lib/mbswidth.c

   1 /* Determine the number of screen columns needed for a string.
   2    Copyright (C) 2000-2001 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_CUR_MAX.  */
  25 #include <stdlib.h>
  26
  27 #include <string.h>
  28
  29 /* Get isprint().  */
  30 #include <ctype.h>
  31
  32 /* Get mbstate_t, mbrtowc(), mbsinit(), wcwidth().  */
  33 #if HAVE_WCHAR_H
  34 # include <wchar.h>
  35 #endif
  36
  37 /* Get iswprint(), iswcntrl().  */
  38 #if HAVE_WCTYPE_H
  39 # include <wctype.h>
  40 #endif
  41 #if !defined iswprint && !HAVE_ISWPRINT
  42 # define iswprint(wc) 1
  43 #endif
  44 #if !defined iswcntrl && !HAVE_ISWCNTRL
  45 # define iswcntrl(wc) 0
  46 #endif
  47
  48 #ifndef mbsinit
  49 # if !HAVE_MBSINIT
  50 #  define mbsinit(ps) 1
  51 # endif
  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 #undef ISCNTRL
  79 #define ISCNTRL(c) (IN_CTYPE_DOMAIN (c) && iscntrl (c))
  80
  81 #include "mbswidth.h"
  82
  83 /* Returns the number of columns needed to represent the multibyte
  84    character string pointed to by STRING.  If a non-printable character
  85    occurs, and MBSW_REJECT_UNPRINTABLE is specified, -1 is returned.
  86    With flags = MBSW_REJECT_INVALID | MBSW_REJECT_UNPRINTABLE, this is
  87    the multibyte analogon of the wcswidth function.  */
  88 int
  89 mbswidth (string, flags)
  90      const char *string;
  91      int flags;
  92 {
  93   return mbsnwidth (string, strlen (string), flags);
  94 }
  95
  96 /* Returns the number of columns needed to represent the multibyte
  97    character string pointed to by STRING of length NBYTES.  If a
  98    non-printable character occurs, and MBSW_REJECT_UNPRINTABLE is
  99    specified, -1 is returned.  */
 100 int
 101 mbsnwidth (string, nbytes, flags)
 102      const char *string;
 103      size_t nbytes;
 104      int flags;
 105 {
 106   const char *p = string;
 107   const char *plimit = p + nbytes;
 108   int width;
 109
 110   width = 0;
 111 #if HAVE_MBRTOWC
 112   if (MB_CUR_MAX > 1)
 113     {
 114       while (p < plimit)
 115         switch (*p)
 116           {
 117             case ' ': case '!': case '"': case '#': case '%':
 118             case '&': case '\'': case '(': case ')': case '*':
 119             case '+': case ',': case '-': case '.': case '/':
 120             case '0': case '1': case '2': case '3': case '4':
 121             case '5': case '6': case '7': case '8': case '9':
 122             case ':': case ';': case '<': case '=': case '>':
 123             case '?':
 124             case 'A': case 'B': case 'C': case 'D': case 'E':
 125             case 'F': case 'G': case 'H': case 'I': case 'J':
 126             case 'K': case 'L': case 'M': case 'N': case 'O':
 127             case 'P': case 'Q': case 'R': case 'S': case 'T':
 128             case 'U': case 'V': case 'W': case 'X': case 'Y':
 129             case 'Z':
 130             case '[': case '\\': case ']': case '^': case '_':
 131             case 'a': case 'b': case 'c': case 'd': case 'e':
 132             case 'f': case 'g': case 'h': case 'i': case 'j':
 133             case 'k': case 'l': case 'm': case 'n': case 'o':
 134             case 'p': case 'q': case 'r': case 's': case 't':
 135             case 'u': case 'v': case 'w': case 'x': case 'y':
 136             case 'z': case '{': case '|': case '}': case '~':
 137               /* These characters are printable ASCII characters.  */
 138               p++;
 139               width++;
 140               break;
 141             default:
 142               /* If we have a multibyte sequence, scan it up to its end.  */
 143               {
 144                 mbstate_t mbstate;
 145                 memset (&mbstate, 0, sizeof mbstate);
 146                 do
 147                   {
 148                     wchar_t wc;
 149                     size_t bytes;
 150                     int w;
 151
 152                     bytes = mbrtowc (&wc, p, plimit - p, &mbstate);
 153
 154                     if (bytes == (size_t) -1)
 155                       /* An invalid multibyte sequence was encountered.  */
 156                       {
 157                         if (!(flags & MBSW_REJECT_INVALID))
 158                           {
 159                             p++;
 160                             width++;
 161                             break;
 162                           }
 163                         else
 164                           return -1;
 165                       }
 166
 167                     if (bytes == (size_t) -2)
 168                       /* An incomplete multibyte character at the end.  */
 169                       {
 170                         if (!(flags & MBSW_REJECT_INVALID))
 171                           {
 172                             p = plimit;
 173                             width++;
 174                             break;
 175                           }
 176                         else
 177                           return -1;
 178                       }
 179
 180                     if (bytes == 0)
 181                       /* A null wide character was encountered.  */
 182                       bytes = 1;
 183
 184                     w = wcwidth (wc);
 185                     if (w >= 0)
 186                       /* A printable multibyte character.  */
 187                       width += w;
 188                     else
 189                       /* An unprintable multibyte character.  */
 190                       if (!(flags & MBSW_REJECT_UNPRINTABLE))
 191                         width += (iswcntrl (wc) ? 0 : 1);
 192                       else
 193                         return -1;
 194
 195                     p += bytes;
 196                   }
 197                 while (! mbsinit (&mbstate));
 198               }
 199               break;
 200           }
 201       return width;
 202     }
 203 #endif
 204
 205   while (p < plimit)
 206     {
 207       unsigned char c = (unsigned char) *p++;
 208
 209       if (ISPRINT (c))
 210         width++;
 211       else if (!(flags & MBSW_REJECT_UNPRINTABLE))
 212         width += (ISCNTRL (c) ? 0 : 1);
 213       else
 214         return -1;
 215     }
 216   return width;
 217 }