1 /* Conversion UTF-8 to UCS-4.
2 Copyright (C) 2001-2002, 2006-2007, 2009-2010 Free Software Foundation, Inc.
3 Written by Bruno Haible <bruno@clisp.org>, 2001.
5 This program is free software: you can redistribute it and/or modify it
6 under the terms of the GNU Lesser General Public License as published
7 by the Free Software Foundation; either version 3 of the License, or
8 (at your option) any later version.
10 This program 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 License
16 along with this program. If not, see <http://www.gnu.org/licenses/>. */
23 #if defined IN_LIBUNISTRING || HAVE_INLINE
26 u8_mbtouc_aux (ucs4_t *puc, const uint8_t *s, size_t n)
36 if ((s[1] ^ 0x80) < 0x40)
38 *puc = ((unsigned int) (c & 0x1f) << 6)
39 | (unsigned int) (s[1] ^ 0x80);
42 /* invalid multibyte character */
46 /* incomplete multibyte character */
55 if ((s[1] ^ 0x80) < 0x40)
57 if ((s[2] ^ 0x80) < 0x40)
59 if ((c >= 0xe1 || s[1] >= 0xa0)
60 && (c != 0xed || s[1] < 0xa0))
62 *puc = ((unsigned int) (c & 0x0f) << 12)
63 | ((unsigned int) (s[1] ^ 0x80) << 6)
64 | (unsigned int) (s[2] ^ 0x80);
67 /* invalid multibyte character */
71 /* invalid multibyte character */
75 /* invalid multibyte character */
79 /* incomplete multibyte character */
81 if (n == 1 || (s[1] ^ 0x80) >= 0x40)
91 if ((s[1] ^ 0x80) < 0x40)
93 if ((s[2] ^ 0x80) < 0x40)
95 if ((s[3] ^ 0x80) < 0x40)
97 if ((c >= 0xf1 || s[1] >= 0x90)
99 && (c < 0xf4 || (c == 0xf4 && s[1] < 0x90))
103 *puc = ((unsigned int) (c & 0x07) << 18)
104 | ((unsigned int) (s[1] ^ 0x80) << 12)
105 | ((unsigned int) (s[2] ^ 0x80) << 6)
106 | (unsigned int) (s[3] ^ 0x80);
109 /* invalid multibyte character */
113 /* invalid multibyte character */
117 /* invalid multibyte character */
121 /* invalid multibyte character */
125 /* incomplete multibyte character */
127 if (n == 1 || (s[1] ^ 0x80) >= 0x40)
129 else if (n == 2 || (s[2] ^ 0x80) >= 0x40)
140 if ((s[1] ^ 0x80) < 0x40)
142 if ((s[2] ^ 0x80) < 0x40)
144 if ((s[3] ^ 0x80) < 0x40)
146 if ((s[4] ^ 0x80) < 0x40)
148 if (c >= 0xf9 || s[1] >= 0x88)
150 *puc = ((unsigned int) (c & 0x03) << 24)
151 | ((unsigned int) (s[1] ^ 0x80) << 18)
152 | ((unsigned int) (s[2] ^ 0x80) << 12)
153 | ((unsigned int) (s[3] ^ 0x80) << 6)
154 | (unsigned int) (s[4] ^ 0x80);
157 /* invalid multibyte character */
161 /* invalid multibyte character */
165 /* invalid multibyte character */
169 /* invalid multibyte character */
172 /* invalid multibyte character */
176 /* incomplete multibyte character */
185 if ((s[1] ^ 0x80) < 0x40)
187 if ((s[2] ^ 0x80) < 0x40)
189 if ((s[3] ^ 0x80) < 0x40)
191 if ((s[4] ^ 0x80) < 0x40)
193 if ((s[5] ^ 0x80) < 0x40)
195 if (c >= 0xfd || s[1] >= 0x84)
197 *puc = ((unsigned int) (c & 0x01) << 30)
198 | ((unsigned int) (s[1] ^ 0x80) << 24)
199 | ((unsigned int) (s[2] ^ 0x80) << 18)
200 | ((unsigned int) (s[3] ^ 0x80) << 12)
201 | ((unsigned int) (s[4] ^ 0x80) << 6)
202 | (unsigned int) (s[5] ^ 0x80);
205 /* invalid multibyte character */
209 /* invalid multibyte character */
213 /* invalid multibyte character */
217 /* invalid multibyte character */
221 /* invalid multibyte character */
224 /* invalid multibyte character */
228 /* incomplete multibyte character */
235 /* invalid multibyte character */