1 /* Determine the number of screen columns needed for a string. 2 Copyright (C) 2000-2005 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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ 17 18 /* Written by Bruno Haible <haible (at) 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 If STRING is not of length < INT_MAX / 2, integer overflow can occur. */ 96 int 97 mbswidth (const char *string, int flags) 98 { 99 return mbsnwidth (string, strlen (string), flags); 100 } 101 102 /* Returns the number of columns needed to represent the multibyte 103 character string pointed to by STRING of length NBYTES. If a 104 non-printable character occurs, and MBSW_REJECT_UNPRINTABLE is 105 specified, -1 is returned. 106 If NBYTES is not < INT_MAX / 2, integer overflow can occur. */ 107 int 108 mbsnwidth (const char *string, size_t nbytes, int flags) 109 { 110 const char *p = string; 111 const char *plimit = p + nbytes; 112 int width; 113 114 width = 0; 115 #if HAVE_MBRTOWC 116 if (MB_CUR_MAX > 1) 117 { 118 while (p < plimit) 119 switch (*p) 120 { 121 case ' ': case '!': case '"': case '#': case '%': 122 case '&': case '\'': case '(': case ')': case '*': 123 case '+': case ',': case '-': case '.': case '/': 124 case '0': case '1': case '2': case '3': case '4': 125 case '5': case '6': case '7': case '8': case '9': 126 case ':': case ';': case '<': case '=': case '>': 127 case '?': 128 case 'A': case 'B': case 'C': case 'D': case 'E': 129 case 'F': case 'G': case 'H': case 'I': case 'J': 130 case 'K': case 'L': case 'M': case 'N': case 'O': 131 case 'P': case 'Q': case 'R': case 'S': case 'T': 132 case 'U': case 'V': case 'W': case 'X': case 'Y': 133 case 'Z': 134 case '[': case '\\': case ']': case '^': case '_': 135 case 'a': case 'b': case 'c': case 'd': case 'e': 136 case 'f': case 'g': case 'h': case 'i': case 'j': 137 case 'k': case 'l': case 'm': case 'n': case 'o': 138 case 'p': case 'q': case 'r': case 's': case 't': 139 case 'u': case 'v': case 'w': case 'x': case 'y': 140 case 'z': case '{': case '|': case '}': case '~': 141 /* These characters are printable ASCII characters. */ 142 p++; 143 width++; 144 break; 145 default: 146 /* If we have a multibyte sequence, scan it up to its end. */ 147 { 148 mbstate_t mbstate; 149 memset (&mbstate, 0, sizeof mbstate); 150 do 151 { 152 wchar_t wc; 153 size_t bytes; 154 int w; 155 156 bytes = mbrtowc (&wc, p, plimit - p, &mbstate); 157 158 if (bytes == (size_t) -1) 159 /* An invalid multibyte sequence was encountered. */ 160 { 161 if (!(flags & MBSW_REJECT_INVALID)) 162 { 163 p++; 164 width++; 165 break; 166 } 167 else 168 return -1; 169 } 170 171 if (bytes == (size_t) -2) 172 /* An incomplete multibyte character at the end. */ 173 { 174 if (!(flags & MBSW_REJECT_INVALID)) 175 { 176 p = plimit; 177 width++; 178 break; 179 } 180 else 181 return -1; 182 } 183 184 if (bytes == 0) 185 /* A null wide character was encountered. */ 186 bytes = 1; 187 188 w = wcwidth (wc); 189 if (w >= 0) 190 /* A printable multibyte character. */ 191 width += w; 192 else 193 /* An unprintable multibyte character. */ 194 if (!(flags & MBSW_REJECT_UNPRINTABLE)) 195 width += (iswcntrl (wc) ? 0 : 1); 196 else 197 return -1; 198 199 p += bytes; 200 } 201 while (! mbsinit (&mbstate)); 202 } 203 break; 204 } 205 return width; 206 } 207 #endif 208 209 while (p < plimit) 210 { 211 unsigned char c = (unsigned char) *p++; 212 213 if (ISPRINT (c)) 214 width++; 215 else if (!(flags & MBSW_REJECT_UNPRINTABLE)) 216 width += (ISCNTRL (c) ? 0 : 1); 217 else 218 return -1; 219 } 220 return width; 221 } 222
