src/utf8.c

   1 /* the Music Player Daemon (MPD)
   2  * Copyright (C) 2003-2007 by Warren Dukes (warren.dukes@gmail.com)
   3  * This project's homepage is: http://www.musicpd.org
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  13  * GNU General Public License for more details.
  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
  16  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  17  */
  18
  19 #include "utf8.h"
  20 #include "utils.h"
  21
  22 #include <stdio.h>
  23 #include <string.h>
  24 #include <stdlib.h>
  25
  26 static char *latin1ToUtf8(char c)
  27 {
  28         static unsigned char utf8[3];
  29         unsigned char uc = c;
  30
  31         memset(utf8, 0, 3);
  32
  33         if (uc < 128)
  34                 utf8[0] = uc;
  35         else if (uc < 192) {
  36                 utf8[0] = 194;
  37                 utf8[1] = uc;
  38         } else {
  39                 utf8[0] = 195;
  40                 utf8[1] = uc - 64;
  41         }
  42
  43         return (char *)utf8;
  44 }
  45
  46 char *latin1StrToUtf8Dup(char *latin1)
  47 {
  48         /* utf8 should have at most two char's per latin1 char */
  49         int len = strlen(latin1) * 2 + 1;
  50         char *ret = xmalloc(len);
  51         char *cp = ret;
  52         char *utf8;
  53
  54         memset(ret, 0, len);
  55
  56         len = 0;
  57
  58         while (*latin1) {
  59                 utf8 = latin1ToUtf8(*latin1);
  60                 while (*utf8) {
  61                         *(cp++) = *(utf8++);
  62                         len++;
  63                 }
  64                 latin1++;
  65         }
  66
  67         return xrealloc(ret, len + 1);
  68 }
  69
  70 static char utf8ToLatin1(char *inUtf8)
  71 {
  72         unsigned char c = 0;
  73         unsigned char *utf8 = (unsigned char *)inUtf8;
  74
  75         if (utf8[0] < 128)
  76                 return utf8[0];
  77         else if (utf8[0] == 195)
  78                 c += 64;
  79         else if (utf8[0] != 194)
  80                 return '?';
  81         return (char)(c + utf8[1]);
  82 }
  83
  84 static int validateUtf8Char(char *inUtf8Char)
  85 {
  86         unsigned char *utf8Char = (unsigned char *)inUtf8Char;
  87
  88         if (utf8Char[0] < 0x80)
  89                 return 1;
  90
  91         if (utf8Char[0] >= 0xC0 && utf8Char[0] <= 0xFD) {
  92                 int count = 1;
  93                 char t = 1 << 5;
  94                 int i;
  95                 while (count < 6 && (t & utf8Char[0])) {
  96                         t = (t >> 1);
  97                         count++;
  98                 }
  99                 if (count > 5)
 100                         return 0;
 101                 for (i = 1; i <= count; i++) {
 102                         if (utf8Char[i] < 0x80 || utf8Char[i] > 0xBF)
 103                                 return 0;
 104                 }
 105                 return count + 1;
 106         } else
 107                 return 0;
 108 }
 109
 110 int validUtf8String(char *string)
 111 {
 112         int ret;
 113
 114         while (*string) {
 115                 ret = validateUtf8Char(string);
 116                 if (0 == ret)
 117                         return 0;
 118                 string += ret;
 119         }
 120
 121         return 1;
 122 }
 123
 124 char *utf8StrToLatin1Dup(char *utf8)
 125 {
 126         /* utf8 should have at most two char's per latin1 char */
 127         int len = strlen(utf8) + 1;
 128         char *ret = xmalloc(len);
 129         char *cp = ret;
 130         int count;
 131
 132         memset(ret, 0, len);
 133
 134         len = 0;
 135
 136         while (*utf8) {
 137                 count = validateUtf8Char(utf8);
 138                 if (!count) {
 139                         free(ret);
 140                         return NULL;
 141                 }
 142                 *(cp++) = utf8ToLatin1(utf8);
 143                 utf8 += count;
 144                 len++;
 145         }
 146
 147         return xrealloc(ret, len + 1);
 148 }