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

Diff of /code/trunk/pcre_study.c

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

revision 77 by nigel, Sat Feb 24 21:40:45 2007 UTC revision 603 by ph10, Fri May 27 10:14:09 2011 UTC
# Line 6  Line 6 
6  and semantics are as close as possible to those of the Perl 5 language.  and semantics are as close as possible to those of the Perl 5 language.
7    
8                         Written by Philip Hazel                         Written by Philip Hazel
9             Copyright (c) 1997-2005 University of Cambridge             Copyright (c) 1997-2010 University of Cambridge
10    
11  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
12  Redistribution and use in source and binary forms, with or without  Redistribution and use in source and binary forms, with or without
# Line 42  POSSIBILITY OF SUCH DAMAGE. Line 42  POSSIBILITY OF SUCH DAMAGE.
42  supporting functions. */  supporting functions. */
43    
44    
45    #ifdef HAVE_CONFIG_H
46    #include "config.h"
47    #endif
48    
49  #include "pcre_internal.h"  #include "pcre_internal.h"
50    
51    #define SET_BIT(c) start_bits[c/8] |= (1 << (c&7))
52    
53    /* Returns from set_start_bits() */
54    
55    enum { SSB_FAIL, SSB_DONE, SSB_CONTINUE };
56    
57    
58    
59    /*************************************************
60    *   Find the minimum subject length for a group  *
61    *************************************************/
62    
63    /* Scan a parenthesized group and compute the minimum length of subject that
64    is needed to match it. This is a lower bound; it does not mean there is a
65    string of that length that matches. In UTF8 mode, the result is in characters
66    rather than bytes.
67    
68    Arguments:
69      code       pointer to start of group (the bracket)
70      startcode  pointer to start of the whole pattern
71      options    the compiling options
72    
73    Returns:   the minimum length
74               -1 if \C was encountered
75               -2 internal error (missing capturing bracket)
76               -3 internal error (opcode not listed)
77    */
78    
79    static int
80    find_minlength(const uschar *code, const uschar *startcode, int options)
81    {
82    int length = -1;
83    BOOL utf8 = (options & PCRE_UTF8) != 0;
84    BOOL had_recurse = FALSE;
85    register int branchlength = 0;
86    register uschar *cc = (uschar *)code + 1 + LINK_SIZE;
87    
88    if (*code == OP_CBRA || *code == OP_SCBRA) cc += 2;
89    
90    /* Scan along the opcodes for this branch. If we get to the end of the
91    branch, check the length against that of the other branches. */
92    
93    for (;;)
94      {
95      int d, min;
96      uschar *cs, *ce;
97      register int op = *cc;
98    
99      switch (op)
100        {
101        case OP_COND:
102        case OP_SCOND:
103    
104        /* If there is only one branch in a condition, the implied branch has zero
105        length, so we don't add anything. This covers the DEFINE "condition"
106        automatically. */
107    
108        cs = cc + GET(cc, 1);
109        if (*cs != OP_ALT)
110          {
111          cc = cs + 1 + LINK_SIZE;
112          break;
113          }
114    
115        /* Otherwise we can fall through and treat it the same as any other
116        subpattern. */
117    
118        case OP_CBRA:
119        case OP_SCBRA:
120        case OP_BRA:
121        case OP_SBRA:
122        case OP_ONCE:
123        d = find_minlength(cc, startcode, options);
124        if (d < 0) return d;
125        branchlength += d;
126        do cc += GET(cc, 1); while (*cc == OP_ALT);
127        cc += 1 + LINK_SIZE;
128        break;
129    
130        /* Reached end of a branch; if it's a ket it is the end of a nested
131        call. If it's ALT it is an alternation in a nested call. If it is
132        END it's the end of the outer call. All can be handled by the same code. */
133    
134        case OP_ALT:
135        case OP_KET:
136        case OP_KETRMAX:
137        case OP_KETRMIN:
138        case OP_END:
139        if (length < 0 || (!had_recurse && branchlength < length))
140          length = branchlength;
141        if (*cc != OP_ALT) return length;
142        cc += 1 + LINK_SIZE;
143        branchlength = 0;
144        had_recurse = FALSE;
145        break;
146    
147        /* Skip over assertive subpatterns */
148    
149        case OP_ASSERT:
150        case OP_ASSERT_NOT:
151        case OP_ASSERTBACK:
152        case OP_ASSERTBACK_NOT:
153        do cc += GET(cc, 1); while (*cc == OP_ALT);
154        /* Fall through */
155    
156        /* Skip over things that don't match chars */
157    
158        case OP_REVERSE:
159        case OP_CREF:
160        case OP_NCREF:
161        case OP_RREF:
162        case OP_NRREF:
163        case OP_DEF:
164        case OP_CALLOUT:
165        case OP_SOD:
166        case OP_SOM:
167        case OP_EOD:
168        case OP_EODN:
169        case OP_CIRC:
170        case OP_CIRCM:
171        case OP_DOLL:
172        case OP_DOLLM:
173        case OP_NOT_WORD_BOUNDARY:
174        case OP_WORD_BOUNDARY:
