1 /* Conversion UTF-8 to UCS-4.
2 Copyright (C) 2001-2002, 2006-2007 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/>. */
26 u8_mbtouc_unsafe_aux (ucs4_t *puc, const uint8_t *s, size_t n)
36 #if CONFIG_UNICODE_SAFETY
37 if ((s[1] ^ 0x80) < 0x40)
40 *puc = ((unsigned int) (c & 0x1f) << 6)
41 | (unsigned int) (s[1] ^ 0x80);
44 /* invalid multibyte character */
48 /* incomplete multibyte character */
57 #if CONFIG_UNICODE_SAFETY
58 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
59 && (c >= 0xe1 || s[1] >= 0xa0)
60 && (c != 0xed || s[1] < 0xa0))
63 *puc = ((unsigned int) (c & 0x0f) << 12)
64 | ((unsigned int) (s[1] ^ 0x80) << 6)
65 | (unsigned int) (s[2] ^ 0x80);
68 /* invalid multibyte character */
72 /* incomplete multibyte character */
81 #if CONFIG_UNICODE_SAFETY
82 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
83 && (s[3] ^ 0x80) < 0x40
84 && (c >= 0xf1 || s[1] >= 0x90)
86 && (c < 0xf4 || (c == 0xf4 && s[1] < 0x90))
91 *puc = ((unsigned int) (c & 0x07) << 18)
92 | ((unsigned int) (s[1] ^ 0x80) << 12)
93 | ((unsigned int) (s[2] ^ 0x80) << 6)
94 | (unsigned int) (s[3] ^ 0x80);
97 /* invalid multibyte character */
101 /* incomplete multibyte character */
111 #if CONFIG_UNICODE_SAFETY
112 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
113 && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40
114 && (c >= 0xf9 || s[1] >= 0x88))
117 *puc = ((unsigned int) (c & 0x03) << 24)
118 | ((unsigned int) (s[1] ^ 0x80) << 18)
119 | ((unsigned int) (s[2] ^ 0x80) << 12)
120 | ((unsigned int) (s[3] ^ 0x80) << 6)
121 | (unsigned int) (s[4] ^ 0x80);
124 /* invalid multibyte character */
128 /* incomplete multibyte character */
137 #if CONFIG_UNICODE_SAFETY
138 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
139 && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40
140 && (s[5] ^ 0x80) < 0x40
141 && (c >= 0xfd || s[1] >= 0x84))
144 *puc = ((unsigned int) (c & 0x01) << 30)
145 | ((unsigned int) (s[1] ^ 0x80) << 24)
146 | ((unsigned int) (s[2] ^ 0x80) << 18)
147 | ((unsigned int) (s[3] ^ 0x80) << 12)
148 | ((unsigned int) (s[4] ^ 0x80) << 6)
149 | (unsigned int) (s[5] ^ 0x80);
152 /* invalid multibyte character */
156 /* incomplete multibyte character */
163 /* invalid multibyte character */