gnu/dist/gettext/gettext-tools/lib/mbswidth.c

   1 /* Determine the number of screen columns needed for a string.
   2    Copyright (C) 2000-2004 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 /* Specification.  */
  25 #include "mbswidth.h"
  26
  27 /* Get MB_CUR_MAX.  */
  28 #include <stdlib.h>
  29
  30 #include <string.h>
  31
  32 /* Get isprint().  */
  33 #include <ctype.h>
  34
  35 /* Get mbstate_t, mbrtowc(), mbsinit(), wcwidth().  */
  36 #if HAVE_WCHAR_H
  37 /* Tru64 with Desktop Toolkit C has a bug: <stdio.h> must be included before
  38    <wchar.h>.
  39    BSD/OS 4.1 has a bug: <stdio.h> and <time.h> must be included before
  40    <wchar.h>.  */
  41 # include <stdio.h>
  42 # include <time.h>
  43 # include <wchar.h>
  44 #endif
  45
  46 /* Get iswprint(), iswcntrl().  */
  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 #if !defined iswcntrl && !HAVE_ISWCNTRL
  54 # define iswcntrl(wc) 0
  55 #endif
  56
  57 #ifndef mbsinit
  58 # if !HAVE_MBSINIT
  59 #  define mbsinit(ps) 1
  60 # endif
  61 #endif
  62
  63 #ifndef HAVE_DECL_WCWIDTH
  64 "this configure-time declaration test was not run"
  65 #endif
  66 #if !HAVE_DECL_WCWIDTH
  67 int wcwidth ();
  68 #endif
  69
  70 #ifndef wcwidth
  71 # if !HAVE_WCWIDTH
  72 /* wcwidth doesn't exist, so assume all printable characters have
  73    width 1.  */
  74 #  define wcwidth(wc) ((wc) == 0 ? 0 : iswprint (wc) ? 1 : -1)
  75 # endif
  76 #endif
  77
  78 /* Get ISPRINT.  */
  79 #if defined (STDC_HEADERS) || (!defined (isascii) && !defined (HAVE_ISASCII))
  80 # define IN_CTYPE_DOMAIN(c) 1
  81 #else
  82 # define IN_CTYPE_DOMAIN(c) isascii(c)
  83 #endif
  84 /* Undefine to protect against the definition in wctype.h of Solaris 2.6.   */
  85 #undef ISPRINT
  86 #define ISPRINT(c) (IN_CTYPE_DOMAIN (c) && isprint (c))
  87 #undef ISCNTRL
  88 #define ISCNTRL(c) (IN_CTYPE_DOMAIN (c) && iscntrl (c))
  89
  90 /* Returns the number of columns needed to represent the multibyte
  91    character string pointed to by STRING.  If a non-printable character
  92    occurs, and MBSW_REJECT_UNPRINTABLE is specified, -1 is returned.
  93    With flags = MBSW_REJECT_INVALID | MBSW_REJECT_UNPRINTABLE, this is
  94    the multibyte analogue of the wcswidth function.  */
  95 int
  96 mbswidth (const char *string, int flags)
  97 {
  98   return mbsnwidth (string, strlen (string), flags);
  99 }
 100
 101 /* Returns the number of columns needed to represent the multibyte
 102    character string pointed to by STRING of length NBYTES.  If a
 103    non-printable character occurs, and MBSW_REJECT_UNPRINTABLE is
 104    specified, -1 is returned.  */
 105 int
 106 mbsnwidth (const char *string, size_t nbytes, int flags)
 107 {
 108   const char *p = string;
 109   const char *plimit = p + nbytes;
 110   int width;
 111
 112   width = 0;
 113 #if HAVE_MBRTOWC
 114   if (MB_CUR_MAX > 1)
 115     {
 116       while (p < plimit)
 117         switch (*p)
 118           {
 119             case ' ': case '!': case '"': case '#': case '%':
 120             case '&': case '\'': case '(': case ')': case '*':
 121             case '+': case ',': case '-': case '.': case '/':
 122             case '0': case '1': case '2': case '3': case '4':
 123             case '5': case '6': case '7': case '8': case '9':
 124             case ':': case ';': case '<': case '=': case '>':
 125             case '?':
 126             case 'A': case 'B': case 'C': case 'D': case 'E':
 127             case 'F': case 'G': case 'H': case 'I': case 'J':
 128             case 'K': case 'L': case 'M': case 'N': case 'O':
 129             case 'P': case 'Q': case 'R': case 'S': case 'T':
 130             case 'U': case 'V': case 'W': case 'X': case 'Y':
 131             case 'Z':
 132             case '[': case '\\': case ']': case '^': case '_':
 133             case 'a': case 'b': case 'c': case 'd': case 'e':
 134             case 'f': case 'g': case 'h': case 'i': case 'j':
 135             case 'k': case 'l': case 'm': case 'n': case 'o':
 136             case 'p': case 'q': case 'r': case 's': case 't':
 137             case 'u': case 'v': case 'w': case 'x': case 'y':
 138             case 'z': case '{': case '|': case '}': case '~':
 139               /* These characters are printable ASCII characters.  */
 140               p++;
 141               width++;
 142               break;
 143             default:
 144               /* If we have a multibyte sequence, scan it up to its end.  */
 145               {
 146                 mbstate_t mbstate;
 147                 memset (&mbstate, 0, sizeof mbstate);
 148                 do
 149                   {
 150                     wchar_t wc;
 151                     size_t bytes;
 152                     int w;
 153
 154                     bytes = mbrtowc (&wc, p, plimit - p, &mbstate);
 155
 156                     if (bytes == (size_t) -1)
 157                       /* An invalid multibyte sequence was encountered.  */
 158                       {
 159                         if (!(flags & MBSW_REJECT_INVALID))
 160                           {
 161                             p++;
 162                             width++;
 163                             break;
 164                           }
 165                         else
 166                           return -1;
 167                       }
 168
 169                     if (bytes == (size_t) -2)
 170                       /* An incomplete multibyte character at the end.  */
 171                       {
 172                         if (!(flags & MBSW_REJECT_INVALID))
 173                           {
 174                             p = plimit;
 175                             width++;
 176                             break;
 177                           }
 178                         else
 179                           return -1;
 180                       }
 181
 182                     if (bytes == 0)
 183                       /* A null wide character was encountered.  */
 184                       bytes = 1;
 185
 186                     w = wcwidth (wc);
 187                     if (w >= 0)
 188                       /* A printable multibyte character.  */
 189                       width += w;
 190                     else
 191                       /* An unprintable multibyte character.  */
 192                       if (!(flags & MBSW_REJECT_UNPRINTABLE))
 193                         width += (iswcntrl (wc) ? 0 : 1);
 194                       else
 195                         return -1;
 196
 197                     p += bytes;
 198                   }
 199                 while (! mbsinit (&mbstate));
 200               }
 201               break;
 202           }
 203       return width;
 204     }
 205 #endif
 206
 207   while (p < plimit)
 208     {
 209       unsigned char c = (unsigned char) *p++;
 210
 211       if (ISPRINT (c))
 212         width++;
 213       else if (!(flags & MBSW_REJECT_UNPRINTABLE))
 214         width += (ISCNTRL (c) ? 0 : 1);
 215       else
 216         return -1;
 217     }
 218   return width;
 219 }