175        cc += _pcre_OP_lengths[*cc];
176        break;
177    
178        /* Skip over a subpattern that has a {0} or {0,x} quantifier */
179    
180        case OP_BRAZERO:
181        case OP_BRAMINZERO:
182        case OP_SKIPZERO:
183        cc += _pcre_OP_lengths[*cc];
184        do cc += GET(cc, 1); while (*cc == OP_ALT);
185        cc += 1 + LINK_SIZE;
186        break;
187    
188        /* Handle literal characters and + repetitions */
189    
190        case OP_CHAR:
191        case OP_CHARI:
192        case OP_NOT:
193        case OP_NOTI:
194        case OP_PLUS:
195        case OP_PLUSI:
196        case OP_MINPLUS:
197        case OP_MINPLUSI:
198        case OP_POSPLUS:
199        case OP_POSPLUSI:
200        case OP_NOTPLUS:
201        case OP_NOTPLUSI:
202        case OP_NOTMINPLUS:
203        case OP_NOTMINPLUSI:
204        case OP_NOTPOSPLUS:
205        case OP_NOTPOSPLUSI:
206        branchlength++;
207        cc += 2;
208    #ifdef SUPPORT_UTF8
209        if (utf8 && cc[-1] >= 0xc0) cc += _pcre_utf8_table4[cc[-1] & 0x3f];
210    #endif
211        break;
212    
213        case OP_TYPEPLUS:
214        case OP_TYPEMINPLUS:
215        case OP_TYPEPOSPLUS:
216        branchlength++;
217        cc += (cc[1] == OP_PROP || cc[1] == OP_NOTPROP)? 4 : 2;
218        break;
219    
220        /* Handle exact repetitions. The count is already in characters, but we
221        need to skip over a multibyte character in UTF8 mode.  */
222    
223        case OP_EXACT:
224        case OP_EXACTI:
225        case OP_NOTEXACT:
226        case OP_NOTEXACTI:
227        branchlength += GET2(cc,1);
228        cc += 4;
229    #ifdef SUPPORT_UTF8
230        if (utf8 && cc[-1] >= 0xc0) cc += _pcre_utf8_table4[cc[-1] & 0x3f];
231    #endif
232        break;
233    
234        case OP_TYPEEXACT:
235        branchlength += GET2(cc,1);
236        cc += (cc[3] == OP_PROP || cc[3] == OP_NOTPROP)? 6 : 4;
237        break;
238    
239        /* Handle single-char non-literal matchers */
240    
241        case OP_PROP:
242        case OP_NOTPROP:
243        cc += 2;
244        /* Fall through */
245    
246        case OP_NOT_DIGIT:
247        case OP_DIGIT:
248        case OP_NOT_WHITESPACE:
249        case OP_WHITESPACE:
250        case OP_NOT_WORDCHAR:
251        case OP_WORDCHAR:
252        case OP_ANY:
253        case OP_ALLANY:
254        case OP_EXTUNI:
255        case OP_HSPACE:
256        case OP_NOT_HSPACE:
257        case OP_VSPACE:
258        case OP_NOT_VSPACE:
259        branchlength++;
260        cc++;
261        break;
262    
263        /* "Any newline" might match two characters */
264    
265        case OP_ANYNL:
266        branchlength += 2;
267        cc++;
268        break;
269    
270        /* The single-byte matcher means we can't proceed in UTF-8 mode */
271    
272        case OP_ANYBYTE:
273    #ifdef SUPPORT_UTF8
274        if (utf8) return -1;
275    #endif
276        branchlength++;
277        cc++;
278        break;
279    
280        /* For repeated character types, we have to test for \p and \P, which have
281        an extra two bytes of parameters. */
282    
283        case OP_TYPESTAR:
284        case OP_TYPEMINSTAR:
285        case OP_TYPEQUERY:
286        case OP_TYPEMINQUERY:
287        case OP_TYPEPOSSTAR:
288        case OP_TYPEPOSQUERY:
289        if (cc[1] == OP_PROP || cc[1] == OP_NOTPROP) cc += 2;
290        cc += _pcre_OP_lengths[op];
291        break;
292    
293        case OP_TYPEUPTO:
294        case OP_TYPEMINUPTO:
295        case OP_TYPEPOSUPTO:
296        if (cc[3] == OP_PROP || cc[3] == OP_NOTPROP) cc += 2;
297        cc += _pcre_OP_lengths[op];
298        break;
299    
300        /* Check a class for variable quantification */
301    
302    #ifdef SUPPORT_UTF8
303        case OP_XCLASS:
304        cc += GET(cc, 1) - 33;
305        /* Fall through */
306    #endif
307    
308        case OP_CLASS:
309        case OP_NCLASS:
310        cc += 33;
311    
312        switch (*cc)
313          {
314          case OP_CRPLUS:
315          case OP_CRMINPLUS:
316          branchlength++;
317          /* Fall through */
318    
319          case OP_CRSTAR:
320          case OP_CRMINSTAR:
321          case OP_CRQUERY:
322          case OP_CRMINQUERY:
323          cc++;
324          break;
325    
326          case OP_CRRANGE:
327          case OP_CRMINRANGE:
328          branchlength += GET2(cc,1);
329          cc += 5;
330          break;
331    
332          default:
333          branchlength++;
334          break;
335          }
336        break;
337    
338        /* Backreferences and subroutine calls are treated in the same way: we find
339        the minimum length for the subpattern. A recursion, however, causes an
340        a flag to be set that causes the length of this branch to be ignored. The
341        logic is that a recursion can only make sense if there is another
342        alternation that stops the recursing. That will provide the minimum length
343        (when no recursion happens). A backreference within the group that it is
344        referencing behaves in the same way.
