libs/wine/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., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, 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     0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 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     3,3,3,3,3,0,0,0,0,0,0,0,0,0,0,0  /* 0xf0-0xff */
  36 };
  37
  38 /* first byte mask depending on UTF-8 sequence length */
  39 static const unsigned char utf8_mask[4] = { 0x7f, 0x1f, 0x0f, 0x07 };
  40
  41 /* minimum Unicode value depending on UTF-8 sequence length */
  42 static const unsigned int utf8_minval[4] = { 0x0, 0x80, 0x800, 0x10000 };
  43
  44
  45 /* get the next char value taking surrogates into account */
  46 static inline unsigned int get_surrogate_value( const WCHAR *src, unsigned int srclen )
  47 {
  48     if (src[0] >= 0xd800 && src[0] <= 0xdfff)  /* surrogate pair */
  49     {
  50         if (src[0] > 0xdbff || /* invalid high surrogate */
  51             srclen <= 1 ||     /* missing low surrogate */
  52             src[1] < 0xdc00 || src[1] > 0xdfff) /* invalid low surrogate */
  53             return 0;
  54         return 0x10000 + ((src[0] & 0x3ff) << 10) + (src[1] & 0x3ff);
  55     }
  56     return src[0];
  57 }
  58
  59 /* query necessary dst length for src string */
  60 static inline int get_length_wcs_utf8( int flags, const WCHAR *src, unsigned int srclen )
  61 {
  62     int len;
  63     unsigned int val;
  64
  65     for (len = 0; srclen; srclen--, src++)
  66     {
  67         if (*src < 0x80)  /* 0x00-0x7f: 1 byte */
  68         {
  69             len++;
  70             continue;
  71         }
  72         if (*src < 0x800)  /* 0x80-0x7ff: 2 bytes */
  73         {
  74             len += 2;
  75             continue;
  76         }
  77         if (!(val = get_surrogate_value( src, srclen )))
  78         {
  79             if (flags & WC_ERR_INVALID_CHARS) return -2;
  80             continue;
  81         }
  82         if (val < 0x10000)  /* 0x800-0xffff: 3 bytes */
  83             len += 3;
  84         else   /* 0x10000-0x10ffff: 4 bytes */
  85         {
  86             len += 4;
  87             src++;
  88             srclen--;
  89         }
  90     }
  91     return len;
  92 }
  93
  94 /* wide char to UTF-8 string conversion */
  95 /* return -1 on dst buffer overflow, -2 on invalid input char */
  96 int wine_utf8_wcstombs( int flags, const WCHAR *src, int srclen, char *dst, int dstlen )
  97 {
  98     int len;
  99
 100     if (!dstlen) return get_length_wcs_utf8( flags, src, srclen );
 101
 102     for (len = dstlen; srclen; srclen--, src++)
 103     {
 104         WCHAR ch = *src;
 105         unsigned int val;
 106
 107         if (ch < 0x80)  /* 0x00-0x7f: 1 byte */
 108         {
 109             if (!len--) return -1;  /* overflow */
 110             *dst++ = ch;
 111             continue;
 112         }
 113
 114         if (ch < 0x800)  /* 0x80-0x7ff: 2 bytes */
 115         {
 116             if ((len -= 2) < 0) return -1;  /* overflow */
 117             dst[1] = 0x80 | (ch & 0x3f);
 118             ch >>= 6;
 119             dst[0] = 0xc0 | ch;
 120             dst += 2;
 121             continue;
 122         }
 123
 124         if (!(val = get_surrogate_value( src, srclen )))
 125         {
 126             if (flags & WC_ERR_INVALID_CHARS) return -2;
 127             continue;
 128         }
 129
 130         if (val < 0x10000)  /* 0x800-0xffff: 3 bytes */
 131         {
 132             if ((len -= 3) < 0) return -1;  /* overflow */
 133             dst[2] = 0x80 | (val & 0x3f);
 134             val >>= 6;
 135             dst[1] = 0x80 | (val & 0x3f);
 136             val >>= 6;
 137             dst[0] = 0xe0 | val;
 138             dst += 3;
 139         }
 140         else   /* 0x10000-0x10ffff: 4 bytes */
 141         {
 142             if ((len -= 4) < 0) return -1;  /* overflow */
