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

Diff of /code/trunk/pcre_exec.c

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

revision 778 by ph10, Thu Dec 1 17:38:47 2011 UTC revision 976 by ph10, Sat Jun 16 17:53:17 2012 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) 1997-2011 University of Cambridge             Copyright (c) 1997-2012 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 37  POSSIBILITY OF SUCH DAMAGE. Line 37  POSSIBILITY OF SUCH DAMAGE.
37  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
38  */  */
39    
   
40  /* This module contains pcre_exec(), the externally visible function that does  /* This module contains pcre_exec(), the externally visible function that does
41  pattern matching using an NFA algorithm, trying to mimic Perl as closely as  pattern matching using an NFA algorithm, trying to mimic Perl as closely as
42  possible. There are also some static supporting functions. */  possible. There are also some static supporting functions. */
# Line 113  Returns:     nothing Line 112  Returns:     nothing
112  */  */
113    
114  static void  static void
115  pchars(const uschar *p, int length, BOOL is_subject, match_data *md)  pchars(const pcre_uchar *p, int length, BOOL is_subject, match_data *md)
116  {  {
117  unsigned int c;  unsigned int c;
118  if (is_subject && length > md->end_subject - p) length = md->end_subject - p;  if (is_subject && length > md->end_subject - p) length = md->end_subject - p;
# Line 140  Arguments: Line 139  Arguments:
139    md          points to match data block    md          points to match data block
140    caseless    TRUE if caseless    caseless    TRUE if caseless
141    
142  Returns:      < 0 if not matched, otherwise the number of subject bytes matched  Returns:      >= 0 the number of subject bytes matched
143                  -1 no match
144                  -2 partial match; always given if at end subject
145  */  */
146    
147  static int  static int
148  match_ref(int offset, register USPTR eptr, int length, match_data *md,  match_ref(int offset, register PCRE_PUCHAR eptr, int length, match_data *md,
149    BOOL caseless)    BOOL caseless)
150  {  {
151  USPTR eptr_start = eptr;  PCRE_PUCHAR eptr_start = eptr;
152  register USPTR p = md->start_subject + md->offset_vector[offset];  register PCRE_PUCHAR p = md->start_subject + md->offset_vector[offset];
153    
154  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
155  if (eptr >= md->end_subject)  if (eptr >= md->end_subject)
# Line 163  pchars(p, length, FALSE, md); Line 164  pchars(p, length, FALSE, md);
164  printf("\n");  printf("\n");
165  #endif  #endif
166    
167  /* Always fail if reference not set (and not JavaScript compatible). */  /* Always fail if reference not set (and not JavaScript compatible - in that
168    case the length is passed as zero). */
169    
170  if (length < 0) return -1;  if (length < 0) return -1;
171    
# Line 173  ASCII characters. */ Line 175  ASCII characters. */
175    
176  if (caseless)  if (caseless)
177    {    {
178  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
179  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
180    if (md->utf8)    if (md->utf)
181      {      {
182      /* Match characters up to the end of the reference. NOTE: the number of      /* Match characters up to the end of the reference. NOTE: the number of
183      bytes matched may differ, because there are some characters whose upper and      bytes matched may differ, because there are some characters whose upper and
# Line 185  if (caseless) Line 187  if (caseless)
187      the latter. It is important, therefore, to check the length along the      the latter. It is important, therefore, to check the length along the
188      reference, not along the subject (earlier code did this wrong). */      reference, not along the subject (earlier code did this wrong). */
189    
190      USPTR endptr = p + length;      PCRE_PUCHAR endptr = p + length;
191      while (p < endptr)      while (p < endptr)
192        {        {
193        int c, d;        int c, d;
194        if (eptr >= md->end_subject) return -1;        if (eptr >= md->end_subject) return -2;   /* Partial match */
195        GETCHARINC(c, eptr);        GETCHARINC(c, eptr);
196        GETCHARINC(d, p);        GETCHARINC(d, p);
197        if (c != d && c != UCD_OTHERCASE(d)) return -1;        if (c != d && c != UCD_OTHERCASE(d)) return -1;
# Line 202  if (caseless) Line 204  if (caseless)
204    /* The same code works when not in UTF-8 mode and in UTF-8 mode when there    /* The same code works when not in UTF-8 mode and in UTF-8 mode when there
