO_10646__ */
150 * g_utf8_collate_key:
151 * @str: a UTF-8 encoded string.
152 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
154 * Converts a string into a collation key that can be compared
155 * with other collation keys using <function>strcmp()</function>.
156 * The results of comparing the collation keys of two strings
157 * with <function>strcmp()</function> will always be the same as
158 * comparing the two original keys with g_utf8_collate().
160 * Return value: a newly allocated string. This string should
161 * be freed with g_free() when you are done with it.
164 g_utf8_collate_key (const gchar *str,
170 #ifdef __STDC_ISO_10646__
172 gunichar *str_norm = _g_utf8_normalize_wc (str, len, G_NORMALIZE_ALL_COMPOSE);
175 size_t result_len = 0;
177 setlocale (LC_COLLATE, "");
179 xfrm_len = wcsxfrm (NULL, (wchar_t *)str_norm, 0);
180 result_wc = g_new (wchar_t, xfrm_len + 1);
181 wcsxfrm (result_wc, (wchar_t *)str_norm, xfrm_len + 1);
183 for (i=0; i < xfrm_len; i++)
184 result_len += utf8_encode (NULL, result_wc[i]);
186 result = g_malloc (result_len + 1);
188 for (i=0; i < xfrm_len; i++)
189 result_len += utf8_encode (result + result_len, result_wc[i]);
191 result[result_len] = '\0';
197 #else /* !__STDC_ISO_10646__ */
199 const gchar *charset;
200 gchar *str_norm = g_utf8_normalize (str, len, G_NORMALIZE_ALL_COMPOSE);
202 if (g_get_charset (&charset))
204 xfrm_len = strxfrm (NULL, str_norm, 0);
205 result = g_malloc (xfrm_len + 1);
206 strxfrm (result, str_norm, xfrm_len + 1);
210 gchar *str_locale = g_convert (str_norm, -1, "UTF-8", charset, NULL, NULL, NULL);
214 xfrm_len = strxfrm (NULL, str_locale, 0);
215 result = g_malloc (xfrm_len + 2);
217 strxfrm (result + 1, str_locale, xfrm_len + 1);
223 xfrm_len = strlen (str_norm);
224 result = g_malloc (xfrm_len + 2);
226 memcpy (result + 1, str_norm, xfrm_len);
227 result[xfrm_len+1] = '\0';
232 #endif /* __STDC_ISO_10646__ */