Coverage Report

Created: 2025-07-17 06:51

/src/libidn/lib/gl/unistr/u8-check.c
Line
Count
Source
1
/* Check UTF-8 string.
2
   Copyright (C) 2002, 2006-2007, 2009-2025 Free Software Foundation, Inc.
3
   Written by Bruno Haible <bruno@clisp.org>, 2002.
4
5
   This file is free software: you can redistribute it and/or modify
6
   it under the terms of the GNU Lesser General Public License as
7
   published by the Free Software Foundation; either version 2.1 of the
8
   License, or (at your option) any later version.
9
10
   This file 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
13
   GNU Lesser General Public License for more details.
14
15
   You should have received a copy of the GNU Lesser General Public License
16
   along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
17
18
/* Don't use the const-improved function macros in this compilation unit.  */
19
#define _LIBUNISTRING_NO_CONST_GENERICS
20
21
#include <config.h>
22
23
/* Specification.  */
24
#include "unistr.h"
25
26
const uint8_t *
27
u8_check (const uint8_t *s, size_t n)
28
195k
{
29
195k
  const uint8_t *s_end = s + n;
30
31
6.99M
  while (s < s_end)
32
6.81M
    {
33
      /* Keep in sync with unistr.h and u8-mbtouc-aux.c.  */
34
6.81M
      uint8_t c = *s;
35
36
6.81M
      if (c < 0x80)
37
1.16M
        {
38
1.16M
          s++;
39
1.16M
          continue;
40
1.16M
        }
41
5.64M
      if (c >= 0xc2)
42
5.64M
        {
43
5.64M
          if (c < 0xe0)
44
1.64M
            {
45
1.64M
              if (s + 2 <= s_end
46
1.64M
                  && (s[1] ^ 0x80) < 0x40)
47
1.64M
                {
48
1.64M
                  s += 2;
49
1.64M
                  continue;
50
1.64M
                }
51
1.64M
            }
52
4.00M
          else if (c < 0xf0)
53
3.96M
            {
54
3.96M
              if (s + 3 <= s_end
55
3.96M
                  && (s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
56
3.96M
                  && (c >= 0xe1 || s[1] >= 0xa0)
57
3.96M
                  && (c != 0xed || s[1] < 0xa0))
58
3.96M
                {
59
3.96M
                  s += 3;
60
3.96M
                  continue;
61
3.96M
                }
62
3.96M
            }
63
38.8k
          else if (c <= 0xf4)
64
32.5k
            {
65
32.5k
              if (s + 4 <= s_end
66
32.5k
                  && (s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
67
32.5k
                  && (s[3] ^ 0x80) < 0x40
68
32.5k
                  && (c >= 0xf1 || s[1] >= 0x90)
69
32.5k
                  && (c < 0xf4 || (/* c == 0xf4 && */ s[1] < 0x90)))
70
31.8k
                {
71
31.8k
                  s += 4;
72
31.8k
                  continue;
73
31.8k
                }
74
32.5k
            }
75
5.64M
        }
76
      /* invalid or incomplete multibyte character */
77
10.1k
      return s;
78
5.64M
    }
79
185k
  return NULL;
80
195k
}