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,
34 * @str1: a UTF-8 encoded string
35 * @str2: a UTF-8 encoded string
37 * Compares two strings for ordering using the linguistically
38 * correct rules for the current locale. When sorting a large
39 * number of strings, it will be significantly faster to
40 * obtain collation keys with g_utf8_collate_key() and
41 * compare the keys with strcmp() when sorting instead of
42 * sorting the original strings.
44 * Return value: -1 if str1 compares before str2, 0 if they
45 * compare equal, 1 if str1 compares after str2.
48 g_utf8_collate (const gchar *str1,
53 #ifdef __STDC_ISO_10646__
55 gunichar *str1_norm = _g_utf8_normalize_wc (str1, G_NORMALIZE_ALL_COMPOSE);
56 gunichar *str2_norm = _g_utf8_normalize_wc (str2, G_NORMALIZE_ALL_COMPOSE);
58 result = wcscoll ((wchar_t *)str1_norm, (wchar_t *)str2_norm);
63 #else /* !__STDC_ISO_10646__ */
66 gchar *str1_norm = g_utf8_normalize (str1, G_NORMALIZE_ALL_COMPOSE);
67 gchar *str2_norm = g_utf8_normalize (str2, G_NORMALIZE_ALL_COMPOSE);
69 if (g_get_charset (&charset))
71 result = strcoll (str1_norm, str2_norm);
75 gchar *str1_locale = g_convert (str1_norm, -1, "UTF-8", charset, NULL, NULL, NULL);
76 gchar *str2_locale = g_convert (str2_norm, -1, "UTF-8", charset, NULL, NULL, NULL);
78 if (str1_locale && str2_locale)
79 result = strcoll (str1_locale, str2_locale);
85 result = strcmp (str1_norm, str2_norm);
94 #endif /* __STDC_ISO_10646__ */
99 #ifdef __STDC_ISO_10646__
100 /* We need UTF-8 encoding of numbers to encode the weights if
101 * we are using wcsxfrm. However, we aren't encoding Unicode
102 * characters, so we can't simply use g_unichar_to_utf8.
104 * The following routine is taken (with modification) from GNU
105 * libc's strxfrm routine:
107 * Copyright (C) 1995-1999,2000,2001 Free Software Foundation, Inc.
108 * Written by Ulrich Drepper <drepper@cygnus.com>, 1995.
111 utf8_encode (char *buf, wchar_t val)
125 for (step = 2; step < 6; ++step)
126 if ((val & (~(guint32)0 << (5 * step + 1))) == 0)
132 *buf = (unsigned char) (~0xff >> step);
136 buf[step] = 0x80 | (val & 0x3f);
146 #endif /* __STDC_ISO_10646__ */
149 * g_utf8_collate_key:
150 * @str: a UTF-8 encoded string.
152 * Converts a string into a collation key that can be compared
153 * with other collation keys using strcmp(). The results of
154 * comparing the collation keys of two strings with strcmp()
155 * will always be the same as comparing the two original
156 * keys with g_utf8_collate().
158 * Return value: a newly allocated string. This string should
159 * be freed with g_free when you are done with it.
162 g_utf8_collate_key (const gchar *str)
167 #ifdef __STDC_ISO_10646__
169 gunichar *str_norm = _g_utf8_normalize_wc (str, G_NORMALIZE_ALL_COMPOSE);
172 size_t result_len = 0;
174 setlocale (LC_COLLATE, "");
176 len = wcsxfrm (NULL, (wchar_t *)str_norm, 0);
177 result_wc = g_new (wchar_t, len + 1);
178 wcsxfrm (result_wc, (wchar_t *)str_norm, len + 1);
180 for (i=0; i < len; i++)
181 result_len += utf8_encode (NULL, result_wc[i]);
183 result = g_malloc (result_len + 1);
185 for (i=0; i < len; i++)
186 result_len += utf8_encode (result + result_len, result_wc[i]);
188 result[result_len] = '\0';
194 #else /* !__STDC_ISO_10646__ */
196 const gchar *charset;
197 gchar *str_norm = g_utf8_normalize (str, G_NORMALIZE_ALL_COMPOSE);
199 if (g_get_charset (&charset))
201 len = strxfrm (NULL, str_norm, 0);
202 result = g_malloc (len + 1);
203 strxfrm (result, str_norm, len + 1);
207 gchar *str_locale = g_convert (str_norm, -1, "UTF-8", charset, NULL, NULL, NULL);
211 len = strxfrm (NULL, str_locale, 0);
212 result = g_malloc (len + 2);
214 strxfrm (result + 1, str_locale, len + 1);
220 len = strlen (str_norm);
221 result = g_malloc (len + 2);
223 memcpy (result + 1, str_norm, len);
224 result[len+1] = '\0';
230 #endif /* __STDC_ISO_10646__ */