345    
346        If PCRE_JAVASCRIPT_COMPAT is set, a backreference to an unset bracket
347        matches an empty string (by default it causes a matching failure), so in
348        that case we must set the minimum length to zero. */
349    
350        case OP_REF:
351        case OP_REFI:
352        if ((options & PCRE_JAVASCRIPT_COMPAT) == 0)
353          {
354          ce = cs = (uschar *)_pcre_find_bracket(startcode, utf8, GET2(cc, 1));
355          if (cs == NULL) return -2;
356          do ce += GET(ce, 1); while (*ce == OP_ALT);
357          if (cc > cs && cc < ce)
358            {
359            d = 0;
360            had_recurse = TRUE;
361            }
362          else d = find_minlength(cs, startcode, options);
363          }
364        else d = 0;
365        cc += 3;
366    
367        /* Handle repeated back references */
368    
369        switch (*cc)
370          {
371          case OP_CRSTAR:
372          case OP_CRMINSTAR:
373          case OP_CRQUERY:
374          case OP_CRMINQUERY:
375          min = 0;
376          cc++;
377          break;
378    
379          case OP_CRRANGE:
380          case OP_CRMINRANGE:
381          min = GET2(cc, 1);
382          cc += 5;
383          break;
384    
385          default:
386          min = 1;
387          break;
388          }
389    
390        branchlength += min * d;
391        break;
392    
393        case OP_RECURSE:
394        cs = ce = (uschar *)startcode + GET(cc, 1);
395        if (cs == NULL) return -2;
396        do ce += GET(ce, 1); while (*ce == OP_ALT);
397        if (cc > cs && cc < ce)
398          had_recurse = TRUE;
399        else
400          branchlength += find_minlength(cs, startcode, options);
401        cc += 1 + LINK_SIZE;
402        break;
403    
404        /* Anything else does not or need not match a character. We can get the
405        item's length from the table, but for those that can match zero occurrences
406        of a character, we must take special action for UTF-8 characters. As it
407        happens, the "NOT" versions of these opcodes are used at present only for
408        ASCII characters, so they could be omitted from this list. However, in
409        future that may change, so we include them here so as not to leave a
410        gotcha for a future maintainer. */
411    
412        case OP_UPTO:
413        case OP_UPTOI:
414        case OP_NOTUPTO:
415        case OP_NOTUPTOI:
416        case OP_MINUPTO:
417        case OP_MINUPTOI:
418        case OP_NOTMINUPTO:
419        case OP_NOTMINUPTOI:
420        case OP_POSUPTO:
421        case OP_POSUPTOI:
422        case OP_NOTPOSUPTO:
423        case OP_NOTPOSUPTOI:
424    
425        case OP_STAR:
426        case OP_STARI:
427        case OP_NOTSTAR:
428        case OP_NOTSTARI:
429        case OP_MINSTAR:
430        case OP_MINSTARI:
431        case OP_NOTMINSTAR:
432        case OP_NOTMINSTARI:
433        case OP_POSSTAR:
434        case OP_POSSTARI:
435        case OP_NOTPOSSTAR:
436        case OP_NOTPOSSTARI:
437    
438        case OP_QUERY:
439        case OP_QUERYI:
440        case OP_NOTQUERY:
441        case OP_NOTQUERYI:
442        case OP_MINQUERY:
443        case OP_MINQUERYI:
444        case OP_NOTMINQUERY:
445        case OP_NOTMINQUERYI:
446        case OP_POSQUERY:
447        case OP_POSQUERYI:
448        case OP_NOTPOSQUERY:
449        case OP_NOTPOSQUERYI:
450    
451        cc += _pcre_OP_lengths[op];
452    #ifdef SUPPORT_UTF8
453        if (utf8 && cc[-1] >= 0xc0) cc += _pcre_utf8_table4[cc[-1] & 0x3f];
454    #endif
455        break;
456    
457        /* Skip these, but we need to add in the name length. */
458    
459        case OP_MARK:
460        case OP_PRUNE_ARG:
461        case OP_SKIP_ARG:
462        cc += _pcre_OP_lengths[op] + cc[1];
463        break;
464    
465        case OP_THEN_ARG:
466        cc += _pcre_OP_lengths[op] + cc[1+LINK_SIZE];
467        break;
468    
469        /* The remaining opcodes are just skipped over. */
470    
471        case OP_ACCEPT:
472        case OP_CLOSE:
473        case OP_COMMIT:
474        case OP_FAIL:
475        case OP_PRUNE:
476        case OP_SET_SOM:
477        case OP_SKIP:
478        case OP_THEN:
479        cc += _pcre_OP_lengths[op];
480        break;
481    
482        /* This should not occur: we list all opcodes explicitly so that when
483        new ones get added they are properly considered. */
484    
485        default:
486        return -3;
487        }
488      }
489    /* Control never gets here */
490    }
491    
492    
493    
494  /*************************************************  /*************************************************
495  *      Set a bit and maybe its alternate case    *  *      Set a bit and maybe its alternate case    *
496  *************************************************/  *************************************************/
497    
498  /* Given a character, set its bit in the table, and also the bit for the other  /* Given a character, set its first byte's bit in the table, and also the
499  version of a letter if we are caseless.  corresponding bit for the other version of a letter if we are caseless. In
500    UTF-8 mode, for characters greater than 127, we can only do the caseless thing
501    when Unicode property support is available.
502    
503  Arguments:  Arguments:
504    start_bits    points to the bit map    start_bits    points to the bit map
505    c             is the character    p             points to the character
506    caseless      the caseless flag    caseless      the caseless flag
507    cd            the block with char table pointers    cd            the block with char table pointers
508      utf8          TRUE for UTF-8 mode
509    
510  Returns:        nothing  Returns:        pointer after the character
511    */
512    
513    static const uschar *
514    set_table_bit(uschar *start_bits, const uschar *p, BOOL caseless,
515      compile_data *cd, BOOL utf8)
516    {
517    unsigned int c = *p;
518    
519    SET_BIT(c);
520    
521    #ifdef SUPPORT_UTF8
522    if (utf8 && c > 127)
523      {
524      GETCHARINC(c, p);
525    #ifdef SUPPORT_UCP
526      if (caseless)
527        {
528        uschar buff[8];
529        c = UCD_OTHERCASE(c);
530        (void)_pcre_ord2utf8(c, buff);
531        SET_BIT(buff[0]);
532        }
533    #endif
534      return p;
535      }
536    #endif
537    
538    /* Not UTF-8 mode, or character is less than 127. */
539    
540    if (caseless && (cd->ctypes[c] & ctype_letter) != 0) SET_BIT(cd->fcc[c]);
541    return p + 1;
542    }
543    
544    
545    
546    /*************************************************
547    *     Set bits for a positive character type     *
548    *************************************************/
549    
550    /* This function sets starting bits for a character type. In UTF-8 mode, we can
551    only do a direct setting for bytes less than 128, as otherwise there can be
552    confusion with bytes in the middle of UTF-8 characters. In a "traditional"
553    environment, the tables will only recognize ASCII characters anyway, but in at
554    least one Windows environment, some higher bytes bits were set in the tables.