205    is no UCP support. */    is no UCP support. */
206      {      {
     if (eptr + length > md->end_subject) return -1;  
207      while (length-- > 0)      while (length-- > 0)
208        { if (md->lcc[*p++] != md->lcc[*eptr++]) return -1; }        {
209          if (eptr >= md->end_subject) return -2;   /* Partial match */
210          if (TABLE_GET(*p, md->lcc, *p) != TABLE_GET(*eptr, md->lcc, *eptr)) return -1;
211          p++;
212          eptr++;
213          }
214      }      }
215    }    }
216    
# Line 213  are in UTF-8 mode. */ Line 219  are in UTF-8 mode. */
219    
220  else  else
221    {    {
222    if (eptr + length > md->end_subject) return -1;    while (length-- > 0)
223    while (length-- > 0) if (*p++ != *eptr++) return -1;      {
224        if (eptr >= md->end_subject) return -2;   /* Partial match */
225        if (*p++ != *eptr++) return -1;
226        }
227    }    }
228    
229  return eptr - eptr_start;  return (int)(eptr - eptr_start);
230  }  }
231    
232    
# Line 307  argument of match(), which never changes Line 316  argument of match(), which never changes
316    
317  #define RMATCH(ra,rb,rc,rd,re,rw)\  #define RMATCH(ra,rb,rc,rd,re,rw)\
318    {\    {\
319    heapframe *newframe = (heapframe *)(pcre_stack_malloc)(sizeof(heapframe));\    heapframe *newframe = frame->Xnextframe;\
320    if (newframe == NULL) RRETURN(PCRE_ERROR_NOMEMORY);\    if (newframe == NULL)\
321    frame->Xwhere = rw; \      {\
322        newframe = (heapframe *)(PUBL(stack_malloc))(sizeof(heapframe));\
323        if (newframe == NULL) RRETURN(PCRE_ERROR_NOMEMORY);\
324        newframe->Xnextframe = NULL;\
325        frame->Xnextframe = newframe;\
326        }\
327      frame->Xwhere = rw;\
328    newframe->Xeptr = ra;\    newframe->Xeptr = ra;\
329    newframe->Xecode = rb;\    newframe->Xecode = rb;\
330    newframe->Xmstart = mstart;\    newframe->Xmstart = mstart;\
# Line 328  argument of match(), which never changes Line 343  argument of match(), which never changes
343    {\    {\
344    heapframe *oldframe = frame;\    heapframe *oldframe = frame;\
345    frame = oldframe->Xprevframe;\    frame = oldframe->Xprevframe;\
   (pcre_stack_free)(oldframe);\  
346    if (frame != NULL)\    if (frame != NULL)\
347      {\      {\
348      rrc = ra;\      rrc = ra;\
# Line 342  argument of match(), which never changes Line 356  argument of match(), which never changes
356    
357  typedef struct heapframe {  typedef struct heapframe {
358    struct heapframe *Xprevframe;    struct heapframe *Xprevframe;
359      struct heapframe *Xnextframe;
360    
361    /* Function arguments that may change */    /* Function arguments that may change */
362    
363    USPTR Xeptr;    PCRE_PUCHAR Xeptr;
364    const uschar *Xecode;    const pcre_uchar *Xecode;
365    USPTR Xmstart;    PCRE_PUCHAR Xmstart;
366    int Xoffset_top;    int Xoffset_top;
367    eptrblock *Xeptrb;    eptrblock *Xeptrb;
368    unsigned int Xrdepth;    unsigned int Xrdepth;
369    
370    /* Function local variables */    /* Function local variables */
371    
372    USPTR Xcallpat;    PCRE_PUCHAR Xcallpat;
373  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
374    USPTR Xcharptr;    PCRE_PUCHAR Xcharptr;
375  #endif  #endif
376    USPTR Xdata;    PCRE_PUCHAR Xdata;
377    USPTR Xnext;    PCRE_PUCHAR Xnext;
378    USPTR Xpp;    PCRE_PUCHAR Xpp;
379    USPTR Xprev;    PCRE_PUCHAR Xprev;
380    USPTR Xsaved_eptr;    PCRE_PUCHAR Xsaved_eptr;
381    
382    recursion_info Xnew_recursive;    recursion_info Xnew_recursive;
383    
# Line 375  typedef struct heapframe { Line 390  typedef struct heapframe {
390    int Xprop_value;    int Xprop_value;
391    int Xprop_fail_result;    int Xprop_fail_result;
392    int Xoclength;    int Xoclength;
393    uschar Xocchars[8];    pcre_uchar Xocchars[6];
394  #endif  #endif
395    
396    int Xcodelink;    int Xcodelink;
# Line 440  the subject. */ Line 455  the subject. */
455    
456    
457  /* Performance note: It might be tempting to extract commonly used fields from  /* Performance note: It might be tempting to extract commonly used fields from
458  the md structure (e.g. utf8, end_subject) into individual variables to improve  the md structure (e.g. utf, end_subject) into individual variables to improve
459  performance. Tests using gcc on a SPARC disproved this; in the first case, it  performance. Tests using gcc on a SPARC disproved this; in the first case, it
460  made performance worse.  made performance worse.
