1 /* Determine the number of screen columns needed for a string. 2 Copyright (C) 2000-2006 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@clisp.cons.org>. */ 19 20 #include <config.h> 21 22 /* Specification. */ 23 #include "mbswidth.h" 24 25 /* Get MB_CUR_MAX. */ 26 #include <stdlib.h> 27 28 #include <string.h> 29 30 /* Get isprint(). */ 31 #include <ctype.h> 32 33 /* Get mbstate_t, mbrtowc(), mbsinit(). */ 34 #if HAVE_WCHAR_H 35 /* Tru64 with Desktop Toolkit C has a bug: <stdio.h> must be included before 36 <wchar.h>. 37 BSD/OS 4.1 has a bug: <stdio.h> and <time.h> must be included before 38 <wchar.h>. */ 39 # include <stdio.h> 40 # include <time.h> 41 # include <wchar.h> 42 #endif 43 44 /* Get wcwidth(). */ 45 #include "wcwidth.h" 46 47 /* Get iswcntrl(). */ 48 #if HAVE_WCTYPE_H 49 # include <wctype.h> 50 #endif 51 #if !defined iswcntrl && !HAVE_ISWCNTRL 52 # define iswcntrl(wc) (((wc) & ~0x1f) == 0 || (wc) == 0x7f) 53 #endif 54 55 #ifndef mbsinit 56 # if !HAVE_MBSINIT 57 # define mbsinit(ps) 1 58 # endif 59 #endif 60 61 /* Returns the number of columns needed to represent the multibyte 62 character string pointed to by STRING. If a non-printable character 63 occurs, and MBSW_REJECT_UNPRINTABLE is specified, -1 is returned. 64 With flags = MBSW_REJECT_INVALID | MBSW_REJECT_UNPRINTABLE, this is 65 the multibyte analogue of the wcswidth function. 66 If STRING is not of length < INT_MAX / 2, integer overflow can occur. */ 67 int 68 mbswidth (const char *string, int flags) 69 { 70 return mbsnwidth (string, strlen (string), flags); 71 } 72 73 /* Returns the number of columns needed to represent the multibyte 74 character string pointed to by STRING of length NBYTES. If a 75 non-printable character occurs, and MBSW_REJECT_UNPRINTABLE is 76 specified, -1 is returned. 77 If NBYTES is not < INT_MAX / 2, integer overflow can occur. */ 78 int 79 mbsnwidth (const char *string, size_t nbytes, int flags) 80 { 81 const char *p = string; 82 const char *plimit = p + nbytes; 83 int width; 84 85 width = 0; 86 #if HAVE_MBRTOWC 87 if (MB_CUR_MAX > 1) 88 { 89 while (p < plimit) 90 switch (*p) 91 { 92 case ' ': case '!': case '"': case '#': case '%': 93 case '&': case '\'': case '(': case ')': case '*': 94 case '+': case ',': case '-': case '.': case '/': 95 case '0': case '1': case '2': case '3': case '4': 96 case '5': case '6': case '7': case '8': case '9': 97 case ':': case ';': case '<': case '=': case '>': 98 case '?': 99 case 'A': case 'B': case 'C': case 'D': case 'E': 100 case 'F': case 'G': case 'H': case 'I': case 'J': 101 case 'K': case 'L': case 'M': case 'N': case 'O': 102 case 'P': case 'Q': case 'R': case 'S': case 'T': 103 case 'U': case 'V': case 'W': case 'X': case 'Y': 104 case 'Z': 105 case '[': case '\\': case ']': case '^': case '_': 106 case 'a': case 'b': case 'c': case 'd': case 'e': 107 case 'f': case 'g': case 'h': case 'i': case 'j': 108 case 'k': case 'l': case 'm': case 'n': case 'o': 109 case 'p': case 'q': case 'r': case 's': case 't': 110 case 'u': case 'v': case 'w': case 'x': case 'y': 111 case 'z': case '{': case '|': case '}': case '~': 112 /* These characters are printable ASCII characters. */ 113 p++; 114 width++; 115 break; 116 default: 117 /* If we have a multibyte sequence, scan it up to its end. */ 118 { 119 mbstate_t mbstate; 120 memset (&mbstate, 0, sizeof mbstate); 121 do 122 { 123 wchar_t wc; 124 size_t bytes; 125 int w; 126 127 bytes = mbrtowc (&wc, p, plimit - p, &mbstate); 128 129 if (bytes == (size_t) -1) 130 /* An invalid multibyte sequence was encountered. */ 131 { 132 if (!(flags & MBSW_REJECT_INVALID)) 133 { 134 p++; 135 width++; 136 break; 137 } 138 else 139 return -1; 140 } 141 142 if (bytes == (size_t) -2) 143 /* An incomplete multibyte character at the end. */ 144 { 145 if (!(flags & MBSW_REJECT_INVALID)) 146 { 147 p = plimit; 148 width++; 149 break; 150 } 151 else 152 return -1; 153 } 154 155 if (bytes == 0) 156 /* A null wide character was encountered. */ 157 bytes = 1; 158 159 w = wcwidth (wc); 160 if (w >= 0) 161 /* A printable multibyte character. */ 162 width += w; 163 else 164 /* An unprintable multibyte character. */ 165 if (!(flags & MBSW_REJECT_UNPRINTABLE)) 166 width += (iswcntrl (wc) ? 0 : 1); 167 else 168 return -1; 169 170 p += bytes; 171 } 172 while (! mbsinit (&mbstate)); 173 } 174 break; 175 } 176 return width; 177 } 178 #endif 179 180 while (p < plimit) 181 { 182 unsigned char c = (unsigned char) *p++; 183 184 if (isprint (c)) 185 width++; 186 else if (!(flags & MBSW_REJECT_UNPRINTABLE)) 187 width += (iscntrl (c) ? 0 : 1); 188 else 189 return -1; 190 } 191 return width; 192 } 193