555    So we deal with that case by considering the UTF-8 encoding.
556    
557    Arguments:
558      start_bits     the starting bitmap
559      cbit type      the type of character wanted
560      table_limit    32 for non-UTF-8; 16 for UTF-8
561      cd             the block with char table pointers
562    
563    Returns:         nothing
564  */  */
565    
566  static void  static void
567  set_bit(uschar *start_bits, unsigned int c, BOOL caseless, compile_data *cd)  set_type_bits(uschar *start_bits, int cbit_type, int table_limit,
568      compile_data *cd)
569  {  {
570  start_bits[c/8] |= (1 << (c&7));  register int c;
571  if (caseless && (cd->ctypes[c] & ctype_letter) != 0)  for (c = 0; c < table_limit; c++) start_bits[c] |= cd->cbits[c+cbit_type];
572    start_bits[cd->fcc[c]/8] |= (1 << (cd->fcc[c]&7));  if (table_limit == 32) return;
573    for (c = 128; c < 256; c++)
574      {
575      if ((cd->cbits[c/8] & (1 << (c&7))) != 0)
576        {
577        uschar buff[8];
578        (void)_pcre_ord2utf8(c, buff);
579        SET_BIT(buff[0]);
580        }
581      }
582    }
583    
584    
585    /*************************************************
586    *     Set bits for a negative character type     *
587    *************************************************/
588    
589    /* This function sets starting bits for a negative character type such as \D.
590    In UTF-8 mode, we can only do a direct setting for bytes less than 128, as
591    otherwise there can be confusion with bytes in the middle of UTF-8 characters.
592    Unlike in the positive case, where we can set appropriate starting bits for
593    specific high-valued UTF-8 characters, in this case we have to set the bits for
594    all high-valued characters. The lowest is 0xc2, but we overkill by starting at
595    0xc0 (192) for simplicity.
596    
597    Arguments:
598      start_bits     the starting bitmap
599      cbit type      the type of character wanted
600      table_limit    32 for non-UTF-8; 16 for UTF-8
601      cd             the block with char table pointers
602    
603    Returns:         nothing
604    */
605    
606    static void
607    set_nottype_bits(uschar *start_bits, int cbit_type, int table_limit,
608      compile_data *cd)
609    {
610    register int c;
611    for (c = 0; c < table_limit; c++) start_bits[c] |= ~cd->cbits[c+cbit_type];
612    if (table_limit != 32) for (c = 24; c < 32; c++) start_bits[c] = 0xff;
613  }  }
614    
615    
616    
617  /*************************************************  /*************************************************
618  *          Create bitmap of starting chars       *  *          Create bitmap of starting bytes       *
619  *************************************************/  *************************************************/
620    
621  /* This function scans a compiled unanchored expression and attempts to build a  /* This function scans a compiled unanchored expression recursively and
622  bitmap of the set of initial characters. If it can't, it returns FALSE. As time  attempts to build a bitmap of the set of possible starting bytes. As time goes
623  goes by, we may be able to get more clever at doing this.  by, we may be able to get more clever at doing this. The SSB_CONTINUE return is
624    useful for parenthesized groups in patterns such as (a*)b where the group
625    provides some optional starting bytes but scanning must continue at the outer
626    level to find at least one mandatory byte. At the outermost level, this
627    function fails unless the result is SSB_DONE.
628    
629  Arguments:  Arguments:
630    code         points to an expression    code         points to an expression
631    start_bits   points to a 32-byte table, initialized to 0    start_bits   points to a 32-byte table, initialized to 0
   caseless     the current state of the caseless flag  
632    utf8         TRUE if in UTF-8 mode    utf8         TRUE if in UTF-8 mode
633    cd           the block with char table pointers    cd           the block with char table pointers
634    
635  Returns:       TRUE if table built, FALSE otherwise  Returns:       SSB_FAIL     => Failed to find any starting bytes
636                   SSB_DONE     => Found mandatory starting bytes
637                   SSB_CONTINUE => Found optional starting bytes
638  */  */
639    
640  static BOOL  static int
641  set_start_bits(const uschar *code, uschar *start_bits, BOOL caseless,  set_start_bits(const uschar *code, uschar *start_bits, BOOL utf8,
642    BOOL utf8, compile_data *cd)    compile_data *cd)
643  {  {
644  register int c;  register int c;
645    int yield = SSB_DONE;
646    int table_limit = utf8? 16:32;
647    
648    #if 0
649    /* ========================================================================= */
650    /* The following comment and code was inserted in January 1999. In May 2006,
651    when it was observed to cause compiler warnings about unused values, I took it
652    out again. If anybody is still using OS/2, they will have to put it back
653    manually. */
654    
655  /* This next statement and the later reference to dummy are here in order to  /* This next statement and the later reference to dummy are here in order to
656  trick the optimizer of the IBM C compiler for OS/2 into generating correct  trick the optimizer of the IBM C compiler for OS/2 into generating correct
# Line 102  disable optimization (in this module it Line 659  disable optimization (in this module it
659  the pcre module can use all the optimization it can get). */  the pcre module can use all the optimization it can get). */
660    
661  volatile int dummy;  volatile int dummy;
