Annotation of embedaddon/php/ext/pcre/pcrelib/pcre_valid_utf8.c, revision 1.1
1.1 ! misho 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) 1997-2009 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: #include "config.h"
! 46:
! 47: #include "pcre_internal.h"
! 48:
! 49:
! 50: /*************************************************
! 51: * Validate a UTF-8 string *
! 52: *************************************************/
! 53:
! 54: /* This function is called (optionally) at the start of compile or match, to
! 55: validate that a supposed UTF-8 string is actually valid. The early check means
! 56: that subsequent code can assume it is dealing with a valid string. The check
! 57: can be turned off for maximum performance, but the consequences of supplying
! 58: an invalid string are then undefined.
! 59:
! 60: Originally, this function checked according to RFC 2279, allowing for values in
! 61: the range 0 to 0x7fffffff, up to 6 bytes long, but ensuring that they were in
! 62: the canonical format. Once somebody had pointed out RFC 3629 to me (it
! 63: obsoletes 2279), additional restrictions were applied. The values are now
! 64: limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the
! 65: subrange 0xd000 to 0xdfff is excluded.
! 66:
! 67: Arguments:
! 68: string points to the string
! 69: length length of string, or -1 if the string is zero-terminated
! 70:
! 71: Returns: < 0 if the string is a valid UTF-8 string
! 72: >= 0 otherwise; the value is the offset of the bad byte
! 73:
! 74: Bad bytes can be:
! 75:
! 76: . An isolated byte whose most significant bits are 0x80, because this
! 77: can only correctly appear within a UTF-8 character;
! 78:
! 79: . A byte whose most significant bits are 0xc0, but whose other bits indicate
! 80: that there are more than 3 additional bytes (i.e. an RFC 2279 starting
! 81: byte, which is no longer valid under RFC 3629);
! 82:
! 83: .
! 84:
! 85: The returned offset may also be equal to the length of the string; this means
! 86: that one or more bytes is missing from the final UTF-8 character.
! 87: */
! 88:
! 89: int
! 90: _pcre_valid_utf8(USPTR string, int length)
! 91: {
! 92: #ifdef SUPPORT_UTF8
! 93: register USPTR p;
! 94:
! 95: if (length < 0)
! 96: {
! 97: for (p = string; *p != 0; p++);
! 98: length = p - string;
! 99: }
! 100:
! 101: for (p = string; length-- > 0; p++)
! 102: {
! 103: register int ab;
! 104: register int c = *p;
! 105: if (c < 128) continue;
! 106: if (c < 0xc0) return p - string;
! 107: ab = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */
! 108: if (ab > 3) return p - string; /* Too many for RFC 3629 */
! 109: if (length < ab) return p + 1 + length - string; /* Missing bytes */
! 110: length -= ab;
! 111:
! 112: /* Check top bits in the second byte */
! 113: if ((*(++p) & 0xc0) != 0x80) return p - string;
! 114:
! 115: /* Check for overlong sequences for each different length, and for the
! 116: excluded range 0xd000 to 0xdfff. */
! 117:
! 118: switch (ab)
! 119: {
! 120: /* Check for xx00 000x (overlong sequence) */
! 121:
! 122: case 1:
! 123: if ((c & 0x3e) == 0) return p - string;
! 124: continue; /* We know there aren't any more bytes to check */
! 125:
! 126: /* Check for 1110 0000, xx0x xxxx (overlong sequence) or
! 127: 1110 1101, 1010 xxxx (0xd000 - 0xdfff) */
! 128:
! 129: case 2:
! 130: if ((c == 0xe0 && (*p & 0x20) == 0) ||
! 131: (c == 0xed && *p >= 0xa0))
! 132: return p - string;
! 133: break;
! 134:
! 135: /* Check for 1111 0000, xx00 xxxx (overlong sequence) or
! 136: greater than 0x0010ffff (f4 8f bf bf) */
! 137:
! 138: case 3:
! 139: if ((c == 0xf0 && (*p & 0x30) == 0) ||
! 140: (c > 0xf4 ) ||
! 141: (c == 0xf4 && *p > 0x8f))
! 142: return p - string;
! 143: break;
! 144:
! 145: #if 0
! 146: /* These cases can no longer occur, as we restrict to a maximum of four
! 147: bytes nowadays. Leave the code here in case we ever want to add an option
! 148: for longer sequences. */
! 149:
! 150: /* Check for 1111 1000, xx00 0xxx */
! 151: case 4:
! 152: if (c == 0xf8 && (*p & 0x38) == 0) return p - string;
! 153: break;
! 154:
! 155: /* Check for leading 0xfe or 0xff, and then for 1111 1100, xx00 00xx */
! 156: case 5:
! 157: if (c == 0xfe || c == 0xff ||
! 158: (c == 0xfc && (*p & 0x3c) == 0)) return p - string;
! 159: break;
! 160: #endif
! 161:
! 162: }
! 163:
! 164: /* Check for valid bytes after the 2nd, if any; all must start 10 */
! 165: while (--ab > 0)
! 166: {
! 167: if ((*(++p) & 0xc0) != 0x80) return p - string;
! 168: }
! 169: }
! 170: #else
! 171: (void)(string); /* Keep picky compilers happy */
! 172: (void)(length);
! 173: #endif
! 174:
! 175: return -1;
! 176: }
! 177:
! 178: /* End of pcre_valid_utf8.c */
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>