461    
# Line 463  Returns:       MATCH_MATCH if matched Line 478  Returns:       MATCH_MATCH if matched
478  */  */
479    
480  static int  static int
481  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, USPTR mstart,  match(REGISTER PCRE_PUCHAR eptr, REGISTER const pcre_uchar *ecode,
482    int offset_top, match_data *md, eptrblock *eptrb, unsigned int rdepth)    PCRE_PUCHAR mstart, int offset_top, match_data *md, eptrblock *eptrb,
483      unsigned int rdepth)
484  {  {
485  /* These variables do not need to be preserved over recursion in this function,  /* These variables do not need to be preserved over recursion in this function,
486  so they can be ordinary variables in all cases. Mark some of them with  so they can be ordinary variables in all cases. Mark some of them with
# Line 473  so they can be ordinary variables in all Line 489  so they can be ordinary variables in all
489  register int  rrc;         /* Returns from recursive calls */  register int  rrc;         /* Returns from recursive calls */
490  register int  i;           /* Used for loops not involving calls to RMATCH() */  register int  i;           /* Used for loops not involving calls to RMATCH() */
491  register unsigned int c;   /* Character values not kept over RMATCH() calls */  register unsigned int c;   /* Character values not kept over RMATCH() calls */
492  register BOOL utf8;        /* Local copy of UTF-8 flag for speed */  register BOOL utf;         /* Local copy of UTF flag for speed */
493    
494  BOOL minimize, possessive; /* Quantifier options */  BOOL minimize, possessive; /* Quantifier options */
495  BOOL caseless;  BOOL caseless;
496  int condcode;  int condcode;
497    
498  /* When recursion is not being used, all "local" variables that have to be  /* When recursion is not being used, all "local" variables that have to be
499  preserved over calls to RMATCH() are part of a "frame" which is obtained from  preserved over calls to RMATCH() are part of a "frame". We set up the top-level
500  heap storage. Set up the top-level frame here; others are obtained from the  frame on the stack here; subsequent instantiations are obtained from the heap
501  heap whenever RMATCH() does a "recursion". See the macro definitions above. */  whenever RMATCH() does a "recursion". See the macro definitions above. Putting
502    the top-level on the stack rather than malloc-ing them all gives a performance
503    boost in many cases where there is not much "recursion". */
504    
505  #ifdef NO_RECURSE  #ifdef NO_RECURSE
506  heapframe *frame = (heapframe *)(pcre_stack_malloc)(sizeof(heapframe));  heapframe *frame = (heapframe *)md->match_frames_base;
 if (frame == NULL) RRETURN(PCRE_ERROR_NOMEMORY);  
 frame->Xprevframe = NULL;            /* Marks the top level */  
507    
508  /* Copy in the original argument variables */  /* Copy in the original argument variables */
509    
# Line 513  HEAP_RECURSE: Line 529  HEAP_RECURSE:
529    
530  /* Ditto for the local variables */  /* Ditto for the local variables */
531    
532  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
533  #define charptr            frame->Xcharptr  #define charptr            frame->Xcharptr
534  #endif  #endif
535  #define callpat            frame->Xcallpat  #define callpat            frame->Xcallpat
# Line 571  declarations can be cut out in a block. Line 587  declarations can be cut out in a block.
587  below are for variables that do not have to be preserved over a recursive call  below are for variables that do not have to be preserved over a recursive call
588  to RMATCH(). */  to RMATCH(). */
589    
590  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
591  const uschar *charptr;  const pcre_uchar *charptr;
592  #endif  #endif
593  const uschar *callpat;  const pcre_uchar *callpat;
594  const uschar *data;  const pcre_uchar *data;
595  const uschar *next;  const pcre_uchar *next;
596  USPTR         pp;  PCRE_PUCHAR       pp;
597  const uschar *prev;  const pcre_uchar *prev;
598  USPTR         saved_eptr;  PCRE_PUCHAR       saved_eptr;
599    
600  recursion_info new_recursive;  recursion_info new_recursive;
601    
# Line 592  int prop_type; Line 608  int prop_type;
608  int prop_value;  int prop_value;
609  int prop_fail_result;  int prop_fail_result;
610  int oclength;  int oclength;
611  uschar occhars[8];  pcre_uchar occhars[6];
612  #endif  #endif
613    
614  int codelink;  int codelink;
# Line 608  int save_offset1, save_offset2, save_off Line 624  int save_offset1, save_offset2, save_off
624  int stacksave[REC_STACK_SAVE_MAX];  int stacksave[REC_STACK_SAVE_MAX];
625    
626  eptrblock newptrb;  eptrblock newptrb;
627    
628    /* There is a special fudge for calling match() in a way that causes it to
629    measure the size of its basic stack frame when the stack is being used for
630    recursion. The second argument (ecode) being NULL triggers this behaviour. It
631    cannot normally ever be NULL. The return is the negated value of the frame
632    size. */
633    
634    if (ecode == NULL)
635      {
636      if (rdepth == 0)
637        return match((PCRE_PUCHAR)&rdepth, NULL, NULL, 0, NULL, NULL, 1);
638      else
639        {
640        int len = (char *)&rdepth - (char *)eptr;
641        return (len > 0)? -len : len;
642        }
643      }
644  #endif     /* NO_RECURSE */  #endif     /* NO_RECURSE */
645    
646  /* To save space on the stack and in the heap frame, I have doubled up on some  /* To save space on the stack and in the heap frame, I have doubled up on some
# Line 620  the alternative names that are used. */ Line 653  the alternative names that are used. */
653  #define code_offset   codelink  #define code_offset   codelink
654  #define condassert    condition  #define condassert    condition
655  #define matched_once  prev_is_word  #define matched_once  prev_is_word
656    #define foc           number
657    #define save_mark     data
658    