662    /* ========================================================================= */
663    #endif
664    
665  do  do
666    {    {
667    const uschar *tcode = code + 1 + LINK_SIZE;    const uschar *tcode = code + (((int)*code == OP_CBRA)? 3:1) + LINK_SIZE;
668    BOOL try_next = TRUE;    BOOL try_next = TRUE;
669    
670    while (try_next)    while (try_next)    /* Loop for items in this branch */
671      {      {
672      /* If a branch starts with a bracket or a positive lookahead assertion,      int rc;
673      recurse to set bits from within them. That's all for this branch. */      switch(*tcode)
   
     if ((int)*tcode >= OP_BRA || *tcode == OP_ASSERT)  
674        {        {
675        if (!set_start_bits(tcode, start_bits, caseless, utf8, cd))        /* Fail if we reach something we don't understand */
         return FALSE;  
       try_next = FALSE;  
       }  
676    
     else switch(*tcode)  
       {  
677        default:        default:
678        return FALSE;        return SSB_FAIL;
679    
680        /* Skip over callout */        /* If we hit a bracket or a positive lookahead assertion, recurse to set
681          bits from within the subpattern. If it can't find anything, we have to
682          give up. If it finds some mandatory character(s), we are done for this
683          branch. Otherwise, carry on scanning after the subpattern. */
684    
685          case OP_BRA:
686          case OP_SBRA:
687          case OP_CBRA:
688          case OP_SCBRA:
689          case OP_ONCE:
690          case OP_ASSERT:
691          rc = set_start_bits(tcode, start_bits, utf8, cd);
692          if (rc == SSB_FAIL) return SSB_FAIL;
693          if (rc == SSB_DONE) try_next = FALSE; else
694            {
695            do tcode += GET(tcode, 1); while (*tcode == OP_ALT);
696            tcode += 1 + LINK_SIZE;
697            }
698          break;
699    
700        case OP_CALLOUT:        /* If we hit ALT or KET, it means we haven't found anything mandatory in
701        tcode += 2 + 2*LINK_SIZE;        this branch, though we might have found something optional. For ALT, we
702          continue with the next alternative, but we have to arrange that the final
703          result from subpattern is SSB_CONTINUE rather than SSB_DONE. For KET,
704          return SSB_CONTINUE: if this is the top level, that indicates failure,
705          but after a nested subpattern, it causes scanning to continue. */
706    
707          case OP_ALT:
708          yield = SSB_CONTINUE;
709          try_next = FALSE;
710        break;        break;
711    
712        /* Skip over extended extraction bracket number */        case OP_KET:
713          case OP_KETRMAX:
714          case OP_KETRMIN:
715          return SSB_CONTINUE;
716    
717        case OP_BRANUMBER:        /* Skip over callout */
718        tcode += 3;  
719          case OP_CALLOUT:
720          tcode += 2 + 2*LINK_SIZE;
721        break;        break;
722    
723        /* Skip over lookbehind and negative lookahead assertions */        /* Skip over lookbehind and negative lookahead assertions */
# Line 143  do Line 726  do
726        case OP_ASSERTBACK:        case OP_ASSERTBACK:
727        case OP_ASSERTBACK_NOT:        case OP_ASSERTBACK_NOT:
728        do tcode += GET(tcode, 1); while (*tcode == OP_ALT);        do tcode += GET(tcode, 1); while (*tcode == OP_ALT);
729        tcode += 1+LINK_SIZE;        tcode += 1 + LINK_SIZE;
       break;  
   
       /* Skip over an option setting, changing the caseless flag */  
   
       case OP_OPT:  
       caseless = (tcode[1] & PCRE_CASELESS) != 0;  
       tcode += 2;  
730        break;        break;
731    
732        /* BRAZERO does the bracket, but carries on. */        /* BRAZERO does the bracket, but carries on. */
733    
734        case OP_BRAZERO:        case OP_BRAZERO:
735        case OP_BRAMINZERO:        case OP_BRAMINZERO:
736        if (!set_start_bits(++tcode, start_bits, caseless, utf8, cd))        if (set_start_bits(++tcode, start_bits, utf8, cd) == SSB_FAIL)
737          return FALSE;          return SSB_FAIL;
738    /* =========================================================================
739          See the comment at the head of this function concerning the next line,
740          which was an old fudge for the benefit of OS/2.
741        dummy = 1;        dummy = 1;
742      ========================================================================= */
743          do tcode += GET(tcode,1); while (*tcode == OP_ALT);
744          tcode += 1 + LINK_SIZE;
745          break;
746    
747          /* SKIPZERO skips the bracket. */
748    
749          case OP_SKIPZERO:
750          tcode++;
751        do tcode += GET(tcode,1); while (*tcode == OP_ALT);        do tcode += GET(tcode,1); while (*tcode == OP_ALT);
752        tcode += 1+LINK_SIZE;        tcode += 1 + LINK_SIZE;
753        break;        break;
754    
755        /* Single-char * or ? sets the bit and tries the next item */        /* Single-char * or ? sets the bit and tries the next item */
756    
757        case OP_STAR:        case OP_STAR:
758        case OP_MINSTAR:        case OP_MINSTAR:
759          case OP_POSSTAR:
760        case OP_QUERY:        case OP_QUERY:
761        case OP_MINQUERY:        case OP_MINQUERY:
762        set_bit(start_bits, tcode[1], caseless, cd);        case OP_POSQUERY:
763        tcode += 2;        tcode = set_table_bit(start_bits, tcode + 1, FALSE, cd, utf8);
764  #ifdef SUPPORT_UTF8        break;
765        if (utf8) while ((*tcode & 0xc0) == 0x80) tcode++;  
766  #endif        case OP_STARI:
767          case OP_MINSTARI:
768          case OP_POSSTARI:
769          case OP_QUERYI:
770          case OP_MINQUERYI:
771          case OP_POSQUERYI:
772          tcode = set_table_bit(start_bits, tcode + 1, TRUE, cd, utf8);
773        break;        break;
774    
775        /* Single-char upto sets the bit and tries the next */        /* Single-char upto sets the bit and tries the next */
776    
777        case OP_UPTO:        case OP_UPTO:
778        case OP_MINUPTO:        case OP_MINUPTO:
779        set_bit(start_bits, tcode[3], caseless, cd);        case OP_POSUPTO:
780        tcode += 4;        tcode = set_table_bit(start_bits, tcode + 3, FALSE, cd, utf8);
781  #ifdef SUPPORT_UTF8        break;
782        if (utf8) while ((*tcode & 0xc0) == 0x80) tcode++;  
783  #endif        case OP_UPTOI:
784          case OP_MINUPTOI:
785          case OP_POSUPTOI:
786          tcode = set_table_bit(start_bits, tcode + 3, TRUE, cd, utf8);
787        break;        break;
788    
789        /* At least one single char sets the bit and stops */        /* At least one single char sets the bit and stops */
790    
791        case OP_EXACT:       /* Fall through */        case OP_EXACT:
792        tcode += 2;        tcode += 2;
793          /* Fall through */
794        case OP_CHAR:        case OP_CHAR:
       case OP_CHARNC:  
795        case OP_PLUS:        case OP_PLUS:
796        case OP_MINPLUS:        case OP_MINPLUS:
797        set_bit(start_bits, tcode[1], caseless, cd);        case OP_POSPLUS:
798          (void)set_table_bit(start_bits, tcode + 1, FALSE, cd, utf8);
799        try_next = FALSE;        try_next = FALSE;
800        break;        break;
801    
802        /* Single character type sets the bits and stops */        case OP_EXACTI:
803          tcode += 2;
804          /* Fall through */
805          case OP_CHARI:
806          case OP_PLUSI:
807          case OP_MINPLUSI:
808          case OP_POSPLUSI:
809          (void)set_table_bit(start_bits, tcode + 1, TRUE, cd, utf8);
810          try_next = FALSE;
811          break;
812    
813          /* Special spacing and line-terminating items. These recognize specific
814          lists of characters. The difference between VSPACE and ANYNL is that the
815          latter can match the two-character CRLF sequence, but that is not
816          relevant for finding the first character, so their code here is
817          identical. */
818    
819          case OP_HSPACE:
820          SET_BIT(0x09);
821          SET_BIT(0x20);
822          if (utf8)
823            {
824            SET_BIT(0xC2);  /* For U+00A0 */
825            SET_BIT(0xE1);  /* For U+1680, U+180E */
826            SET_BIT(0xE2);  /* For U+2000 - U+200A, U+202F, U+205F */
827            SET_BIT(0xE3);  /* For U+3000 */
828            }
829          else SET_BIT(0xA0);
830          try_next = FALSE;
831          break;
832    
833          case OP_ANYNL:
834          case OP_VSPACE:
835          SET_BIT(0x0A);
836          SET_BIT(0x0B);
837          SET_BIT(0x0C);
838          SET_BIT(0x0D);
839          if (utf8)
840            {
841            SET_BIT(0xC2);  /* For U+0085 */
842            SET_BIT(0xE2);  /* For U+2028, U+2029 */
843            }
844          else SET_BIT(0x85);
845          try_next = FALSE;
846          break;
847    
848          /* Single character types set the bits and stop. Note that if PCRE_UCP
849          is set, we do not see these op codes because \d etc are converted to
850          properties. Therefore, these apply in the case when only characters less
851          than 256 are recognized to match the types. */
852    
853        case OP_NOT_DIGIT:        case OP_NOT_DIGIT:
854        for (c = 0; c < 32; c++)        set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
         start_bits[c] |= ~cd->cbits[c+cbit_digit];  
855        try_next = FALSE;        try_next = FALSE;
856        break;        break;
857    
858        case OP_DIGIT:        case OP_DIGIT:
859        for (c = 0; c < 32; c++)        set_type_bits(start_bits, cbit_digit, table_limit, cd);
         start_bits[c] |= cd->cbits[c+cbit_digit];  
860        try_next = FALSE;        try_next = FALSE;
861        break;        break;
862    
863          /* The cbit_space table has vertical tab as whitespace; we have to
864          ensure it is set as not whitespace. */
865    
866        case OP_NOT_WHITESPACE:        case OP_NOT_WHITESPACE:
867        for (c = 0; c < 32; c++)        set_nottype_bits(start_bits, cbit_space, table_limit, cd);
868          start_bits[c] |= ~cd->cbits[c+cbit_space];        start_bits[1] |= 0x08;
869        try_next = FALSE;        try_next = FALSE;
870        break;        break;
871    
872          /* The cbit_space table has vertical tab as whitespace; we have to
873          not set it from the table. */
874    
875        case OP_WHITESPACE:        case OP_WHITESPACE:
876        for (c = 0; c < 32; c++)        c = start_bits[1];    /* Save in case it was already set */
877          start_bits[c] |= cd->cbits[c+cbit_space];        set_type_bits(start_bits, cbit_space, table_limit, cd);
878          start_bits[1] = (start_bits[1] & ~0x08) | c;
879        try_next = FALSE;        try_next = FALSE;
880        break;        break;
881    
882        case OP_NOT_WORDCHAR:        case OP_NOT_WORDCHAR:
883        for (c = 0; c < 32; c++)        set_nottype_bits(start_bits, cbit_word, table_limit, cd);
         start_bits[c] |= ~cd->cbits[c+cbit_word];  
884        try_next = FALSE;        try_next = FALSE;
885        break;        break;
886    
887        case OP_WORDCHAR:        case OP_WORDCHAR:
888        for (c = 0; c < 32; c++)        set_type_bits(start_bits, cbit_word, table_limit, cd);
         start_bits[c] |= cd->cbits[c+cbit_word];  