 143             dst[3] = 0x80 | (val & 0x3f);
 144             val >>= 6;
 145             dst[2] = 0x80 | (val & 0x3f);
 146             val >>= 6;
 147             dst[1] = 0x80 | (val & 0x3f);
 148             val >>= 6;
 149             dst[0] = 0xf0 | val;
 150             dst += 4;
 151             src++;
 152             srclen--;
 153         }
 154     }
 155     return dstlen - len;
 156 }
 157
 158 /* query necessary dst length for src string */
 159 static inline int get_length_mbs_utf8( int flags, const char *src, int srclen )
 160 {
 161     int len, ret = 0;
 162     unsigned int res;
 163     const char *srcend = src + srclen;
 164
 165     while (src < srcend)
 166     {
 167         unsigned char ch = *src++;
 168         if (ch < 0x80)  /* special fast case for 7-bit ASCII */
 169         {
 170             ret++;
 171             continue;
 172         }
 173         len = utf8_length[ch-0x80];
 174         if (src + len > srcend) goto bad;
 175         res = ch & utf8_mask[len];
 176
 177         switch(len)
 178         {
 179         case 3:
 180             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 181             res = (res << 6) | ch;
 182             src++;
 183         case 2:
 184             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 185             res = (res << 6) | ch;
 186             src++;
 187         case 1:
 188             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 189             res = (res << 6) | ch;
 190             src++;
 191             if (res < utf8_minval[len]) goto bad;
 192             if (res > 0x10ffff) goto bad;
 193             if (res > 0xffff) ret++;
 194             ret++;
 195             continue;
 196         }
 197     bad:
 198         if (flags & MB_ERR_INVALID_CHARS) return -2;  /* bad char */
 199         /* otherwise ignore it */
 200     }
 201     return ret;
 202 }
 203
 204 /* UTF-8 to wide char string conversion */
 205 /* return -1 on dst buffer overflow, -2 on invalid input char */
 206 int wine_utf8_mbstowcs( int flags, const char *src, int srclen, WCHAR *dst, int dstlen )
 207 {
 208     int len;
 209     unsigned int res;
 210     const char *srcend = src + srclen;
 211     WCHAR *dstend = dst + dstlen;
 212
 213     if (!dstlen) return get_length_mbs_utf8( flags, src, srclen );
 214
 215     while ((dst < dstend) && (src < srcend))
 216     {
 217         unsigned char ch = *src++;
 218         if (ch < 0x80)  /* special fast case for 7-bit ASCII */
 219         {
 220             *dst++ = ch;
 221             continue;
 222         }
 223         len = utf8_length[ch-0x80];
 224         if (src + len > srcend) goto bad;
 225         res = ch & utf8_mask[len];
 226
 227         switch(len)
 228         {
 229         case 3:
 230             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 231             res = (res << 6) | ch;
 232             src++;
 233         case 2:
 234             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 235             res = (res << 6) | ch;
 236             src++;
 237         case 1:
 238             if ((ch = *src ^ 0x80) >= 0x40) goto bad;
 239             res = (res << 6) | ch;
 240             src++;
 241             if (res < utf8_minval[len]) goto bad;
 242             if (res > 0x10ffff) goto bad;
 243             if (res <= 0xffff) *dst++ = res;
 244             else /* we need surrogates */
 245             {
 246                 if (dst == dstend - 1) return -1;  /* overflow */
 247                 res -= 0x10000;
 248                 *dst++ = 0xd800 | (res >> 10);
 249                 *dst++ = 0xdc00 | (res & 0x3ff);
 250             }
 251             continue;
 252         }
 253     bad:
 254         if (flags & MB_ERR_INVALID_CHARS) return -2;  /* bad char */
 255         /* otherwise ignore it */
 256     }
 257     if (src < srcend) return -1;  /* overflow */
 258     return dstlen - (dstend - dst);
 259 }