659  /* These statements are here to stop the compiler complaining about unitialized  /* These statements are here to stop the compiler complaining about unitialized
660  variables. */  variables. */
# Line 645  defined). However, RMATCH isn't like a f Line 680  defined). However, RMATCH isn't like a f
680  complicated macro. It has to be used in one particular way. This shouldn't,  complicated macro. It has to be used in one particular way. This shouldn't,
681  however, impact performance when true recursion is being used. */  however, impact performance when true recursion is being used. */
682    
683  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
684  utf8 = md->utf8;       /* Local copy of the flag */  utf = md->utf;       /* Local copy of the flag */
685  #else  #else
686  utf8 = FALSE;  utf = FALSE;
687  #endif  #endif
688    
689  /* First check that we haven't called match() too many times, or that we  /* First check that we haven't called match() too many times, or that we
# Line 689  for (;;) Line 724  for (;;)
724      case OP_MARK:      case OP_MARK:
725      md->nomatch_mark = ecode + 2;      md->nomatch_mark = ecode + 2;
726      md->mark = NULL;    /* In case previously set by assertion */      md->mark = NULL;    /* In case previously set by assertion */
727      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
728        eptrb, RM55);        eptrb, RM55);
729      if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&      if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
730           md->mark == NULL) md->mark = ecode + 2;           md->mark == NULL) md->mark = ecode + 2;
# Line 702  for (;;) Line 737  for (;;)
737      unaltered. */      unaltered. */
738    
739      else if (rrc == MATCH_SKIP_ARG &&      else if (rrc == MATCH_SKIP_ARG &&
740          strcmp((char *)(ecode + 2), (char *)(md->start_match_ptr)) == 0)          STRCMP_UC_UC(ecode + 2, md->start_match_ptr) == 0)
741        {        {
742        md->start_match_ptr = eptr;        md->start_match_ptr = eptr;
743        RRETURN(MATCH_SKIP);        RRETURN(MATCH_SKIP);
# Line 715  for (;;) Line 750  for (;;)
750      /* COMMIT overrides PRUNE, SKIP, and THEN */      /* COMMIT overrides PRUNE, SKIP, and THEN */
751    
752      case OP_COMMIT:      case OP_COMMIT:
753      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
754        eptrb, RM52);        eptrb, RM52);
755      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE &&      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE &&
756          rrc != MATCH_SKIP && rrc != MATCH_SKIP_ARG &&          rrc != MATCH_SKIP && rrc != MATCH_SKIP_ARG &&
# Line 726  for (;;) Line 761  for (;;)
761      /* PRUNE overrides THEN */      /* PRUNE overrides THEN */
762    
763      case OP_PRUNE:      case OP_PRUNE:
764      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
765        eptrb, RM51);        eptrb, RM51);
766      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
767      RRETURN(MATCH_PRUNE);      RRETURN(MATCH_PRUNE);
# Line 734  for (;;) Line 769  for (;;)
769      case OP_PRUNE_ARG:      case OP_PRUNE_ARG:
770      md->nomatch_mark = ecode + 2;      md->nomatch_mark = ecode + 2;
771      md->mark = NULL;    /* In case previously set by assertion */      md->mark = NULL;    /* In case previously set by assertion */
772      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
773        eptrb, RM56);        eptrb, RM56);
774      if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&      if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
775           md->mark == NULL) md->mark = ecode + 2;           md->mark == NULL) md->mark = ecode + 2;
# Line 744  for (;;) Line 779  for (;;)
779      /* SKIP overrides PRUNE and THEN */      /* SKIP overrides PRUNE and THEN */
780    
781      case OP_SKIP:      case OP_SKIP:
782      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
783        eptrb, RM53);        eptrb, RM53);
784      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)
785        RRETURN(rrc);        RRETURN(rrc);
# Line 752  for (;;) Line 787  for (;;)
787      RRETURN(MATCH_SKIP);      RRETURN(MATCH_SKIP);
788    
789      /* Note that, for Perl compatibility, SKIP with an argument does NOT set      /* Note that, for Perl compatibility, SKIP with an argument does NOT set
790      nomatch_mark. There is a flag that disables this opcode when re-matching a      nomatch_mark. There is a flag that disables this opcode when re-matching a
791      pattern that ended with a SKIP for which there was not a matching MARK. */      pattern that ended with a SKIP for which there was not a matching MARK. */
792    
793      case OP_SKIP_ARG:      case OP_SKIP_ARG:
794      if (md->ignore_skip_arg)      if (md->ignore_skip_arg)
795        {        {
796        ecode += _pcre_OP_lengths[*ecode] + ecode[1];        ecode += PRIV(OP_lengths)[*ecode] + ecode[1];
797        break;        break;
798        }        }
799      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
800        eptrb, RM57);        eptrb, RM57);
801      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)
802        RRETURN(rrc);        RRETURN(rrc);
803    
804      /* Pass back the current skip name by overloading md->start_match_ptr and      /* Pass back the current skip name by overloading md->start_match_ptr and
805      returning the special MATCH_SKIP_ARG return code. This will either be      returning the special MATCH_SKIP_ARG return code. This will either be
806      caught by a matching MARK, or get to the top, where it causes a rematch      caught by a matching MARK, or get to the top, where it causes a rematch
807      with the md->ignore_skip_arg flag set. */      with the md->ignore_skip_arg flag set. */
808    
809      md->start_match_ptr = ecode + 2;      md->start_match_ptr = ecode + 2;
# Line 779  for (;;) Line 814  for (;;)
814      match pointer to do this. */      match pointer to do this. */
815    
816      case OP_THEN:      case OP_THEN:
817      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
818        eptrb, RM54);        eptrb, RM54);
819      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
820      md->start_match_ptr = ecode;      md->start_match_ptr = ecode;
# Line 788  for (;;) Line 823  for (;;)
823      case OP_THEN_ARG:      case OP_THEN_ARG:
824      md->nomatch_mark = ecode + 2;      md->nomatch_mark = ecode + 2;
825      md->mark = NULL;    /* In case previously set by assertion */      md->mark = NULL;    /* In case previously set by assertion */
826      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top,
827        md, eptrb, RM58);        md, eptrb, RM58);
828      if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&      if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
829           md->mark == NULL) md->mark = ecode + 2;           md->mark == NULL) md->mark = ecode + 2;
# Line 812  for (;;) Line 847  for (;;)
847      case OP_ONCE_NC:      case OP_ONCE_NC:
848      prev = ecode;      prev = ecode;
849      saved_eptr = eptr;      saved_eptr = eptr;
850        save_mark = md->mark;
851      do      do
852        {        {
853        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM64);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM64);
# Line 830  for (;;) Line 866  for (;;)
866    
867        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
868        ecode += GET(ecode,1);        ecode += GET(ecode,1);
869          md->mark = save_mark;
870        }        }
871      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
872    
# Line 869  for (;;) Line 906  for (;;)
906        }        }
907      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
908        {        {
       md->match_function_type = MATCH_CBEGROUP;  
909        RMATCH(eptr, prev, offset_top, md, eptrb, RM66);        RMATCH(eptr, prev, offset_top, md, eptrb, RM66);
910        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
911        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
# Line 909  for (;;) Line 945  for (;;)
945        save_offset2 = md->offset_vector[offset+1];        save_offset2 = md->offset_vector[offset+1];
946        save_offset3 = md->offset_vector[md->offset_end - number];        save_offset3 = md->offset_vector[md->offset_end - number];
947        save_capture_last = md->capture_last;        save_capture_last = md->capture_last;
948          save_mark = md->mark;
949    
950        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));
951        md->offset_vector[md->offset_end - number] =        md->offset_vector[md->offset_end - number] =
# Line 917  for (;;) Line 954  for (;;)
954        for (;;)        for (;;)
955          {          {
956          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
957          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
958            eptrb, RM1);            eptrb, RM1);
959          if (rrc == MATCH_ONCE) break;  /* Backing up through an atomic group */          if (rrc == MATCH_ONCE) break;  /* Backing up through an atomic group */
960    
# Line 945  for (;;) Line 982  for (;;)
982          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
983          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
984          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
985            md->mark = save_mark;
986          if (*ecode != OP_ALT) break;          if (*ecode != OP_ALT) break;
987          }          }
988    
# Line 996  for (;;) Line 1034  for (;;)
1034    
1035      for (;;)      for (;;)
1036        {        {
1037        if (op >= OP_SBRA || op == OP_ONCE) md->match_function_type = MATCH_CBEGROUP;        if (op >= OP_SBRA || op == OP_ONCE)
1038            md->match_function_type = MATCH_CBEGROUP;
1039    
1040        /* If this is not a possibly empty group, and there are no (*THEN)s in        /* If this is not a possibly empty group, and there are no (*THEN)s in
1041        the pattern, and this is the final alternative, optimize as described        the pattern, and this is the final alternative, optimize as described
# Line 1004  for (;;) Line 1043  for (;;)
1043    
1044        else if (!md->hasthen && ecode[GET(ecode, 1)] != OP_ALT)        else if (!md->hasthen && ecode[GET(ecode, 1)] != OP_ALT)
1045          {          {
1046          ecode += _pcre_OP_lengths[*ecode];          ecode += PRIV(OP_lengths)[*ecode];
1047          goto TAIL_RECURSE;          goto TAIL_RECURSE;
1048          }          }
1049    
1050        /* In all other cases, we have to make another call to match(). */        /* In all other cases, we have to make another call to match(). */
1051    
1052        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, eptrb,        save_mark = md->mark;
1053          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb,
1054          RM2);          RM2);
1055    
1056        /* See comment in the code for capturing groups above about handling        /* See comment in the code for capturing groups above about handling
# Line 1028  for (;;) Line 1068  for (;;)
1068          {          {
1069          if (rrc == MATCH_ONCE)          if (rrc == MATCH_ONCE)
1070            {            {
1071            const uschar *scode = ecode;            const pcre_uchar *scode = ecode;
1072            if (*scode != OP_ONCE)           /* If not at start, find it */            if (*scode != OP_ONCE)           /* If not at start, find it */
1073              {              {
1074              while (*scode == OP_ALT) scode += GET(scode, 1);              while (*scode == OP_ALT) scode += GET(scode, 1);
# Line 1039  for (;;) Line 1079  for (;;)
1079          RRETURN(rrc);          RRETURN(rrc);
1080          }          }
1081        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1082          md->mark = save_mark;
1083        if (*ecode != OP_ALT) break;        if (*ecode != OP_ALT) break;
1084        }        }
1085    
# Line 1070  for (;;) Line 1111  for (;;)
1111      if (offset < md->offset_max)      if (offset < md->offset_max)