889        try_next = FALSE;        try_next = FALSE;
890        break;        break;
891    
# Line 244  do Line 894  do
894    
895        case OP_TYPEPLUS:        case OP_TYPEPLUS:
896        case OP_TYPEMINPLUS:        case OP_TYPEMINPLUS:
897          case OP_TYPEPOSPLUS:
898        tcode++;        tcode++;
899        break;        break;
900    
# Line 256  do Line 907  do
907    
908        case OP_TYPEUPTO:        case OP_TYPEUPTO:
909        case OP_TYPEMINUPTO:        case OP_TYPEMINUPTO:
910          case OP_TYPEPOSUPTO:
911        tcode += 2;               /* Fall through */        tcode += 2;               /* Fall through */
912    
913        case OP_TYPESTAR:        case OP_TYPESTAR:
914        case OP_TYPEMINSTAR:        case OP_TYPEMINSTAR:
915          case OP_TYPEPOSSTAR:
916        case OP_TYPEQUERY:        case OP_TYPEQUERY:
917        case OP_TYPEMINQUERY:        case OP_TYPEMINQUERY:
918          case OP_TYPEPOSQUERY:
919        switch(tcode[1])        switch(tcode[1])
920          {          {
921            default:
922          case OP_ANY:          case OP_ANY:
923          return FALSE;          case OP_ALLANY:
924            return SSB_FAIL;
925    
926            case OP_HSPACE:
927            SET_BIT(0x09);
928            SET_BIT(0x20);
929            if (utf8)
930              {
931              SET_BIT(0xC2);  /* For U+00A0 */
932              SET_BIT(0xE1);  /* For U+1680, U+180E */
933              SET_BIT(0xE2);  /* For U+2000 - U+200A, U+202F, U+205F */
934              SET_BIT(0xE3);  /* For U+3000 */
935              }
936            else SET_BIT(0xA0);
937            break;
938    
939            case OP_ANYNL:
940            case OP_VSPACE:
941            SET_BIT(0x0A);
942            SET_BIT(0x0B);
943            SET_BIT(0x0C);
944            SET_BIT(0x0D);
945            if (utf8)
946              {
947              SET_BIT(0xC2);  /* For U+0085 */
948              SET_BIT(0xE2);  /* For U+2028, U+2029 */
949              }
950            else SET_BIT(0x85);
951            break;
952    
953          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
954          for (c = 0; c < 32; c++)          set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
           start_bits[c] |= ~cd->cbits[c+cbit_digit];  
955          break;          break;
956    
957          case OP_DIGIT:          case OP_DIGIT:
958          for (c = 0; c < 32; c++)          set_type_bits(start_bits, cbit_digit, table_limit, cd);
           start_bits[c] |= cd->cbits[c+cbit_digit];  
959          break;          break;
960    
961            /* The cbit_space table has vertical tab as whitespace; we have to
962            ensure it gets set as not whitespace. */
963    
964          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
965          for (c = 0; c < 32; c++)          set_nottype_bits(start_bits, cbit_space, table_limit, cd);
966            start_bits[c] |= ~cd->cbits[c+cbit_space];          start_bits[1] |= 0x08;
967          break;          break;
968    
969            /* The cbit_space table has vertical tab as whitespace; we have to
970            avoid setting it. */
971    
972          case OP_WHITESPACE:          case OP_WHITESPACE:
973          for (c = 0; c < 32; c++)          c = start_bits[1];    /* Save in case it was already set */
974            start_bits[c] |= cd->cbits[c+cbit_space];          set_type_bits(start_bits, cbit_space, table_limit, cd);
975            start_bits[1] = (start_bits[1] & ~0x08) | c;
976          break;          break;
977    
978          case OP_NOT_WORDCHAR:          case OP_NOT_WORDCHAR:
979          for (c = 0; c < 32; c++)          set_nottype_bits(start_bits, cbit_word, table_limit, cd);
           start_bits[c] |= ~cd->cbits[c+cbit_word];  
980          break;          break;
981    
982          case OP_WORDCHAR:          case OP_WORDCHAR:
983          for (c = 0; c < 32; c++)          set_type_bits(start_bits, cbit_word, table_limit, cd);
           start_bits[c] |= cd->cbits[c+cbit_word];  
984          break;          break;
985          }          }
986    
# Line 308  do Line 994  do
994        character with a value > 255. */        character with a value > 255. */
995    
996        case OP_NCLASS:        case OP_NCLASS:
997    #ifdef SUPPORT_UTF8
998        if (utf8)        if (utf8)
999          {          {
1000          start_bits[24] |= 0xf0;              /* Bits for 0xc4 - 0xc8 */          start_bits[24] |= 0xf0;              /* Bits for 0xc4 - 0xc8 */
1001          memset(start_bits+25, 0xff, 7);      /* Bits for 0xc9 - 0xff */          memset(start_bits+25, 0xff, 7);      /* Bits for 0xc9 - 0xff */
1002          }          }
1003    #endif
1004        /* Fall through */        /* Fall through */
1005    
1006        case OP_CLASS:        case OP_CLASS:
# Line 325  do Line 1013  do
1013          value is > 127. In fact, there are only two possible starting bytes for          value is > 127. In fact, there are only two possible starting bytes for
1014          characters in the range 128 - 255. */          characters in the range 128 - 255. */
1015    
1016    #ifdef SUPPORT_UTF8
1017          if (utf8)          if (utf8)
1018            {            {
1019            for (c = 0; c < 16; c++) start_bits[c] |= tcode[c];            for (c = 0; c < 16; c++) start_bits[c] |= tcode[c];
# Line 342  do Line 1031  do
1031          /* In non-UTF-8 mode, the two bit maps are completely compatible. */          /* In non-UTF-8 mode, the two bit maps are completely compatible. */
1032    
1033          else          else
1034    #endif
1035            {            {
1036            for (c = 0; c < 32; c++) start_bits[c] |= tcode[c];            for (c = 0; c < 32; c++) start_bits[c] |= tcode[c];
