lib/mbswidth.c

   1 /* Determine the number of screen columns needed for a string.
   2    Copyright (C) 2000 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 2, or (at your option)
   7    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, write to the Free Software Foundation,
  16    Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  */
  17
  18 /* Written by Bruno Haible <haible@clisp.cons.org>.  */
  19
  20 #ifdef HAVE_CONFIG_H
  21 # include <config.h>
  22 #endif
  23
  24 /* Get MB_LEN_MAX.  */
  25 #if HAVE_LIMITS_H
  26 # include <limits.h>
  27 #endif
  28
  29 /* Get MB_CUR_MAX.  */
  30 #if HAVE_STDLIB_H
  31 # include <stdlib.h>
  32 #endif
  33
  34 #if HAVE_STRING_H
  35 # include <string.h>
  36 #endif
  37
  38 /* Get isprint().  */
  39 #include <ctype.h>
  40
  41 /* Get mbstate_t, mbrtowc(), mbsinit(), wcwidth().  */
  42 #if HAVE_WCHAR_H
  43 # include <wchar.h>
  44 #endif
  45
  46 /* Get iswprint().  */
  47 #if HAVE_WCTYPE_H
  48 # include <wctype.h>
  49 #endif
  50 #if !defined iswprint && !HAVE_ISWPRINT
  51 # define iswprint(wc) 1
  52 #endif
  53
  54 #ifndef HAVE_DECL_WCWIDTH
  55 "this configure-time declaration test was not run"
  56 #endif
  57 #if !HAVE_DECL_WCWIDTH
  58 int wcwidth ();
  59 #endif
  60
  61 #ifndef wcwidth
  62 # if !HAVE_WCWIDTH
  63 /* wcwidth doesn't exist, so assume all printable characters have
  64    width 1.  */
  65 #  define wcwidth(wc) ((wc) == 0 ? 0 : iswprint (wc) ? 1 : -1)
  66 # endif
  67 #endif
  68
  69 /* Get ISPRINT.  */
  70 #if defined (STDC_HEADERS) || (!defined (isascii) && !defined (HAVE_ISASCII))
  71 # define IN_CTYPE_DOMAIN(c) 1
  72 #else
  73 # define IN_CTYPE_DOMAIN(c) isascii(c)
  74 #endif
  75 /* Undefine to protect against the definition in wctype.h of solaris2.6.   */
  76 #undef ISPRINT
  77 #define ISPRINT(c) (IN_CTYPE_DOMAIN (c) && isprint (c))
  78
  79 #include "mbswidth.h"
  80
  81 /* Returns the number of columns needed to represent the multibyte
  82    character string pointed to by STRING.  If a non-printable character
  83    occurs, -1 is returned, unless MBSW_ACCEPT_UNPRINTABLE is specified.
  84    With flags = 0, this is the multibyte analogon of the wcswidth function.  */
  85 int
  86 mbswidth (const char *string, int flags)
  87 {
  88   return mbsnwidth (string, strlen (string), flags);
  89 }
  90
  91 /* Returns the number of columns needed to represent the multibyte
  92    character string pointed to by STRING of length NBYTES.  If a
  93    non-printable character occurs, -1 is returned, unless
  94    MBSW_ACCEPT_UNPRINTABLE is specified.  */
  95 int
  96 mbsnwidth (const char *string, size_t nbytes, int flags)
  97 {
  98   const char *p = string;
  99   const char *plimit = p + nbytes;
 100   int width;
 101
 102   width = 0;
 103 #if HAVE_MBRTOWC && (MB_LEN_MAX > 1)
 104   if (MB_CUR_MAX > 1)
 105     {
 106       while (p < plimit)
 107         switch (*p)
 108           {
 109             case ' ': case '!': case '"': case '#': case '%':
 110             case '&': case '\'': case '(': case ')': case '*':
 111             case '+': case ',': case '-': case '.': case '/':
 112             case '0': case '1': case '2': case '3': case '4':
 113             case '5': case '6': case '7': case '8': case '9':
 114             case ':': case ';': case '<': case '=': case '>':
 115             case '?':
 116             case 'A': case 'B': case 'C': case 'D': case 'E':
 117             case 'F': case 'G': case 'H': case 'I': case 'J':
 118             case 'K': case 'L': case 'M': case 'N': case 'O':
 119             case 'P': case 'Q': case 'R': case 'S': case 'T':
 120             case 'U': case 'V': case 'W': case 'X': case 'Y':
 121             case 'Z':
 122             case '[': case '\\': case ']': case '^': case '_':
 123             case 'a': case 'b': case 'c': case 'd': case 'e':
 124             case 'f': case 'g': case 'h': case 'i': case 'j':
 125             case 'k': case 'l': case 'm': case 'n': case 'o':
 126             case 'p': case 'q': case 'r': case 's': case 't':
 127             case 'u': case 'v': case 'w': case 'x': case 'y':
 128             case 'z': case '{': case '|': case '}': case '~':
 129               /* These characters are printable ASCII characters.  */
 130               p++;
 131               width++;
 132               break;
 133             default:
 134               /* If we have a multibyte sequence, scan it up to its end.  */
 135               {
 136                 mbstate_t mbstate;
 137                 memset (&mbstate, 0, sizeof mbstate);
 138                 do
 139                   {
 140                     wchar_t wc;
 141                     size_t bytes;
 142                     int w;
 143
 144                     bytes = mbrtowc (&wc, p, plimit - p, &mbstate);
 145
 146                     if (bytes == (size_t) -1)
 147                       /* An invalid multibyte sequence was encountered.  */
 148                       {
 149                         if (flags & MBSW_ACCEPT_INVALID)
 150                           {
 151                             p++;
 152                             width++;
 153                             break;
 154                           }
 155                         else
 156                           return -1;
 157                       }
 158
 159                     if (bytes == (size_t) -2)
 160                       /* An incomplete multibyte character at the end.  */
 161                       {
 162                         if (flags & MBSW_ACCEPT_INVALID)
 163                           {
 164                             p = plimit;
 165                             width++;
 166                             break;
 167                           }
 168                         else
 169                           return -1;
 170                       }
 171
 172                     if (bytes == 0)
 173                       /* A null wide character was encountered.  */
 174                       bytes = 1;
 175
 176                     w = wcwidth (wc);
 177                     if (w >= 0)
 178                       /* A printable multibyte character.  */
 179                       width += w;
 180                     else
 181                       /* An unprintable multibyte character.  */
 182                       if (flags & MBSW_ACCEPT_UNPRINTABLE)
 183                         width += 1;
 184                       else
 185                         return -1;
 186
 187                     p += bytes;
 188                   }
 189                 while (! mbsinit (&mbstate));
 190               }
 191               break;
 192           }
 193       return width;
 194     }
 195 #endif
 196
 197   while (p < plimit)
 198     {
 199       unsigned char c = (unsigned char) *p++;
 200
 201       if ((flags & MBSW_ACCEPT_UNPRINTABLE) || ISPRINT (c))
 202         width++;
 203       else
 204         return -1;
 205     }
 206   return width;
 207 }