1 /* Look at first character in UTF-8 string.
2 Copyright (C) 1999-2000, 2002, 2006-2007 Free Software Foundation, Inc.
3 Written by Bruno Haible <bruno@clisp.org>, 2002.
5 This program is free software; you can redistribute it and/or modify it
6 under the terms of the GNU Library General Public License as published
7 by the Free Software Foundation; either version 2, or (at your option)
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 Library General Public License for more details.
15 You should have received a copy of the GNU Library General Public
16 License along with this program; if not, write to the Free Software
17 Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,
26 u8_strmbtouc (ucs4_t *puc, const uint8_t *s)
28 /* Keep in sync with utf8-ucs4.h and utf8-ucs4.c. */
34 return (c != 0 ? 1 : 0);
40 #if CONFIG_UNICODE_SAFETY
41 if ((s[1] ^ 0x80) < 0x40)
46 *puc = ((unsigned int) (c & 0x1f) << 6)
47 | (unsigned int) (s[1] ^ 0x80);
53 #if CONFIG_UNICODE_SAFETY
54 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
55 && (c >= 0xe1 || s[1] >= 0xa0)
56 && (c != 0xed || s[1] < 0xa0))
58 if (s[1] != 0 && s[2] != 0)
61 *puc = ((unsigned int) (c & 0x0f) << 12)
62 | ((unsigned int) (s[1] ^ 0x80) << 6)
63 | (unsigned int) (s[2] ^ 0x80);
69 #if CONFIG_UNICODE_SAFETY
70 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
71 && (s[3] ^ 0x80) < 0x40
72 && (c >= 0xf1 || s[1] >= 0x90)
74 && (c < 0xf4 || (c == 0xf4 && s[1] < 0x90))
78 if (s[1] != 0 && s[2] != 0 && s[3] != 0)
81 *puc = ((unsigned int) (c & 0x07) << 18)
82 | ((unsigned int) (s[1] ^ 0x80) << 12)
83 | ((unsigned int) (s[2] ^ 0x80) << 6)
84 | (unsigned int) (s[3] ^ 0x80);
91 #if CONFIG_UNICODE_SAFETY
92 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
93 && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40
94 && (c >= 0xf9 || s[1] >= 0x88))
96 if (s[1] != 0 && s[2] != 0 && s[3] != 0 && s[4] != 0)
99 *puc = ((unsigned int) (c & 0x03) << 24)
100 | ((unsigned int) (s[1] ^ 0x80) << 18)
101 | ((unsigned int) (s[2] ^ 0x80) << 12)
102 | ((unsigned int) (s[3] ^ 0x80) << 6)
103 | (unsigned int) (s[4] ^ 0x80);
109 #if CONFIG_UNICODE_SAFETY
110 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
111 && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40
112 && (s[5] ^ 0x80) < 0x40
113 && (c >= 0xfd || s[1] >= 0x84))
115 if (s[1] != 0 && s[2] != 0 && s[3] != 0 && s[4] != 0 && s[5] != 0)
118 *puc = ((unsigned int) (c & 0x01) << 30)
119 | ((unsigned int) (s[1] ^ 0x80) << 24)
120 | ((unsigned int) (s[2] ^ 0x80) << 18)
121 | ((unsigned int) (s[3] ^ 0x80) << 12)
122 | ((unsigned int) (s[4] ^ 0x80) << 6)
123 | (unsigned int) (s[5] ^ 0x80);
129 /* invalid or incomplete multibyte character */