1037            }            }
# Line 377  do Line 1067  do
1067    code += GET(code, 1);   /* Advance to next branch */    code += GET(code, 1);   /* Advance to next branch */
1068    }    }
1069  while (*code == OP_ALT);  while (*code == OP_ALT);
1070  return TRUE;  return yield;
1071  }  }
1072    
1073    
# Line 397  Arguments: Line 1087  Arguments:
1087              set NULL unless error              set NULL unless error
1088    
1089  Returns:    pointer to a pcre_extra block, with study_data filled in and the  Returns:    pointer to a pcre_extra block, with study_data filled in and the
1090                appropriate flag set;                appropriate flags set;
1091              NULL on error or if no optimization possible              NULL on error or if no optimization possible
1092  */  */
1093    
1094  EXPORT pcre_extra *  PCRE_EXP_DEFN pcre_extra * PCRE_CALL_CONVENTION
1095  pcre_study(const pcre *external_re, int options, const char **errorptr)  pcre_study(const pcre *external_re, int options, const char **errorptr)
1096  {  {
1097    int min;
1098    BOOL bits_set = FALSE;
1099  uschar start_bits[32];  uschar start_bits[32];
1100  pcre_extra *extra;  pcre_extra *extra;
1101  pcre_study_data *study;  pcre_study_data *study;
1102  const uschar *tables;  const uschar *tables;
1103  const real_pcre *re = (const real_pcre *)external_re;  uschar *code;
 uschar *code = (uschar *)re + re->name_table_offset +  
   (re->name_count * re->name_entry_size);  
1104  compile_data compile_block;  compile_data compile_block;
1105    const real_pcre *re = (const real_pcre *)external_re;
1106    
1107  *errorptr = NULL;  *errorptr = NULL;
1108    
# Line 427  if ((options & ~PUBLIC_STUDY_OPTIONS) != Line 1118  if ((options & ~PUBLIC_STUDY_OPTIONS) !=
1118    return NULL;    return NULL;
1119    }    }
1120    
1121    code = (uschar *)re + re->name_table_offset +
1122      (re->name_count * re->name_entry_size);
1123    
1124  /* For an anchored pattern, or an unanchored pattern that has a first char, or  /* For an anchored pattern, or an unanchored pattern that has a first char, or
1125  a multiline pattern that matches only at "line starts", no further processing  a multiline pattern that matches only at "line starts", there is no point in
1126  at present. */  seeking a list of starting bytes. */
1127    
1128  if ((re->options & (PCRE_ANCHORED|PCRE_FIRSTSET|PCRE_STARTLINE)) != 0)  if ((re->options & PCRE_ANCHORED) == 0 &&
1129    return NULL;      (re->flags & (PCRE_FIRSTSET|PCRE_STARTLINE)) == 0)
1130      {
1131      /* Set the character tables in the block that is passed around */
1132    
1133      tables = re->tables;
1134      if (tables == NULL)
1135        (void)pcre_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,
1136        (void *)(&tables));
1137    
1138      compile_block.lcc = tables + lcc_offset;
1139      compile_block.fcc = tables + fcc_offset;
1140      compile_block.cbits = tables + cbits_offset;
1141      compile_block.ctypes = tables + ctypes_offset;
1142    
1143      /* See if we can find a fixed set of initial characters for the pattern. */
1144    
1145      memset(start_bits, 0, 32 * sizeof(uschar));
1146      bits_set = set_start_bits(code, start_bits, (re->options & PCRE_UTF8) != 0,
1147        &compile_block) == SSB_DONE;
1148      }
1149    
1150    /* Find the minimum length of subject string. */
1151    
1152    switch(min = find_minlength(code, code, re->options))
1153      {
1154      case -2: *errorptr = "internal error: missing capturing bracket"; break;
1155      case -3: *errorptr = "internal error: opcode not recognized"; break;
1156      default: break;
1157      }
1158    
1159  /* Set the character tables in the block that is passed around */  /* Return NULL if no optimization is possible. */
1160    
1161  tables = re->tables;  if (!bits_set && min < 0) return NULL;
 if (tables == NULL)  
   (void)pcre_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,  
   (void *)(&tables));  
   
 compile_block.lcc = tables + lcc_offset;  
 compile_block.fcc = tables + fcc_offset;  
 compile_block.cbits = tables + cbits_offset;  
 compile_block.ctypes = tables + ctypes_offset;  
   
 /* See if we can find a fixed set of initial characters for the pattern. */  
   
 memset(start_bits, 0, 32 * sizeof(uschar));  
 if (!set_start_bits(code, start_bits, (re->options & PCRE_CASELESS) != 0,  
   (re->options & PCRE_UTF8) != 0, &compile_block)) return NULL;  
1162    
1163  /* Get a pcre_extra block and a pcre_study_data block. The study data is put in  /* Get a pcre_extra block and a pcre_study_data block. The study data is put in
1164  the latter, which is pointed to by the former, which may also get additional  the latter, which is pointed to by the former, which may also get additional
# Line 473  extra->flags = PCRE_EXTRA_STUDY_DATA; Line 1181  extra->flags = PCRE_EXTRA_STUDY_DATA;
1181  extra->study_data = study;  extra->study_data = study;
1182    
1183  study->size = sizeof(pcre_study_data);  study->size = sizeof(pcre_study_data);
1184  study->options = PCRE_STUDY_MAPPED;  study->flags = 0;
1185  memcpy(study->start_bits, start_bits, sizeof(start_bits));  
1186    if (bits_set)
1187      {
1188      study->flags |= PCRE_STUDY_MAPPED;
1189      memcpy(study->start_bits, start_bits, sizeof(start_bits));
1190      }
1191    
1192    if (min >= 0)
1193      {
1194      study->flags |= PCRE_STUDY_MINLEN;
1195      study->minlength = min;
1196      }
1197    
1198  return extra;  return extra;
1199  }  }
Legend:
Removed from v.77  
changed lines
  Added in v.603
webmaster@exim.org
 ViewVC Help
Powered by ViewVC 1.1.12