pcre_valid_utf8.c

   1 /*************************************************
   2 *      Perl-Compatible Regular Expressions       *
   3 *************************************************/
   4
   5 /* PCRE is a library of functions to support regular expressions whose syntax
   6 and semantics are as close as possible to those of the Perl 5 language.
   7
   8                        Written by Philip Hazel
   9            Copyright (c) 1997-2013 University of Cambridge
  10
  11 -----------------------------------------------------------------------------
  12 Redistribution and use in source and binary forms, with or without
  13 modification, are permitted provided that the following conditions are met:
  14
  15     * Redistributions of source code must retain the above copyright notice,
  16       this list of conditions and the following disclaimer.
  17
  18     * Redistributions in binary form must reproduce the above copyright
  19       notice, this list of conditions and the following disclaimer in the
  20       documentation and/or other materials provided with the distribution.
  21
  22     * Neither the name of the University of Cambridge nor the names of its
  23       contributors may be used to endorse or promote products derived from
  24       this software without specific prior written permission.
  25
  26 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
  27 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  28 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  29 ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
  30 LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
  31 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
  32 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
  33 INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
  34 CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
  35 ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
  36 POSSIBILITY OF SUCH DAMAGE.
  37 -----------------------------------------------------------------------------
  38 */
  39
  40
  41 /* This module contains an internal function for validating UTF-8 character
  42 strings. */
  43
  44
  45 #ifdef HAVE_CONFIG_H
  46 #include "config.h"
  47 #endif
  48
  49 #include "pcre_internal.h"
  50
  51
  52 /*************************************************
  53 *         Validate a UTF-8 string                *
  54 *************************************************/
  55
  56 /* This function is called (optionally) at the start of compile or match, to
  57 check that a supposed UTF-8 string is actually valid. The early check means
  58 that subsequent code can assume it is dealing with a valid string. The check
  59 can be turned off for maximum performance, but the consequences of supplying an
  60 invalid string are then undefined.
  61
  62 Originally, this function checked according to RFC 2279, allowing for values in
  63 the range 0 to 0x7fffffff, up to 6 bytes long, but ensuring that they were in
  64 the canonical format. Once somebody had pointed out RFC 3629 to me (it
  65 obsoletes 2279), additional restrictions were applied. The values are now
  66 limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the
  67 subrange 0xd000 to 0xdfff is excluded. However, the format of 5-byte and 6-byte
  68 characters is still checked.
  69
  70 From release 8.13 more information about the details of the error are passed
  71 back in the returned value:
  72
  73 PCRE_UTF8_ERR0   No error
  74 PCRE_UTF8_ERR1   Missing 1 byte at the end of the string
  75 PCRE_UTF8_ERR2   Missing 2 bytes at the end of the string
  76 PCRE_UTF8_ERR3   Missing 3 bytes at the end of the string
  77 PCRE_UTF8_ERR4   Missing 4 bytes at the end of the string
  78 PCRE_UTF8_ERR5   Missing 5 bytes at the end of the string
