1 /* Look at first character in UTF-8 string.
2 Copyright (C) 1999-2002, 2006-2007, 2009-2013 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/>. */
20 #if defined IN_LIBUNISTRING
21 /* Tell unistr.h to declare u8_mbtouc as 'extern', not 'static inline'. */
22 # include "unistring-notinline.h"
31 u8_mbtouc (ucs4_t *puc, const uint8_t *s, size_t n)
46 if ((s[1] ^ 0x80) < 0x40)
48 *puc = ((unsigned int) (c & 0x1f) << 6)
49 | (unsigned int) (s[1] ^ 0x80);
52 /* invalid multibyte character */
56 /* incomplete multibyte character */
65 if ((s[1] ^ 0x80) < 0x40)
67 if ((s[2] ^ 0x80) < 0x40)
69 if ((c >= 0xe1 || s[1] >= 0xa0)
70 && (c != 0xed || s[1] < 0xa0))
72 *puc = ((unsigned int) (c & 0x0f) << 12)
73 | ((unsigned int) (s[1] ^ 0x80) << 6)
74 | (unsigned int) (s[2] ^ 0x80);
77 /* invalid multibyte character */
81 /* invalid multibyte character */
85 /* invalid multibyte character */
89 /* incomplete multibyte character */
91 if (n == 1 || (s[1] ^ 0x80) >= 0x40)
101 if ((s[1] ^ 0x80) < 0x40)
103 if ((s[2] ^ 0x80) < 0x40)
105 if ((s[3] ^ 0x80) < 0x40)
107 if ((c >= 0xf1 || s[1] >= 0x90)
109 && (c < 0xf4 || (c == 0xf4 && s[1] < 0x90))
113 *puc = ((unsigned int) (c & 0x07) << 18)
114 | ((unsigned int) (s[1] ^ 0x80) << 12)
115 | ((unsigned int) (s[2] ^ 0x80) << 6)
116 | (unsigned int) (s[3] ^ 0x80);
119 /* invalid multibyte character */
123 /* invalid multibyte character */
127 /* invalid multibyte character */
131 /* invalid multibyte character */
135 /* incomplete multibyte character */
137 if (n == 1 || (s[1] ^ 0x80) >= 0x40)
139 else if (n == 2 || (s[2] ^ 0x80) >= 0x40)
150 if ((s[1] ^ 0x80) < 0x40)
152 if ((s[2] ^ 0x80) < 0x40)
154 if ((s[3] ^ 0x80) < 0x40)
156 if ((s[4] ^ 0x80) < 0x40)
158 if (c >= 0xf9 || s[1] >= 0x88)
160 *puc = ((unsigned int) (c & 0x03) << 24)
161 | ((unsigned int) (s[1] ^ 0x80) << 18)
162 | ((unsigned int) (s[2] ^ 0x80) << 12)
163 | ((unsigned int) (s[3] ^ 0x80) << 6)
164 | (unsigned int) (s[4] ^ 0x80);
167 /* invalid multibyte character */
171 /* invalid multibyte character */
175 /* invalid multibyte character */
179 /* invalid multibyte character */
182 /* invalid multibyte character */
186 /* incomplete multibyte character */
195 if ((s[1] ^ 0x80) < 0x40)
197 if ((s[2] ^ 0x80) < 0x40)
199 if ((s[3] ^ 0x80) < 0x40)
201 if ((s[4] ^ 0x80) < 0x40)
203 if ((s[5] ^ 0x80) < 0x40)
205 if (c >= 0xfd || s[1] >= 0x84)
207 *puc = ((unsigned int) (c & 0x01) << 30)
208 | ((unsigned int) (s[1] ^ 0x80) << 24)
209 | ((unsigned int) (s[2] ^ 0x80) << 18)
210 | ((unsigned int) (s[3] ^ 0x80) << 12)
211 | ((unsigned int) (s[4] ^ 0x80) << 6)
212 | (unsigned int) (s[5] ^ 0x80);
215 /* invalid multibyte character */
219 /* invalid multibyte character */
223 /* invalid multibyte character */
227 /* invalid multibyte character */
231 /* invalid multibyte character */
234 /* invalid multibyte character */
238 /* incomplete multibyte character */
245 /* invalid multibyte character */