1112        {        {
1113        matched_once = FALSE;        matched_once = FALSE;
1114        code_offset = ecode - md->start_code;        code_offset = (int)(ecode - md->start_code);
1115    
1116        save_offset1 = md->offset_vector[offset];        save_offset1 = md->offset_vector[offset];
1117        save_offset2 = md->offset_vector[offset+1];        save_offset2 = md->offset_vector[offset+1];
# Line 1093  for (;;) Line 1134  for (;;)
1134          md->offset_vector[md->offset_end - number] =          md->offset_vector[md->offset_end - number] =
1135            (int)(eptr - md->start_subject);            (int)(eptr - md->start_subject);
1136          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
1137          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
1138            eptrb, RM63);            eptrb, RM63);
1139          if (rrc == MATCH_KETRPOS)          if (rrc == MATCH_KETRPOS)
1140            {            {
# Line 1160  for (;;) Line 1201  for (;;)
1201    
1202      POSSESSIVE_NON_CAPTURE:      POSSESSIVE_NON_CAPTURE:
1203      matched_once = FALSE;      matched_once = FALSE;
1204      code_offset = ecode - md->start_code;      code_offset = (int)(ecode - md->start_code);
1205    
1206      for (;;)      for (;;)
1207        {        {
1208        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
1209        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
1210          eptrb, RM48);          eptrb, RM48);
1211        if (rrc == MATCH_KETRPOS)        if (rrc == MATCH_KETRPOS)
1212          {          {
# Line 1215  for (;;) Line 1256  for (;;)
1256    
1257      if (ecode[LINK_SIZE+1] == OP_CALLOUT)      if (ecode[LINK_SIZE+1] == OP_CALLOUT)
1258        {        {
1259        if (pcre_callout != NULL)        if (PUBL(callout) != NULL)
1260          {          {
1261          pcre_callout_block cb;          PUBL(callout_block) cb;
1262          cb.version          = 2;   /* Version 1 of the callout block */          cb.version          = 2;   /* Version 1 of the callout block */
1263          cb.callout_number   = ecode[LINK_SIZE+2];          cb.callout_number   = ecode[LINK_SIZE+2];
1264          cb.offset_vector    = md->offset_vector;          cb.offset_vector    = md->offset_vector;
1265    #ifdef COMPILE_PCRE8
1266          cb.subject          = (PCRE_SPTR)md->start_subject;          cb.subject          = (PCRE_SPTR)md->start_subject;
1267    #else
1268            cb.subject          = (PCRE_SPTR16)md->start_subject;
1269    #endif
1270          cb.subject_length   = (int)(md->end_subject - md->start_subject);          cb.subject_length   = (int)(md->end_subject - md->start_subject);
1271          cb.start_match      = (int)(mstart - md->start_subject);          cb.start_match      = (int)(mstart - md->start_subject);
1272          cb.current_position = (int)(eptr - md->start_subject);          cb.current_position = (int)(eptr - md->start_subject);
# Line 1231  for (;;) Line 1276  for (;;)
1276          cb.capture_last     = md->capture_last;          cb.capture_last     = md->capture_last;
1277          cb.callout_data     = md->callout_data;          cb.callout_data     = md->callout_data;
1278          cb.mark             = md->nomatch_mark;          cb.mark             = md->nomatch_mark;
1279          if ((rrc = (*pcre_callout)(&cb)) > 0) RRETURN(MATCH_NOMATCH);          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1280          if (rrc < 0) RRETURN(rrc);          if (rrc < 0) RRETURN(rrc);
1281          }          }
1282        ecode += _pcre_OP_lengths[OP_CALLOUT];        ecode += PRIV(OP_lengths)[OP_CALLOUT];
1283        }        }
1284    
1285      condcode = ecode[LINK_SIZE+1];      condcode = ecode[LINK_SIZE+1];
# Line 1260  for (;;) Line 1305  for (;;)
1305    
1306          if (!condition && condcode == OP_NRREF)          if (!condition && condcode == OP_NRREF)
1307            {            {
1308            uschar *slotA = md->name_table;            pcre_uchar *slotA = md->name_table;
1309            for (i = 0; i < md->name_count; i++)            for (i = 0; i < md->name_count; i++)
1310              {              {
1311              if (GET2(slotA, 0) == recno) break;              if (GET2(slotA, 0) == recno) break;
# Line 1273  for (;;) Line 1318  for (;;)
1318    
1319            if (i < md->name_count)            if (i < md->name_count)
1320              {              {
1321              uschar *slotB = slotA;              pcre_uchar *slotB = slotA;
1322              while (slotB > md->name_table)              while (slotB > md->name_table)
1323                {                {
1324                slotB -= md->name_entry_size;                slotB -= md->name_entry_size;
1325                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1326                  {                  {
1327                  condition = GET2(slotB, 0) == md->recursive->group_num;                  condition = GET2(slotB, 0) == md->recursive->group_num;
1328                  if (condition) break;                  if (condition) break;
# Line 1293  for (;;) Line 1338  for (;;)
1338                for (i++; i < md->name_count; i++)                for (i++; i < md->name_count; i++)
1339                  {                  {
1340                  slotB += md->name_entry_size;                  slotB += md->name_entry_size;
1341                  if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                  if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1342                    {                    {
1343                    condition = GET2(slotB, 0) == md->recursive->group_num;                    condition = GET2(slotB, 0) == md->recursive->group_num;
1344                    if (condition) break;                    if (condition) break;
# Line 1306  for (;;) Line 1351  for (;;)
1351    
1352          /* Chose branch according to the condition */          /* Chose branch according to the condition */
