code/trunk/pcre_valid_utf8.c

/*************************************************
*      Perl-Compatible Regular Expressions       *
*************************************************/

/* PCRE is a library of functions to support regular expressions whose syntax
and semantics are as close as possible to those of the Perl 5 language.

                       Written by Philip Hazel
           Copyright (c) 1997-2009 University of Cambridge

-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:

    * Redistributions of source code must retain the above copyright notice,
      this list of conditions and the following disclaimer.

    * Redistributions in binary form must reproduce the above copyright
      notice, this list of conditions and the following disclaimer in the
      documentation and/or other materials provided with the distribution.

    * Neither the name of the University of Cambridge nor the names of its
      contributors may be used to endorse or promote products derived from
      this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
-----------------------------------------------------------------------------
*/


/* This module contains an internal function for validating UTF-8 character
strings. */


#ifdef HAVE_CONFIG_H
#include "config.h"
#endif

#include "pcre_internal.h"


/*************************************************
*         Validate a UTF-8 string                *
*************************************************/

/* This function is called (optionally) at the start of compile or match, to
validate that a supposed UTF-8 string is actually valid. The early check means
that subsequent code can assume it is dealing with a valid string. The check
can be turned off for maximum performance, but the consequences of supplying
an invalid string are then undefined.

Originally, this function checked according to RFC 2279, allowing for values in
the range 0 to 0x7fffffff, up to 6 bytes long, but ensuring that they were in
the canonical format. Once somebody had pointed out RFC 3629 to me (it
obsoletes 2279), additional restrictions were applied. The values are now
limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the
subrange 0xd000 to 0xdfff is excluded.

Arguments:
  string       points to the string
  length       length of string, or -1 if the string is zero-terminated

Returns:       < 0    if the string is a valid UTF-8 string
               >= 0   otherwise; the value is the offset of the bad byte
               
Bad bytes can be:

  . An isolated byte whose most significant bits are 0x80, because this
    can only correctly appear within a UTF-8 character;
      
  . A byte whose most significant bits are 0xc0, but whose other bits indicate
    that there are more than 3 additional bytes (i.e. an RFC 2279 starting 
    byte, which is no longer valid under RFC 3629);
    
  .
  
The returned offset may also be equal to the length of the string; this means 
that one or more bytes is missing from the final UTF-8 character.
*/

int
_pcre_valid_utf8(USPTR string, int length)
{
#ifdef SUPPORT_UTF8
register USPTR p;

if (length < 0)
  {
  for (p = string; *p != 0; p++);
  length = p - string;
  }

for (p = string; length-- > 0; p++)
  {
  register int ab;
  register int c = *p;
  if (c < 128) continue;
  if (c < 0xc0) return p - string;
  ab = _pcre_utf8_table4[c & 0x3f];     /* Number of additional bytes */
  if (ab > 3) return p - string;        /* Too many for RFC 3629 */
  if (length < ab) return p + 1 + length - string;   /* Missing bytes */
  length -= ab;

  /* Check top bits in the second byte */
  if ((*(++p) & 0xc0) != 0x80) return p - string;

  /* Check for overlong sequences for each different length, and for the
  excluded range 0xd000 to 0xdfff.  */

  switch (ab)
    {
    /* Check for xx00 000x (overlong sequence) */

    case 1:
    if ((c & 0x3e) == 0) return p - string;
    continue;   /* We know there aren't any more bytes to check */

    /* Check for 1110 0000, xx0x xxxx (overlong sequence) or
                 1110 1101, 1010 xxxx (0xd000 - 0xdfff) */

    case 2:
    if ((c == 0xe0 && (*p & 0x20) == 0) ||
        (c == 0xed && *p >= 0xa0))
      return p - string;
    break;

    /* Check for 1111 0000, xx00 xxxx (overlong sequence) or
       greater than 0x0010ffff (f4 8f bf bf) */

    case 3:
    if ((c == 0xf0 && (*p & 0x30) == 0) ||
        (c > 0xf4 ) ||
        (c == 0xf4 && *p > 0x8f))
      return p - string;
    break;

#if 0
    /* These cases can no longer occur, as we restrict to a maximum of four
    bytes nowadays. Leave the code here in case we ever want to add an option
    for longer sequences. */

    /* Check for 1111 1000, xx00 0xxx */
    case 4:
    if (c == 0xf8 && (*p & 0x38) == 0) return p - string;
    break;

    /* Check for leading 0xfe or 0xff, and then for 1111 1100, xx00 00xx */
    case 5:
    if (c == 0xfe || c == 0xff ||
       (c == 0xfc && (*p & 0x3c) == 0)) return p - string;
    break;
#endif

