/[pcre]/code/trunk/pcre_valid_utf8.c
ViewVC logotype

Diff of /code/trunk/pcre_valid_utf8.c

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 597 by ph10, Wed Nov 24 17:39:25 2010 UTC revision 598 by ph10, Sat May 7 15:37:31 2011 UTC
# Line 54  strings. */ Line 54  strings. */
54  *************************************************/  *************************************************/
55    
56  /* This function is called (optionally) at the start of compile or match, to  /* 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  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  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  can be turned off for maximum performance, but the consequences of supplying an
60  an invalid string are then undefined.  invalid string are then undefined.
61    
62  Originally, this function checked according to RFC 2279, allowing for values in  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  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  the canonical format. Once somebody had pointed out RFC 3629 to me (it
65  obsoletes 2279), additional restrictions were applied. The values are now  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  limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the
67  subrange 0xd000 to 0xdfff is excluded.  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 error code:
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    
96  Arguments:  Arguments:
97    string       points to the string    string       points to the string
98    length       length of string, or -1 if the string is zero-terminated    length       length of string, or -1 if the string is zero-terminated
99      errp         pointer to an error code variable
100    
101  Returns:       < 0    if the string is a valid UTF-8 string  Returns:       < 0    if the string is a valid UTF-8 string
102                 >= 0   otherwise; the value is the offset of the bad byte                 >= 0   otherwise; the value is the offset of the bad character
   
