lib/mbrtowc-impl-utf8.h

   1 /* Convert multibyte character to wide character.
   2    Copyright (C) 1999-2002, 2005-2024 Free Software Foundation, Inc.
   3
   4    This file is free software: you can redistribute it and/or modify
   5    it under the terms of the GNU Lesser General Public License as
   6    published by the Free Software Foundation; either version 2.1 of the
   7    License, or (at your option) any later version.
   8
   9    This file is distributed in the hope that it will be useful,
  10    but WITHOUT ANY WARRANTY; without even the implied warranty of
  11    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12    GNU Lesser General Public License for more details.
  13
  14    You should have received a copy of the GNU Lesser General Public License
  15    along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
  16
  17 /* Written by Bruno Haible <bruno@clisp.org>, 2008.  */
  18
  19 /* This file contains the part of the body of the mbrtowc and mbrtoc32 functions
  20    that handles the special case of the UTF-8 encoding.  */
  21
  22         /* Cf. unistr/u8-mbtouc.c.  */
  23         unsigned char c = (unsigned char) p[0];
  24
  25         if (c < 0x80)
  26           {
  27             if (pwc != NULL)
  28               *pwc = c;
  29             res = (c == 0 ? 0 : 1);
  30             goto success;
  31           }
  32         if (c >= 0xc2)
  33           {
  34             if (c < 0xe0)
  35               {
  36                 if (m == 1)
  37                   goto incomplete;
  38                 else /* m >= 2 */
  39                   {
  40                     unsigned char c2 = (unsigned char) p[1];
  41
  42                     if ((c2 ^ 0x80) < 0x40)
  43                       {
  44                         if (pwc != NULL)
  45                           *pwc = ((unsigned int) (c & 0x1f) << 6)
  46                                  | (unsigned int) (c2 ^ 0x80);
  47                         res = 2;
  48                         goto success;
  49                       }
  50                   }
  51               }
  52             else if (c < 0xf0)
  53               {
  54                 if (m == 1)
  55                   goto incomplete;
  56                 else
  57                   {
  58                     unsigned char c2 = (unsigned char) p[1];
  59
  60                     if ((c2 ^ 0x80) < 0x40
  61                         && (c >= 0xe1 || c2 >= 0xa0)
  62                         && (c != 0xed || c2 < 0xa0))
  63                       {
  64                         if (m == 2)
  65                           goto incomplete;
  66                         else /* m >= 3 */
  67                           {
  68                             unsigned char c3 = (unsigned char) p[2];
  69
  70                             if ((c3 ^ 0x80) < 0x40)
  71                               {
  72                                 unsigned int wc =
  73                                   (((unsigned int) (c & 0x0f) << 12)
  74                                    | ((unsigned int) (c2 ^ 0x80) << 6)
  75                                    | (unsigned int) (c3 ^ 0x80));
  76
  77                                 if (FITS_IN_CHAR_TYPE (wc))
  78                                   {
  79                                     if (pwc != NULL)
  80                                       *pwc = wc;
  81                                     res = 3;
  82                                     goto success;
  83                                   }
  84                               }
  85                           }
  86                       }
  87                   }
  88               }
  89             else if (c <= 0xf4)
  90               {
  91                 if (m == 1)
  92                   goto incomplete;
  93                 else
  94                   {
  95                     unsigned char c2 = (unsigned char) p[1];
  96
  97                     if ((c2 ^ 0x80) < 0x40
  98                         && (c >= 0xf1 || c2 >= 0x90)
  99                         && (c < 0xf4 || (/* c == 0xf4 && */ c2 < 0x90)))
 100                       {
 101                         if (m == 2)
 102                           goto incomplete;
 103                         else
 104                           {
 105                             unsigned char c3 = (unsigned char) p[2];
 106
 107                             if ((c3 ^ 0x80) < 0x40)
 108                               {
 109                                 if (m == 3)
 110                                   goto incomplete;
 111                                 else /* m >= 4 */
 112                                   {
 113                                     unsigned char c4 = (unsigned char) p[3];
 114
 115                                     if ((c4 ^ 0x80) < 0x40)
 116                                       {
 117                                         unsigned int wc =
 118                                           (((unsigned int) (c & 0x07) << 18)
 119                                            | ((unsigned int) (c2 ^ 0x80) << 12)
 120                                            | ((unsigned int) (c3 ^ 0x80) << 6)
 121                                            | (unsigned int) (c4 ^ 0x80));
 122
 123                                         if (FITS_IN_CHAR_TYPE (wc))
 124                                           {
 125                                             if (pwc != NULL)
 126                                               *pwc = wc;
 127                                             res = 4;
 128                                             goto success;
 129                                           }
 130                                       }
 131                                   }
 132                               }
 133                           }
 134                       }
 135                   }
 136               }
 137           }
 138         goto invalid;