    }

  /* Check for valid bytes after the 2nd, if any; all must start 10 */
  while (--ab > 0)
    {
    if ((*(++p) & 0xc0) != 0x80) return p - string;
    }
  }
#else
(void)(string);  /* Keep picky compilers happy */
(void)(length);
#endif

return -1;
}

/* End of pcre_valid_utf8.c */
1	nigel	77	/*************************************************
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	ph10	409	Copyright (c) 1997-2009 University of Cambridge
10	nigel	77
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	ph10	200	#ifdef HAVE_CONFIG_H
46	ph10	236	#include "config.h"
47	ph10	200	#endif
48	ph10	199
49	nigel	77	#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			validate 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
60			an invalid string are then undefined.
61
62	ph10	211	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	ph10	293	obsoletes 2279), additional restrictions were applied. The values are now
66	ph10	211	limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the
67			subrange 0xd000 to 0xdfff is excluded.
68
69	nigel	77	Arguments:
70			string points to the string
71			length length of string, or -1 if the string is zero-terminated
72
73			Returns: < 0 if the string is a valid UTF-8 string
74			>= 0 otherwise; the value is the offset of the bad byte
75	ph10	569
76			Bad bytes can be:
77
78			. An isolated byte whose most significant bits are 0x80, because this
79			can only correctly appear within a UTF-8 character;
80
81			. A byte whose most significant bits are 0xc0, but whose other bits indicate
82			that there are more than 3 additional bytes (i.e. an RFC 2279 starting
83			byte, which is no longer valid under RFC 3629);
84
85			.
86
87			The returned offset may also be equal to the length of the string; this means
88			that one or more bytes is missing from the final UTF-8 character.
89	nigel	77	*/
90
91	nigel	87	int
92	ph10	409	_pcre_valid_utf8(USPTR string, int length)
93	nigel	77	{
94	ph10	110	#ifdef SUPPORT_UTF8
95	ph10	409	register USPTR p;
96	nigel	77
97			if (length < 0)
98			{
99			for (p = string; *p != 0; p++);
100			length = p - string;
101			}
102
103			for (p = string; length-- > 0; p++)
104			{
105			register int ab;
106			register int c = *p;
107			if (c < 128) continue;
108	nigel	93	if (c < 0xc0) return p - string;
109	ph10	211	ab = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */
110	ph10	569	if (ab > 3) return p - string; /* Too many for RFC 3629 */
111			if (length < ab) return p + 1 + length - string; /* Missing bytes */
112	nigel	77	length -= ab;
113
114			/* Check top bits in the second byte */
115			if ((*(++p) & 0xc0) != 0x80) return p - string;
116
117	ph10	211	/* Check for overlong sequences for each different length, and for the
118			excluded range 0xd000 to 0xdfff. */
119
120	nigel	77	switch (ab)
121			{
122	ph10	211	/* Check for xx00 000x (overlong sequence) */
123
124	nigel	77	case 1:
125			if ((c & 0x3e) == 0) return p - string;
126			continue; /* We know there aren't any more bytes to check */
127
128	ph10	211	/* Check for 1110 0000, xx0x xxxx (overlong sequence) or
129			1110 1101, 1010 xxxx (0xd000 - 0xdfff) */
130
131	nigel	77	case 2:
132	ph10	211	if ((c == 0xe0 && (*p & 0x20) == 0) \|\|
133			(c == 0xed && *p >= 0xa0))
134			return p - string;
135	nigel	77	break;
136
137	ph10	211	/* Check for 1111 0000, xx00 xxxx (overlong sequence) or
138			greater than 0x0010ffff (f4 8f bf bf) */
139
140	nigel	77	case 3:
141	ph10	211	if ((c == 0xf0 && (*p & 0x30) == 0) \|\|
142			(c > 0xf4 ) \|\|
143			(c == 0xf4 && *p > 0x8f))
144			return p - string;
145	nigel	77	break;
146
147	ph10	211	#if 0
148			/* These cases can no longer occur, as we restrict to a maximum of four
149			bytes nowadays. Leave the code here in case we ever want to add an option
150			for longer sequences. */
151
152	nigel	77	/* Check for 1111 1000, xx00 0xxx */
153			case 4:
154			if (c == 0xf8 && (*p & 0x38) == 0) return p - string;
155			break;
156
157			/* Check for leading 0xfe or 0xff, and then for 1111 1100, xx00 00xx */
158			case 5:
159			if (c == 0xfe \|\| c == 0xff \|\|
160			(c == 0xfc && (*p & 0x3c) == 0)) return p - string;
161			break;
162	ph10	211	#endif
163
164	nigel	77	}
165
166			/* Check for valid bytes after the 2nd, if any; all must start 10 */
167			while (--ab > 0)
168			{
169			if ((*(++p) & 0xc0) != 0x80) return p - string;
170			}
171			}
172	ph10	369	#else
173			(void)(string); /* Keep picky compilers happy */
174	ph10	371	(void)(length);
175	ph10	110	#endif
176	nigel	77
177			return -1;
178			}
179
180			/* End of pcre_valid_utf8.c */
Name	Value
svn:eol-style	native
svn:keywords	"Author Date Id Revision Url"