libs/unicode/utf8.c

   1 /*
   2  * UTF-8 support routines
   3  *
   4  * Copyright 2000 Alexandre Julliard
   5  *
   6  * This library is free software; you can redistribute it and/or
   7  * modify it under the terms of the GNU Lesser General Public
   8  * License as published by the Free Software Foundation; either
   9  * version 2.1 of the License, or (at your option) any later version.
  10  *
  11  * This library is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14  * Lesser General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU Lesser General Public
  17  * License along with this library; if not, write to the Free Software
  18  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  19  */
  20
  21 #include <string.h>
  22
  23 #include "wine/unicode.h"
  24
  25 /* number of following bytes in sequence based on first byte value (for bytes above 0x7f) */
  26 static const char utf8_length[128] =
  27 {
  28     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x80-0x8f */
  29     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x90-0x9f */
  30     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xa0-0xaf */
  31     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xb0-0xbf */
  32     1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xc0-0xcf */
  33     1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xd0-0xdf */
  34     2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* 0xe0-0xef */
  35     3,3,3,3,3,3,3,3,4,4,4,4,5,5,0,0  /* 0xf0-0xff */
  36 };
  37
  38 /* first byte mask depending on UTF-8 sequence length */
  39 static const unsigned char utf8_mask[6] = { 0x7f, 0x1f, 0x0f, 0x07, 0x03, 0x01 };
  40
  41 /* minimum Unicode value depending on UTF-8 sequence length */
  42 static const unsigned int utf8_minval[6] = { 0x0, 0x80, 0x800, 0x10000, 0x200000, 0x4000000 };
  43
  44
  45 /* query necessary dst length for src string */
  46 inline static int get_length_wcs_utf8( const WCHAR *src, unsigned int srclen )
  47 {
  48     int len;
  49     for (len = 0; srclen; srclen--, src++, len++)
  50     {
  51         if (*src >= 0x80)
  52         {
  53             len++;
  54             if (*src >= 0x800) len++;
  55         }
  56     }
  57     return len;
  58 }
  59
  60 /* wide char to UTF-8 string conversion */
  61 /* return -1 on dst buffer overflow */
  62 int wine_utf8_wcstombs( const WCHAR *src, int srclen, char *dst, int dstlen )
  63 {
  64     int len;
  65
  66     if (!dstlen) return get_length_wcs_utf8( src, srclen );
  67
  68     for (len = dstlen; srclen; srclen--, src++)
  69     {
  70         WCHAR ch = *src;
  71
  72         if (ch < 0x80)  /* 0x00-0x7f: 1 byte */
  73         {
  74             if (!len--) return -1;  /* overflow */
  75             *dst++ = ch;
  76             continue;
  77         }
  78
  79         if (ch < 0x800)  /* 0x80-0x7ff: 2 bytes */
  80         {
  81             if ((len -= 2) < 0) return -1;  /* overflow */
  82             dst[1] = 0x80 | (ch & 0x3f);
  83             ch >>= 6;
  84             dst[0] = 0xc0 | ch;
  85             dst += 2;
  86             continue;
  87         }
  88
  89         /* 0x800-0xffff: 3 bytes */
  90
  91         if ((len -= 3) < 0) return -1;  /* overflow */
  92         dst[2] = 0x80 | (ch & 0x3f);
  93         ch >>= 6;
  94         dst[1] = 0x80 | (ch & 0x3f);
  95         ch >>= 6;
  96         dst[0] = 0xe0 | ch;
  97         dst += 3;
  98     }
  99     return dstlen - len;
 100 }
 101
 102 /* query necessary dst length for src string */
 103 inline static int get_length_mbs_utf8( const unsigned char *src, int srclen )
 104 {
 105     int ret;
 106     const unsigned char *srcend = src + srclen;
 107
 108     for (ret = 0; src < srcend; ret++)
 109     {
 110         unsigned char ch = *src++;
 111         if (ch < 0xc0) continue;
 112
 113         switch(utf8_length[ch-0x80])
 114         {
 115         case 5:
 116             if (src >= srcend) return ret;  /* ignore partial char */
 117             if ((ch = *src ^ 0x80) >= 0x40) continue;
 118             src++;
 119         case 4:
 120             if (src >= srcend) return ret;  /* ignore partial char */
 121             if ((ch = *src ^ 0x80) >= 0x40) continue;
 122             src++;
 123         case 3:
 124             if (src >= srcend) return ret;  /* ignore partial char */
 125             if ((ch = *src ^ 0x80) >= 0x40) continue;
 126             src++;
 127         case 2:
 128             if (src >= srcend) return ret;  /* ignore partial char */
 129             if ((ch = *src ^ 0x80) >= 0x40) continue;
 130             src++;
 131         case 1:
 132             if (src >= srcend) return ret;  /* ignore partial char */
 133             if ((ch = *src ^ 0x80) >= 0x40) continue;
 134             src++;
 135         }
 136     }
 137     return ret;
 138 }
 139
 140 /* UTF-8 to wide char string conversion */
 141 /* return -1 on dst buffer overflow, -2 on invalid input char */
 142 int wine_utf8_mbstowcs( int flags, const char *src, int srclen, WCHAR *dst, int dstlen )
 143 {
 144     int len, count;
 145     unsigned int res;
 146     const char *srcend = src + srclen;
 147
 148     if (!dstlen) return get_length_mbs_utf8( src, srclen );
 149
 150     for (count = dstlen; count && (src < srcend); count--, dst++)
 151     {
 152         unsigned char ch = *src++;
 153         if (ch < 0x80)  /* special fast case for 7-bit ASCII */
 154         {
 155             *dst = ch;
 156             continue;
 157         }
 158         len = utf8_length[ch-0x80];
 159         res = ch & utf8_mask[len];
 160
 161         switch(len)
 162         {
 163         case 5:
 164             if (src >= srcend) goto done;  /* ignore partial char */
 165             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 166             res = (res << 6) | ch;
 167             src++;
 168         case 4:
 169             if (src >= srcend) goto done;  /* ignore partial char */
 170             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 171             res = (res << 6) | ch;
 172             src++;
 173         case 3:
 174             if (src >= srcend) goto done;  /* ignore partial char */
 175             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 176             res = (res << 6) | ch;
 177             src++;
 178         case 2:
 179             if (src >= srcend) goto done;  /* ignore partial char */
 180             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 181             res = (res << 6) | ch;
 182             src++;
 183         case 1:
 184             if (src >= srcend) goto done;  /* ignore partial char */
 185             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 186             res = (res << 6) | ch;
 187             src++;
 188             if (res < utf8_minval[len]) goto bad;
 189             if (res >= 0x10000) goto bad;  /* FIXME: maybe we should do surrogates here */
 190             *dst = res;
 191             continue;
 192         }
 193     bad:
 194         if (flags & MB_ERR_INVALID_CHARS) return -2;  /* bad char */
 195         *dst = (WCHAR)'?';
 196     }
 197     if (src < srcend) return -1;  /* overflow */
 198 done:
 199     return dstlen - count;
 200 }