 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.  
103  */  */
104    
105  int  int
106  _pcre_valid_utf8(USPTR string, int length)  _pcre_valid_utf8(USPTR string, int length, int *errorcode)
107  {  {
108  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
109  register USPTR p;  register USPTR p;
# Line 100  if (length < 0) Line 114  if (length < 0)
114    length = p - string;    length = p - string;
115    }    }
116    
117    *errorcode = PCRE_UTF8_ERR0;
118    
119  for (p = string; length-- > 0; p++)  for (p = string; length-- > 0; p++)
120    {    {
121    register int ab;    register int ab, c, d;
122    register int c = *p;  
123    if (c < 128) continue;    c = *p;
124    if (c < 0xc0) return p - string;    if (c < 128) continue;                /* ASCII character */
125    
126      if (c < 0xc0)                         /* Isolated 10xx xxxx byte */
127        {
128        *errorcode = PCRE_UTF8_ERR20;
129        return p - string;
130        }
131    
132      if (c >= 0xfe)                        /* Invalid 0xfe or 0xff bytes */
133        {
134        *errorcode = PCRE_UTF8_ERR21;
135        return p - string;
136        }
137    
138    ab = _pcre_utf8_table4[c & 0x3f];     /* Number of additional bytes */    ab = _pcre_utf8_table4[c & 0x3f];     /* Number of additional bytes */
139    if (ab > 3) return p - string;        /* Too many for RFC 3629 */    if (length < ab)
140    if (length < ab) return p + 1 + length - string;   /* Missing bytes */      {
141    length -= ab;      *errorcode = ab - length;           /* Codes ERR1 to ERR5 */
142        return p - string;                  /* Missing bytes */
143        }
144      length -= ab;                         /* Length remaining */
145    
146    /* Check top bits in the second byte */    /* Check top bits in the second byte */
147    if ((*(++p) & 0xc0) != 0x80) return p - string;  
148      if (((d = *(++p)) & 0xc0) != 0x80)
149    /* Check for overlong sequences for each different length, and for the      {
150    excluded range 0xd000 to 0xdfff.  */      *errorcode = PCRE_UTF8_ERR6;
151        return p - string - 1;
152        }
153    
154      /* For each length, check that the remaining bytes start with the 0x80 bit
155      set and not the 0x40 bit. Then check for an overlong sequence, and for the
156      excluded range 0xd800 to 0xdfff. */
157    
158    switch (ab)    switch (ab)
159      {      {
160      /* Check for xx00 000x (overlong sequence) */      /* 2-byte character. No further bytes to check for 0x80. Check first byte
161        for for xx00 000x (overlong sequence). */
162      case 1:  
163      if ((c & 0x3e) == 0) return p - string;      case 1: if ((c & 0x3e) == 0)
164      continue;   /* We know there aren't any more bytes to check */        {
165          *errorcode = PCRE_UTF8_ERR15;
166      /* Check for 1110 0000, xx0x xxxx (overlong sequence) or        return p - string - 1;
167                   1110 1101, 1010 xxxx (0xd000 - 0xdfff) */        }
168        break;
169    
170        /* 3-byte character. Check third byte for 0x80. Then check first 2 bytes
171          for 1110 0000, xx0x xxxx (overlong sequence) or
172              1110 1101, 1010 xxxx (0xd800 - 0xdfff) */
173    
174      case 2:      case 2:
175      if ((c == 0xe0 && (*p & 0x20) == 0) ||      if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
176          (c == 0xed && *p >= 0xa0))        {
177        return p - string;        *errorcode = PCRE_UTF8_ERR7;
178          return p - string - 2;
179          }
180        if (c == 0xe0 && (d & 0x20) == 0)
181          {
182          *errorcode = PCRE_UTF8_ERR16;
183          return p - string - 2;
184          }
185        if (c == 0xed && d >= 0xa0)
186          {
187          *errorcode = PCRE_UTF8_ERR14;
188          return p - string - 2;
189          }
190      break;      break;
191    
192      /* Check for 1111 0000, xx00 xxxx (overlong sequence) or      /* 4-byte character. Check 3rd and 4th bytes for 0x80. Then check first 2
193         greater than 0x0010ffff (f4 8f bf bf) */         bytes for for 1111 0000, xx00 xxxx (overlong sequence), then check for a
194           character greater than 0x0010ffff (f4 8f bf bf) */
195    
196      case 3:      case 3:
197      if ((c == 0xf0 && (*p & 0x30) == 0) ||      if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
198          (c > 0xf4 ) ||        {
199          (c == 0xf4 && *p > 0x8f))        *errorcode = PCRE_UTF8_ERR7;
200        return p - string;        return p - string - 2;
201          }
202        if ((*(++p) & 0xc0) != 0x80)     /* Fourth byte */
203          {
204          *errorcode = PCRE_UTF8_ERR8;
205          return p - string - 3;
206          }
207        if (c == 0xf0 && (d & 0x30) == 0)
208          {
209          *errorcode = PCRE_UTF8_ERR17;
210          return p - string - 3;
211          }
212        if (c > 0xf4 || (c == 0xf4 && d > 0x8f))
213          {
214          *errorcode = PCRE_UTF8_ERR13;
215          return p - string - 3;
216          }
217      break;      break;
218    
219  #if 0      /* 5-byte and 6-byte characters are not allowed by RFC 3629, and will be
220      /* These cases can no longer occur, as we restrict to a maximum of four      rejected by the length test below. However, we do the appropriate tests
221      bytes nowadays. Leave the code here in case we ever want to add an option      here so that overlong sequences get diagnosed, and also in case there is
222      for longer sequences. */      ever an option for handling these larger code points. */
223    
224      /* Check for 1111 1000, xx00 0xxx */      /* 5-byte character. Check 3rd, 4th, and 5th bytes for 0x80. Then check for
225      case 4:      1111 1000, xx00 0xxx */
226      if (c == 0xf8 && (*p & 0x38) == 0) return p - string;  
227        case 4:
228        if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
229          {
230          *errorcode = PCRE_UTF8_ERR7;
231          return p - string - 2;
232          }
233        if ((*(++p) & 0xc0) != 0x80)     /* Fourth byte */
234          {
235          *errorcode = PCRE_UTF8_ERR8;
236          return p - string - 3;
237          }
238        if ((*(++p) & 0xc0) != 0x80)     /* Fifth byte */
239          {
240          *errorcode = PCRE_UTF8_ERR9;
241          return p - string - 4;
242          }
243        if (c == 0xf8 && (d & 0x38) == 0)
244          {
245          *errorcode = PCRE_UTF8_ERR18;
246          return p - string - 4;
247          }
248      break;      break;
249    
250      /* Check for leading 0xfe or 0xff, and then for 1111 1100, xx00 00xx */      /* 6-byte character. Check 3rd-6th bytes for 0x80. Then check for
251        1111 1100, xx00 00xx. */
252    
253      case 5:      case 5:
254      if (c == 0xfe || c == 0xff ||      if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
255         (c == 0xfc && (*p & 0x3c) == 0)) return p - string;        {
256          *errorcode = PCRE_UTF8_ERR7;
257          return p - string - 2;
258          }
259        if ((*(++p) & 0xc0) != 0x80)     /* Fourth byte */
260          {
261          *errorcode = PCRE_UTF8_ERR8;
262          return p - string - 3;
263          }
264        if ((*(++p) & 0xc0) != 0x80)     /* Fifth byte */
265          {
266          *errorcode = PCRE_UTF8_ERR9;
267          return p - string - 4;
268          }
269        if ((*(++p) & 0xc0) != 0x80)     /* Sixth byte */
270          {
271          *errorcode = PCRE_UTF8_ERR10;
272          return p - string - 5;
273          }
274        if (c == 0xfc && (d & 0x3c) == 0)
275          {
276          *errorcode = PCRE_UTF8_ERR19;
277          return p - string - 5;
278          }
279      break;      break;
 #endif  
   
280      }      }
281    
282      /* Character is valid under RFC 2279, but 4-byte and 5-byte characters are
283      excluded by RFC 3629. The pointer p is currently at the last byte of the
284      character. */
285    
286    /* Check for valid bytes after the 2nd, if any; all must start 10 */    if (ab > 3)
   while (--ab > 0)  
287      {      {
288      if ((*(++p) & 0xc0) != 0x80) return p - string;      *errorcode = (ab == 4)? PCRE_UTF8_ERR11 : PCRE_UTF8_ERR12;
289      }      return p - string - ab;
290        }
291    }    }
292  #else  
293    #else  /* SUPPORT_UTF8 */
294  (void)(string);  /* Keep picky compilers happy */  (void)(string);  /* Keep picky compilers happy */
295  (void)(length);  (void)(length);
296  #endif  #endif
297    
298  return -1;  return -1;   /* This indicates success */
299  }  }
300    
301  /* End of pcre_valid_utf8.c */  /* End of pcre_valid_utf8.c */

Legend:
Removed from v.597  
changed lines
  Added in v.598

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12