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 /* 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 # include <wchar.h>
  38 #endif
  39
  40 /* Get iswprint(), iswcntrl().  */
  41 #if HAVE_WCTYPE_H
  42 # include <wctype.h>
  43 #endif
  44 #if !defined iswprint && !HAVE_ISWPRINT
  45 # define iswprint(wc) 1
  46 #endif
  47 #if !defined iswcntrl && !HAVE_ISWCNTRL
  48 # define iswcntrl(wc) 0
  49 #endif
  50
  51 #ifndef mbsinit
  52 # if !HAVE_MBSINIT
  53 #  define mbsinit(ps) 1
  54 # endif
  55 #endif
  56
  57 #ifndef HAVE_DECL_WCWIDTH
  58 "this configure-time declaration test was not run"
  59 #endif
  60 #if !HAVE_DECL_WCWIDTH
  61 int wcwidth ();
  62 #endif
  63
  64 #ifndef wcwidth
  65 # if !HAVE_WCWIDTH
  66 /* wcwidth doesn't exist, so assume all printable characters have
  67    width 1.  */
  68 #  define wcwidth(wc) ((wc) == 0 ? 0 : iswprint (wc) ? 1 : -1)
  69 # endif
  70 #endif
  71
  72 /* Get ISPRINT.  */
  73 #if defined (STDC_HEADERS) || (!defined (isascii) && !defined (HAVE_ISASCII))
  74 # define IN_CTYPE_DOMAIN(c) 1
  75 #else
  76 # define IN_CTYPE_DOMAIN(c) isascii(c)
  77 #endif
  78 /* Undefine to protect against the definition in wctype.h of solaris2.6.   */
  79 #undef ISPRINT
  80 #define ISPRINT(c) (IN_CTYPE_DOMAIN (c) && isprint (c))
  81 #undef ISCNTRL
  82 #define ISCNTRL(c) (IN_CTYPE_DOMAIN (c) && iscntrl (c))
  83
  84 /* Returns the number of columns needed to represent the multibyte
  85    character string pointed to by STRING.  If a non-printable character
  86    occurs, and MBSW_REJECT_UNPRINTABLE is specified, -1 is returned.
  87    With flags = MBSW_REJECT_INVALID | MBSW_REJECT_UNPRINTABLE, this is
  88    the multibyte analogon of the wcswidth function.  */
  89 int
  90 mbswidth (string, flags)
  91      const char *string;
  92      int flags;
  93 {
  94   return mbsnwidth (string, strlen (string), flags);
  95 }
  96
  97 /* Returns the number of columns needed to represent the multibyte
  98    character string pointed to by STRING of length NBYTES.  If a
  99    non-printable character occurs, and MBSW_REJECT_UNPRINTABLE is
 100    specified, -1 is returned.  */
 101 int
 102 mbsnwidth (string, nbytes, flags)
 103      const char *string;
 104      size_t nbytes;
 105      int flags;
 106 {
 107   const char *p = string;
 108   const char *plimit = p + nbytes;
 109   int width;
 110
 111   width = 0;
 112 #if HAVE_MBRTOWC
 113   if (MB_CUR_MAX > 1)
 114     {
 115       while (p < plimit)
 116         switch (*p)
 117           {
 118             case ' ': case '!': case '"': case '#': case '%':
 119             case '&': case '\'': case '(': case ')': case '*':
 120             case '+': case ',': case '-': case '.': case '/':
 121             case '0': case '1': case '2': case '3': case '4':
 122             case '5': case '6': case '7': case '8': case '9':
 123             case ':': case ';': case '<': case '=': case '>':
 124             case '?':
 125             case 'A': case 'B': case 'C': case 'D': case 'E':
 126             case 'F': case 'G': case 'H': case 'I': case 'J':
 127             case 'K': case 'L': case 'M': case 'N': case 'O':
 128             case 'P': case 'Q': case 'R': case 'S': case 'T':
 129             case 'U': case 'V': case 'W': case 'X': case 'Y':
 130             case 'Z':
 131             case '[': case '\\': case ']': case '^': case '_':
 132             case 'a': case 'b': case 'c': case 'd': case 'e':
 133             case 'f': case 'g': case 'h': case 'i': case 'j':
 134             case 'k': case 'l': case 'm': case 'n': case 'o':
 135             case 'p': case 'q': case 'r': case 's': case 't':
 136             case 'u': case 'v': case 'w': case 'x': case 'y':
 137             case 'z': case '{': case '|': case '}': case '~':
 138               /* These characters are printable ASCII characters.  */
 139               p++;
 140               width++;
 141               break;
 142             default:
 143               /* If we have a multibyte sequence, scan it up to its end.  */
 144               {
 145                 mbstate_t mbstate;
 146                 memset (&mbstate, 0, sizeof mbstate);
 147                 do
 148                   {
 149                     wchar_t wc;
 150                     size_t bytes;
 151                     int w;
 152
 153                     bytes = mbrtowc (&wc, p, plimit - p, &mbstate);
 154
 155                     if (bytes == (size_t) -1)
 156                       /* An invalid multibyte sequence was encountered.  */
 157                       {
 158                         if (!(flags & MBSW_REJECT_INVALID))
 159                           {
 160                             p++;
 161                             width++;
 162                             break;
 163                           }
 164                         else
 165                           return -1;
 166                       }
 167
 168                     if (bytes == (size_t) -2)
 169                       /* An incomplete multibyte character at the end.  */
 170                       {
 171                         if (!(flags & MBSW_REJECT_INVALID))
 172                           {
 173                             p = plimit;
 174                             width++;
 175                             break;
 176                           }
 177                         else
 178                           return -1;
 179                       }
 180
 181                     if (bytes == 0)
 182                       /* A null wide character was encountered.  */
 183                       bytes = 1;
 184
 185                     w = wcwidth (wc);
 186                     if (w >= 0)
 187                       /* A printable multibyte character.  */
 188                       width += w;
 189                     else
 190                       /* An unprintable multibyte character.  */
 191                       if (!(flags & MBSW_REJECT_UNPRINTABLE))
 192                         width += (iswcntrl (wc) ? 0 : 1);
 193                       else
 194                         return -1;
 195
 196                     p += bytes;
 197                   }
 198                 while (! mbsinit (&mbstate));
 199               }
 200               break;
 201           }
 202       return width;
 203     }
 204 #endif
 205
 206   while (p < plimit)
 207     {
 208       unsigned char c = (unsigned char) *p++;
 209
 210       if (ISPRINT (c))
 211         width++;
 212       else if (!(flags & MBSW_REJECT_UNPRINTABLE))
 213         width += (ISCNTRL (c) ? 0 : 1);
 214       else
 215         return -1;
 216     }
 217   return width;
 218 }