1/* Look at first character in UTF-8 string.
2   Copyright (C) 1999-2000, 2002, 2006-2007, 2009-2010 Free Software
3   Foundation, Inc.
4   Written by Bruno Haible <bruno@clisp.org>, 2002.
5
6   This program is free software: you can redistribute it and/or modify it
7   under the terms of the GNU Lesser General Public License as published
8   by the Free Software Foundation; either version 3 of the License, or
9   (at your option) any later version.
10
11   This program is distributed in the hope that it will be useful,
12   but WITHOUT ANY WARRANTY; without even the implied warranty of
13   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
14   Lesser General Public License for more details.
15
16   You should have received a copy of the GNU Lesser General Public License
17   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
18
19#include <config.h>
20
21/* Specification.  */
22#include "unistr.h"
23
24int
25u8_strmbtouc (ucs4_t *puc, const uint8_t *s)
26{
27  /* Keep in sync with unistr.h and utf8-ucs4.c.  */
28  uint8_t c = *s;
29
30  if (c < 0x80)
31    {
32      *puc = c;
33      return (c != 0 ? 1 : 0);
34    }
35  if (c >= 0xc2)
36    {
37      if (c < 0xe0)
38        {
39#if CONFIG_UNICODE_SAFETY
40          if ((s[1] ^ 0x80) < 0x40)
41#else
42          if (s[1] != 0)
43#endif
44            {
45              *puc = ((unsigned int) (c & 0x1f) << 6)
46                     | (unsigned int) (s[1] ^ 0x80);
47              return 2;
48            }
49        }
50      else if (c < 0xf0)
51        {
52#if CONFIG_UNICODE_SAFETY
53          if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
54              && (c >= 0xe1 || s[1] >= 0xa0)
55              && (c != 0xed || s[1] < 0xa0))
56#else
57          if (s[1] != 0 && s[2] != 0)
58#endif
59            {
60              *puc = ((unsigned int) (c & 0x0f) << 12)
61                     | ((unsigned int) (s[1] ^ 0x80) << 6)
62                     | (unsigned int) (s[2] ^ 0x80);
63              return 3;
64            }
65        }
66      else if (c < 0xf8)
67        {
68#if CONFIG_UNICODE_SAFETY
69          if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
70              && (s[3] ^ 0x80) < 0x40
71              && (c >= 0xf1 || s[1] >= 0x90)
72#if 1
73              && (c < 0xf4 || (c == 0xf4 && s[1] < 0x90))
74#endif
75             )
76#else
77          if (s[1] != 0 && s[2] != 0 && s[3] != 0)
78#endif
79            {
80              *puc = ((unsigned int) (c & 0x07) << 18)
81                     | ((unsigned int) (s[1] ^ 0x80) << 12)
82                     | ((unsigned int) (s[2] ^ 0x80) << 6)
83                     | (unsigned int) (s[3] ^ 0x80);
84              return 4;
85            }
86        }
87#if 0
88      else if (c < 0xfc)
89        {
90#if CONFIG_UNICODE_SAFETY
91          if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
92              && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40
93              && (c >= 0xf9 || s[1] >= 0x88))
94#else
95          if (s[1] != 0 && s[2] != 0 && s[3] != 0 && s[4] != 0)
96#endif
97            {
98              *puc = ((unsigned int) (c & 0x03) << 24)
99                     | ((unsigned int) (s[1] ^ 0x80) << 18)
100                     | ((unsigned int) (s[2] ^ 0x80) << 12)
101                     | ((unsigned int) (s[3] ^ 0x80) << 6)
102                     | (unsigned int) (s[4] ^ 0x80);
103              return 5;
104            }
105        }
106      else if (c < 0xfe)
107        {
108#if CONFIG_UNICODE_SAFETY
109          if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
110              && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40
111              && (s[5] ^ 0x80) < 0x40
112              && (c >= 0xfd || s[1] >= 0x84))
113#else
114          if (s[1] != 0 && s[2] != 0 && s[3] != 0 && s[4] != 0 && s[5] != 0)
115#endif
116            {
117              *puc = ((unsigned int) (c & 0x01) << 30)
118                     | ((unsigned int) (s[1] ^ 0x80) << 24)
119                     | ((unsigned int) (s[2] ^ 0x80) << 18)
120                     | ((unsigned int) (s[3] ^ 0x80) << 12)
121                     | ((unsigned int) (s[4] ^ 0x80) << 6)
122                     | (unsigned int) (s[5] ^ 0x80);
123              return 6;
124            }
125        }
126#endif
127    }
128  /* invalid or incomplete multibyte character */
129  return -1;
130}
131