  79 PCRE_UTF8_ERR6   2nd-byte's two top bits are not 0x80
  80 PCRE_UTF8_ERR7   3rd-byte's two top bits are not 0x80
  81 PCRE_UTF8_ERR8   4th-byte's two top bits are not 0x80
  82 PCRE_UTF8_ERR9   5th-byte's two top bits are not 0x80
  83 PCRE_UTF8_ERR10  6th-byte's two top bits are not 0x80
  84 PCRE_UTF8_ERR11  5-byte character is not permitted by RFC 3629
  85 PCRE_UTF8_ERR12  6-byte character is not permitted by RFC 3629
  86 PCRE_UTF8_ERR13  4-byte character with value > 0x10ffff is not permitted
  87 PCRE_UTF8_ERR14  3-byte character with value 0xd000-0xdfff is not permitted
  88 PCRE_UTF8_ERR15  Overlong 2-byte sequence
  89 PCRE_UTF8_ERR16  Overlong 3-byte sequence
  90 PCRE_UTF8_ERR17  Overlong 4-byte sequence
  91 PCRE_UTF8_ERR18  Overlong 5-byte sequence (won't ever occur)
  92 PCRE_UTF8_ERR19  Overlong 6-byte sequence (won't ever occur)
  93 PCRE_UTF8_ERR20  Isolated 0x80 byte (not within UTF-8 character)
  94 PCRE_UTF8_ERR21  Byte with the illegal value 0xfe or 0xff
  95 PCRE_UTF8_ERR22  Unused (was non-character)
  96
  97 Arguments:
  98   string       points to the string
  99   length       length of string, or -1 if the string is zero-terminated
 100   errp         pointer to an error position offset variable
 101
 102 Returns:       = 0    if the string is a valid UTF-8 string
 103                > 0    otherwise, setting the offset of the bad character
 104 */
 105
 106 int
 107 PRIV(valid_utf)(PCRE_PUCHAR string, int length, int *erroroffset)
 108 {
 109 #ifdef SUPPORT_UTF
 110 register PCRE_PUCHAR p;
 111
 112 if (length < 0)
 113   {
 114   for (p = string; *p != 0; p++);
 115   length = (int)(p - string);
 116   }
 117
 118 for (p = string; length-- > 0; p++)
 119   {
 120   register pcre_uchar ab, c, d;
 121
 122   c = *p;
 123   if (c < 128) continue;                /* ASCII character */
 124
 125   if (c < 0xc0)                         /* Isolated 10xx xxxx byte */
 126     {
 127     *erroroffset = (int)(p - string);
 128     return PCRE_UTF8_ERR20;
 129     }
 130
 131   if (c >= 0xfe)                        /* Invalid 0xfe or 0xff bytes */
 132     {
 133     *erroroffset = (int)(p - string);
 134     return PCRE_UTF8_ERR21;
 135     }
 136
 137   ab = PRIV(utf8_table4)[c & 0x3f];     /* Number of additional bytes */
 138   if (length < ab)
 139     {
 140     *erroroffset = (int)(p - string);          /* Missing bytes */
 141     return ab - length;                 /* Codes ERR1 to ERR5 */
 142     }
 143   length -= ab;                         /* Length remaining */
 144
 145   /* Check top bits in the second byte */
 146
 147   if (((d = *(++p)) & 0xc0) != 0x80)
 148     {
 149     *erroroffset = (int)(p - string) - 1;
 150     return PCRE_UTF8_ERR6;
 151     }
 152
 153   /* For each length, check that the remaining bytes start with the 0x80 bit
 154   set and not the 0x40 bit. Then check for an overlong sequence, and for the
 155   excluded range 0xd800 to 0xdfff. */
 156
 157   switch (ab)
 158     {
 159     /* 2-byte character. No further bytes to check for 0x80. Check first byte
 160     for for xx00 000x (overlong sequence). */
 161
 162     case 1: if ((c & 0x3e) == 0)
 163       {
 164       *erroroffset = (int)(p - string) - 1;
 165       return PCRE_UTF8_ERR15;
 166       }
 167     break;
 168
 169     /* 3-byte character. Check third byte for 0x80. Then check first 2 bytes
 170       for 1110 0000, xx0x xxxx (overlong sequence) or
 171           1110 1101, 1010 xxxx (0xd800 - 0xdfff) */
 172
 173     case 2:
 174     if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
 175       {
 176       *erroroffset = (int)(p - string) - 2;
 177       return PCRE_UTF8_ERR7;
 178       }
 179     if (c == 0xe0 && (d & 0x20) == 0)
 180       {
 181       *erroroffset = (int)(p - string) - 2;
 182       return PCRE_UTF8_ERR16;
 183       }