1353    
1354          ecode += condition? 3 : GET(ecode, 1);          ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
1355          }          }
1356        }        }
1357    
# Line 1323  for (;;) Line 1368  for (;;)
1368        if (!condition && condcode == OP_NCREF)        if (!condition && condcode == OP_NCREF)
1369          {          {
1370          int refno = offset >> 1;          int refno = offset >> 1;
1371          uschar *slotA = md->name_table;          pcre_uchar *slotA = md->name_table;
1372    
1373          for (i = 0; i < md->name_count; i++)          for (i = 0; i < md->name_count; i++)
1374            {            {
# Line 1337  for (;;) Line 1382  for (;;)
1382    
1383          if (i < md->name_count)          if (i < md->name_count)
1384            {            {
1385            uschar *slotB = slotA;            pcre_uchar *slotB = slotA;
1386            while (slotB > md->name_table)            while (slotB > md->name_table)
1387              {              {
1388              slotB -= md->name_entry_size;              slotB -= md->name_entry_size;
1389              if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)              if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1390                {                {
1391                offset = GET2(slotB, 0) << 1;                offset = GET2(slotB, 0) << 1;
1392                condition = offset < offset_top &&                condition = offset < offset_top &&
# Line 1359  for (;;) Line 1404  for (;;)
1404              for (i++; i < md->name_count; i++)              for (i++; i < md->name_count; i++)
1405                {                {
1406                slotB += md->name_entry_size;                slotB += md->name_entry_size;
1407                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1408                  {                  {
1409                  offset = GET2(slotB, 0) << 1;                  offset = GET2(slotB, 0) << 1;
1410                  condition = offset < offset_top &&                  condition = offset < offset_top &&
# Line 1374  for (;;) Line 1419  for (;;)
1419    
1420        /* Chose branch according to the condition */        /* Chose branch according to the condition */
1421    
1422        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
1423        }        }
1424    
1425      else if (condcode == OP_DEF)     /* DEFINE - always false */      else if (condcode == OP_DEF)     /* DEFINE - always false */
# Line 1466  for (;;) Line 1511  for (;;)
1511        md->offset_vector[offset+1] = (int)(eptr - md->start_subject);        md->offset_vector[offset+1] = (int)(eptr - md->start_subject);
1512        if (offset_top <= offset) offset_top = offset + 2;        if (offset_top <= offset) offset_top = offset + 2;
1513        }        }
1514      ecode += 3;      ecode += 1 + IMM2_SIZE;
1515      break;      break;
1516    
1517    
# Line 1513  for (;;) Line 1558  for (;;)
1558    
1559      case OP_ASSERT:      case OP_ASSERT:
1560      case OP_ASSERTBACK:      case OP_ASSERTBACK:
1561        save_mark = md->mark;
1562      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1563        {        {
1564        condassert = TRUE;        condassert = TRUE;
# Line 1528  for (;;) Line 1574  for (;;)
1574          mstart = md->start_match_ptr;   /* In case \K reset it */          mstart = md->start_match_ptr;   /* In case \K reset it */
1575          break;          break;
1576          }          }
1577          md->mark = save_mark;
1578    
1579        /* PCRE does not allow THEN to escape beyond an assertion; it is treated        /* A COMMIT failure must fail the entire assertion, without trying any
1580        as NOMATCH. */        subsequent branches. */
1581    
1582          if (rrc == MATCH_COMMIT) RRETURN(MATCH_NOMATCH);
1583    
1584          /* PCRE does not allow THEN to escape beyond an assertion; it
1585          is treated as NOMATCH. */
1586    
1587        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1588        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
# Line 1557  for (;;) Line 1609  for (;;)
1609    
1610      case OP_ASSERT_NOT:      case OP_ASSERT_NOT:
1611      case OP_ASSERTBACK_NOT:      case OP_ASSERTBACK_NOT:
1612        save_mark = md->mark;
1613      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1614        {        {
1615        condassert = TRUE;        condassert = TRUE;
# Line 1567  for (;;) Line 1620  for (;;)
1620      do      do
1621        {        {
1622        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);
1623          md->mark = save_mark;
1624        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) RRETURN(MATCH_NOMATCH);        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) RRETURN(MATCH_NOMATCH);
1625        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)
1626          {          {
# Line 1593  for (;;) Line 1647  for (;;)
1647      back a number of characters, not bytes. */      back a number of characters, not bytes. */
1648    
1649      case OP_REVERSE:      case OP_REVERSE:
1650  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
1651      if (utf8)      if (utf)
1652        {        {
1653        i = GET(ecode, 1);        i = GET(ecode, 1);
1654        while (i-- > 0)        while (i-- > 0)
# Line 1625  for (;;) Line 1679  for (;;)
1679      function is able to force a failure. */      function is able to force a failure. */
1680    
1681      case OP_CALLOUT:      case OP_CALLOUT:
1682      if (pcre_callout != NULL)      if (PUBL(callout) != NULL)
1683        {        {
1684        pcre_callout_block cb;        PUBL(callout_block) cb;
1685        cb.version          = 2;   /* Version 1 of the callout block */        cb.version          = 2;   /* Version 1 of the callout block */
1686        cb.callout_number   = ecode[1];        cb.callout_number   = ecode[1];
1687        cb.offset_vector    = md->offset_vector;        cb.offset_vector    = md->offset_vector;
