lib/mbswidth.c

   1 /* Determine the number of screen columns needed for a string.
   2    Copyright (C) 2000-2024 Free Software Foundation, Inc.
   3
   4    This program is free software: you can redistribute it and/or modify
   5    it under the terms of the GNU General Public License as published by
   6    the Free Software Foundation, either version 3 of the License, or
   7    (at your option) any later version.
   8
   9    This program 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 General Public License for more details.
  13
  14    You should have received a copy of the GNU General Public License
  15    along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
  16
  17 /* Written by Bruno Haible <haible@clisp.cons.org>.  */
  18
  19 #include <config.h>
  20
  21 /* Specification.  */
  22 #include "mbswidth.h"
  23
  24 /* Get MB_CUR_MAX.  */
  25 #include <stdlib.h>
  26
  27 #include <string.h>
  28
  29 /* Get isprint().  */
  30 #include <ctype.h>
  31
  32 /* Get mbstate_t, mbsinit().  */
  33 #include <wchar.h>
  34
  35 /* Get char32_t, mbrtoc32(), c32iscntrl(), c32width().  */
  36 #include <uchar.h>
  37
  38 /* Get INT_MAX.  */
  39 #include <limits.h>
  40
  41 /* Returns the number of columns needed to represent the multibyte
  42    character string pointed to by STRING.  If a non-printable character
  43    occurs, and MBSW_REJECT_UNPRINTABLE is specified, -1 is returned.
  44    With flags = MBSW_REJECT_INVALID | MBSW_REJECT_UNPRINTABLE, this is
  45    the multibyte analogue of the wcswidth function.  */
  46 int
  47 mbswidth (const char *string, int flags)
  48 {
  49   return mbsnwidth (string, strlen (string), flags);
  50 }
  51
  52 /* Returns the number of columns needed to represent the multibyte
  53    character string pointed to by STRING of length NBYTES.  If a
  54    non-printable character occurs, and MBSW_REJECT_UNPRINTABLE is
  55    specified, -1 is returned.  */
  56 int
  57 mbsnwidth (const char *string, size_t nbytes, int flags)
  58 {
  59   const char *p = string;
  60   const char *plimit = p + nbytes;
  61   int width;
  62
  63   width = 0;
  64   if (MB_CUR_MAX > 1)
  65     {
  66       while (p < plimit)
  67         switch (*p)
  68           {
  69             case ' ': case '!': case '"': case '#': case '$': case '%':
  70             case '&': case '\'': case '(': case ')': case '*':
  71             case '+': case ',': case '-': case '.': case '/':
  72             case '0': case '1': case '2': case '3': case '4':
  73             case '5': case '6': case '7': case '8': case '9':
  74             case ':': case ';': case '<': case '=': case '>':
  75             case '?': case '@':
  76             case 'A': case 'B': case 'C': case 'D': case 'E':
  77             case 'F': case 'G': case 'H': case 'I': case 'J':
  78             case 'K': case 'L': case 'M': case 'N': case 'O':
  79             case 'P': case 'Q': case 'R': case 'S': case 'T':
  80             case 'U': case 'V': case 'W': case 'X': case 'Y':
  81             case 'Z':
  82             case '[': case '\\': case ']': case '^': case '_': case '`':
  83             case 'a': case 'b': case 'c': case 'd': case 'e':
  84             case 'f': case 'g': case 'h': case 'i': case 'j':
  85             case 'k': case 'l': case 'm': case 'n': case 'o':
  86             case 'p': case 'q': case 'r': case 's': case 't':
  87             case 'u': case 'v': case 'w': case 'x': case 'y':
  88             case 'z': case '{': case '|': case '}': case '~':
  89               /* These characters are printable ASCII characters.  */
  90               p++;
  91               width++;
  92               break;
  93             default:
  94               /* If we have a multibyte sequence, scan it up to its end.  */
  95               {
  96                 mbstate_t mbstate;
  97                 mbszero (&mbstate);
  98                 for (;;)
  99                   {
 100                     char32_t wc;
 101                     size_t bytes;
 102                     int w;
 103
 104                     bytes = mbrtoc32 (&wc, p, plimit - p, &mbstate);
 105
 106                     if (bytes == (size_t) -1)
 107                       /* An invalid multibyte sequence was encountered.  */
 108                       {
 109                         if (!(flags & MBSW_REJECT_INVALID))
 110                           {
 111                             p++;
 112                             width++;
 113                             break;
 114                           }
 115                         else
 116                           return -1;
 117                       }
 118
 119                     if (bytes == (size_t) -2)
 120                       /* An incomplete multibyte character at the end.  */
 121                       {
 122                         if (!(flags & MBSW_REJECT_INVALID))
 123                           {
 124                             p = plimit;
 125                             width++;
 126                             break;
 127                           }
 128                         else
 129                           return -1;
 130                       }
 131
 132                     if (bytes == 0)
 133                       /* A null wide character was encountered.  */
 134                       bytes = 1;
 135                     #if !GNULIB_MBRTOC32_REGULAR
 136                     else if (bytes == (size_t) -3)
 137                       bytes = 0;
 138                     #endif
 139
 140                     w = c32width (wc);
 141                     if (w >= 0)
 142                       /* A printable multibyte character.  */
 143                       {
 144                         if (w > INT_MAX - width)
 145                           goto overflow;
 146                         width += w;
 147                       }
 148                     else
 149                       /* An unprintable multibyte character.  */
 150                       if (!(flags & MBSW_REJECT_UNPRINTABLE))
 151                         {
 152                           if (!c32iscntrl (wc))
 153                             {
 154                               if (width == INT_MAX)
 155                                 goto overflow;
 156                               width++;
 157                             }
 158                         }
 159                       else
 160                         return -1;
 161
 162                     p += bytes;
 163                     #if !GNULIB_MBRTOC32_REGULAR
 164                     if (mbsinit (&mbstate))
 165                     #endif
 166                       break;
 167                   }
 168               }
 169               break;
 170           }
 171       return width;
 172     }
 173
 174   while (p < plimit)
 175     {
 176       unsigned char c = (unsigned char) *p++;
 177
 178       if (isprint (c))
 179         {
 180           if (width == INT_MAX)
 181             goto overflow;
 182           width++;
 183         }
 184       else if (!(flags & MBSW_REJECT_UNPRINTABLE))
 185         {
 186           if (!iscntrl (c))
 187             {
 188               if (width == INT_MAX)
 189                 goto overflow;
 190               width++;
 191             }
 192         }
 193       else
 194         return -1;
 195     }
 196   return width;
 197
 198  overflow:
 199   return INT_MAX;
 200 }