 184     if (c == 0xed && d >= 0xa0)
 185       {
 186       *erroroffset = (int)(p - string) - 2;
 187       return PCRE_UTF8_ERR14;
 188       }
 189     break;
 190
 191     /* 4-byte character. Check 3rd and 4th bytes for 0x80. Then check first 2
 192        bytes for for 1111 0000, xx00 xxxx (overlong sequence), then check for a
 193        character greater than 0x0010ffff (f4 8f bf bf) */
 194
 195     case 3:
 196     if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
 197       {
 198       *erroroffset = (int)(p - string) - 2;
 199       return PCRE_UTF8_ERR7;
 200       }
 201     if ((*(++p) & 0xc0) != 0x80)     /* Fourth byte */
 202       {
 203       *erroroffset = (int)(p - string) - 3;
 204       return PCRE_UTF8_ERR8;
 205       }
 206     if (c == 0xf0 && (d & 0x30) == 0)
 207       {
 208       *erroroffset = (int)(p - string) - 3;
 209       return PCRE_UTF8_ERR17;
 210       }
 211     if (c > 0xf4 || (c == 0xf4 && d > 0x8f))
 212       {
 213       *erroroffset = (int)(p - string) - 3;
 214       return PCRE_UTF8_ERR13;
 215       }
 216     break;
 217
 218     /* 5-byte and 6-byte characters are not allowed by RFC 3629, and will be
 219     rejected by the length test below. However, we do the appropriate tests
 220     here so that overlong sequences get diagnosed, and also in case there is
 221     ever an option for handling these larger code points. */
 222
 223     /* 5-byte character. Check 3rd, 4th, and 5th bytes for 0x80. Then check for
 224     1111 1000, xx00 0xxx */
 225
 226     case 4:
 227     if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
 228       {
 229       *erroroffset = (int)(p - string) - 2;
 230       return PCRE_UTF8_ERR7;
 231       }
 232     if ((*(++p) & 0xc0) != 0x80)     /* Fourth byte */
 233       {
 234       *erroroffset = (int)(p - string) - 3;
 235       return PCRE_UTF8_ERR8;
 236       }
 237     if ((*(++p) & 0xc0) != 0x80)     /* Fifth byte */
 238       {
 239       *erroroffset = (int)(p - string) - 4;
 240       return PCRE_UTF8_ERR9;
 241       }
 242     if (c == 0xf8 && (d & 0x38) == 0)
 243       {
 244       *erroroffset = (int)(p - string) - 4;
 245       return PCRE_UTF8_ERR18;
 246       }
 247     break;
 248
 249     /* 6-byte character. Check 3rd-6th bytes for 0x80. Then check for
 250     1111 1100, xx00 00xx. */
 251
 252     case 5:
 253     if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
 254       {
 255       *erroroffset = (int)(p - string) - 2;
 256       return PCRE_UTF8_ERR7;
 257       }
 258     if ((*(++p) & 0xc0) != 0x80)     /* Fourth byte */
 259       {
 260       *erroroffset = (int)(p - string) - 3;
 261       return PCRE_UTF8_ERR8;
 262       }
 263     if ((*(++p) & 0xc0) != 0x80)     /* Fifth byte */
 264       {
 265       *erroroffset = (int)(p - string) - 4;
 266       return PCRE_UTF8_ERR9;
 267       }
 268     if ((*(++p) & 0xc0) != 0x80)     /* Sixth byte */
 269       {
 270       *erroroffset = (int)(p - string) - 5;
 271       return PCRE_UTF8_ERR10;
 272       }
 273     if (c == 0xfc && (d & 0x3c) == 0)
 274       {
 275       *erroroffset = (int)(p - string) - 5;
 276       return PCRE_UTF8_ERR19;
 277       }
 278     break;
 279     }
 280
 281   /* Character is valid under RFC 2279, but 4-byte and 5-byte characters are
 282   excluded by RFC 3629. The pointer p is currently at the last byte of the
 283   character. */
 284
 285   if (ab > 3)
 286     {
 287     *erroroffset = (int)(p - string) - ab;
 288     return (ab == 4)? PCRE_UTF8_ERR11 : PCRE_UTF8_ERR12;
 289     }
 290   }
 291
 292 #else  /* Not SUPPORT_UTF */
 293 (void)(string);  /* Keep picky compilers happy */
 294 (void)(length);
 295 (void)(erroroffset);
 296 #endif
 297
 298 return PCRE_UTF8_ERR0;   /* This indicates success */
 299 }
 300
 301 /* End of pcre_valid_utf8.c */