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 "winnls.h"
  24 #include "wine/unicode.h"
  25
  26 /* number of following bytes in sequence based on first byte value (for bytes above 0x7f) */
  27 static const char utf8_length[128] =
  28 {
  29     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x80-0x8f */
  30     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x90-0x9f */
  31     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xa0-0xaf */
  32     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xb0-0xbf */
  33     1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xc0-0xcf */
  34     1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xd0-0xdf */
  35     2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* 0xe0-0xef */
  36     3,3,3,3,3,3,3,3,4,4,4,4,5,5,0,0  /* 0xf0-0xff */
  37 };
  38
  39 /* first byte mask depending on UTF-8 sequence length */
  40 static const unsigned char utf8_mask[6] = { 0x7f, 0x1f, 0x0f, 0x07, 0x03, 0x01 };
  41
  42 /* minimum Unicode value depending on UTF-8 sequence length */
  43 static const unsigned int utf8_minval[6] = { 0x0, 0x80, 0x800, 0x10000, 0x200000, 0x4000000 };
  44
  45
  46 /* query necessary dst length for src string */
  47 inline static int get_length_wcs_utf8( const WCHAR *src, unsigned int srclen )
  48 {
  49     int len;
  50     for (len = 0; srclen; srclen--, src++, len++)
  51     {
  52         if (*src >= 0x80)
  53         {
  54             len++;
  55             if (*src >= 0x800) len++;
  56         }
  57     }
  58     return len;
  59 }
  60
  61 /* wide char to UTF-8 string conversion */
  62 /* return -1 on dst buffer overflow */
  63 int utf8_wcstombs( const WCHAR *src, int srclen, char *dst, int dstlen )
  64 {
  65     int ret = srclen;
  66
  67     if (!dstlen) return get_length_wcs_utf8( src, srclen );
  68
  69     for (ret = srclen; srclen; srclen--, src++)
  70     {
  71         WCHAR ch = *src;
  72
  73         if (ch < 0x80)  /* 0x00-0x7f: 1 byte */
  74         {
  75             if (!dstlen--) return -1;  /* overflow */
  76             *dst++ = ch;
  77             continue;
  78         }
  79
  80         if (ch < 0x800)  /* 0x80-0x7ff: 2 bytes */
  81         {
  82             if ((dstlen -= 2) < 0) return -1;  /* overflow */
  83             dst[1] = 0x80 | (ch & 0x3f);
  84             ch >>= 6;
  85             dst[0] = 0xc0 | ch;
  86             dst += 2;
  87             continue;
  88         }
  89
  90         /* 0x800-0xffff: 3 bytes */
  91
  92         if ((dstlen -= 3) < 0) return -1;  /* overflow */
  93         dst[2] = 0x80 | (ch & 0x3f);
  94         ch >>= 6;
  95         dst[1] = 0x80 | (ch & 0x3f);
  96         ch >>= 6;
  97         dst[0] = 0xe0 | ch;
  98         dst += 3;
  99     }
 100     return ret;
 101 }
 102
 103 /* query necessary dst length for src string */
 104 inline static int get_length_mbs_utf8( const unsigned char *src, int srclen )
 105 {
 106     int ret;
 107     const unsigned char *srcend = src + srclen;
 108
 109     for (ret = 0; src < srcend; ret++)
 110     {
 111         unsigned char ch = *src++;
 112         if (ch < 0xc0) continue;
 113
 114         switch(utf8_length[ch-0x80])
 115         {
 116         case 5:
 117             if (src >= srcend) return ret;  /* ignore partial char */
 118             if ((ch = *src ^ 0x80) >= 0x40) continue;
 119             src++;
 120         case 4:
 121             if (src >= srcend) return ret;  /* ignore partial char */
 122             if ((ch = *src ^ 0x80) >= 0x40) continue;
 123             src++;
 124         case 3:
 125             if (src >= srcend) return ret;  /* ignore partial char */
 126             if ((ch = *src ^ 0x80) >= 0x40) continue;
 127             src++;
 128         case 2:
 129             if (src >= srcend) return ret;  /* ignore partial char */
 130             if ((ch = *src ^ 0x80) >= 0x40) continue;
 131             src++;
 132         case 1:
 133             if (src >= srcend) return ret;  /* ignore partial char */
 134             if ((ch = *src ^ 0x80) >= 0x40) continue;
 135             src++;
 136         }
 137     }
 138     return ret;
 139 }
 140
 141 /* UTF-8 to wide char string conversion */
 142 /* return -1 on dst buffer overflow, -2 on invalid input char */
 143 int utf8_mbstowcs( int flags, const char *src, int srclen, WCHAR *dst, int dstlen )
 144 {
 145     int len, count;
 146     unsigned int res;
 147     const char *srcend = src + srclen;
 148
 149     if (!dstlen) return get_length_mbs_utf8( src, srclen );
 150
 151     for (count = dstlen; count && (src < srcend); count--, dst++)
 152     {
 153         unsigned char ch = *src++;
 154         if (ch < 0x80)  /* special fast case for 7-bit ASCII */
 155         {
 156             *dst = ch;
 157             continue;
 158         }
 159         len = utf8_length[ch-0x80];
 160         res = ch & utf8_mask[len];
 161
 162         switch(len)
 163         {
 164         case 5:
 165             if (src >= srcend) goto done;  /* ignore partial char */
 166             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 167             res = (res << 6) | ch;
 168             src++;
 169         case 4:
 170             if (src >= srcend) goto done;  /* ignore partial char */
 171             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 172             res = (res << 6) | ch;
 173             src++;
 174         case 3:
 175             if (src >= srcend) goto done;  /* ignore partial char */
 176             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 177             res = (res << 6) | ch;
 178             src++;
 179         case 2:
 180             if (src >= srcend) goto done;  /* ignore partial char */
 181             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 182             res = (res << 6) | ch;
 183             src++;
 184         case 1:
 185             if (src >= srcend) goto done;  /* ignore partial char */
 186             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 187             res = (res << 6) | ch;
 188             src++;
 189             if (res < utf8_minval[len]) goto bad;
 190             if (res >= 0x10000) goto bad;  /* FIXME: maybe we should do surrogates here */
 191             *dst = res;
 192             continue;
 193         }
 194     bad:
 195         if (flags & MB_ERR_INVALID_CHARS) return -2;  /* bad char */
 196         *dst = (WCHAR)'?';
 197     }
 198     if (src < srcend) return -1;  /* overflow */
 199 done:
 200     return dstlen - count;
 201 }