1688    #ifdef COMPILE_PCRE8
1689        cb.subject          = (PCRE_SPTR)md->start_subject;        cb.subject          = (PCRE_SPTR)md->start_subject;
1690    #else
1691          cb.subject          = (PCRE_SPTR16)md->start_subject;
1692    #endif
1693        cb.subject_length   = (int)(md->end_subject - md->start_subject);        cb.subject_length   = (int)(md->end_subject - md->start_subject);
1694        cb.start_match      = (int)(mstart - md->start_subject);        cb.start_match      = (int)(mstart - md->start_subject);
1695        cb.current_position = (int)(eptr - md->start_subject);        cb.current_position = (int)(eptr - md->start_subject);
# Line 1641  for (;;) Line 1699  for (;;)
1699        cb.capture_last     = md->capture_last;        cb.capture_last     = md->capture_last;
1700        cb.callout_data     = md->callout_data;        cb.callout_data     = md->callout_data;
1701        cb.mark             = md->nomatch_mark;        cb.mark             = md->nomatch_mark;
1702        if ((rrc = (*pcre_callout)(&cb)) > 0) RRETURN(MATCH_NOMATCH);        if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1703        if (rrc < 0) RRETURN(rrc);        if (rrc < 0) RRETURN(rrc);
1704        }        }
1705      ecode += 2 + 2*LINK_SIZE;      ecode += 2 + 2*LINK_SIZE;
# Line 1700  for (;;) Line 1758  for (;;)
1758        else        else
1759          {          {
1760          new_recursive.offset_save =          new_recursive.offset_save =
1761            (int *)(pcre_malloc)(new_recursive.saved_max * sizeof(int));            (int *)(PUBL(malloc))(new_recursive.saved_max * sizeof(int));
1762          if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY);          if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY);
1763          }          }
1764        memcpy(new_recursive.offset_save, md->offset_vector,        memcpy(new_recursive.offset_save, md->offset_vector,
# Line 1715  for (;;) Line 1773  for (;;)
1773        do        do
1774          {          {
1775          if (cbegroup) md->match_function_type = MATCH_CBEGROUP;          if (cbegroup) md->match_function_type = MATCH_CBEGROUP;
1776          RMATCH(eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,          RMATCH(eptr, callpat + PRIV(OP_lengths)[*callpat], offset_top,
1777            md, eptrb, RM6);            md, eptrb, RM6);
1778          memcpy(md->offset_vector, new_recursive.offset_save,          memcpy(md->offset_vector, new_recursive.offset_save,
1779              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
# Line 1724  for (;;) Line 1782  for (;;)
1782            {            {
1783            DPRINTF(("Recursion matched\n"));            DPRINTF(("Recursion matched\n"));
1784            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
1785              (pcre_free)(new_recursive.offset_save);              (PUBL(free))(new_recursive.offset_save);
1786    
1787            /* Set where we got to in the subject, and reset the start in case            /* Set where we got to in the subject, and reset the start in case
1788            it was changed by \K. This *is* propagated back out of a recursion,            it was changed by \K. This *is* propagated back out of a recursion,
# Line 1735  for (;;) Line 1793  for (;;)
1793            goto RECURSION_MATCHED;        /* Exit loop; end processing */            goto RECURSION_MATCHED;        /* Exit loop; end processing */
1794            }            }
1795    
1796          /* PCRE does not allow THEN to escape beyond a recursion; it is treated          /* PCRE does not allow THEN or COMMIT to escape beyond a recursion; it
1797          as NOMATCH. */          is treated as NOMATCH. */
1798    
1799          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN &&
1800                     rrc != MATCH_COMMIT)
1801            {            {
1802            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1803            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
1804              (pcre_free)(new_recursive.offset_save);              (PUBL(free))(new_recursive.offset_save);
1805            RRETURN(rrc);            RRETURN(rrc);
1806            }            }
1807    
# Line 1754  for (;;) Line 1813  for (;;)
1813        DPRINTF(("Recursion didn't match\n"));        DPRINTF(("Recursion didn't match\n"));
1814        md->recursive = new_recursive.prevrec;        md->recursive = new_recursive.prevrec;
1815        if (new_recursive.offset_save != stacksave)        if (new_recursive.offset_save != stacksave)
1816          (pcre_free)(new_recursive.offset_save);          (PUBL(free))(new_recursive.offset_save);
1817        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
1818        }        }
1819    
# Line 1949  for (;;) Line 2008  for (;;)
2008          }          }
2009        if (*prev >= OP_SBRA)    /* Could match an empty string */        if (*prev >= OP_SBRA)    /* Could match an empty string */
2010          {          {
         md->match_function_type = MATCH_CBEGROUP;  
2011          RMATCH(eptr, prev, offset_top, md, eptrb, RM50);          RMATCH(eptr, prev, offset_top, md, eptrb, RM50);
2012          RRETURN(rrc);          RRETURN(rrc);
2013          }          }
# Line 1958  for (;;) Line 2016  for (;;)
2016        }        }
2017      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
2018        {        {
       if (*prev >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;  
2019        RMATCH(eptr, prev, offset_top, md, eptrb, RM13);        RMATCH(eptr, prev, offset_top, md, eptrb, RM13);
2020        if (rrc == MATCH_ONCE && md->once_target == prev) rrc = MATCH_NOMATCH;        if (rrc == MATCH_ONCE && md->once_target == prev) rrc = MATCH_NOMATCH;
2021        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
# Line 2015  for (;;) Line 2072  for (;;)
2072    
2073      case OP_DOLLM:      case OP_DOLLM:
2074      if (eptr < md->end_subject)      if (eptr < md->end_subject)
2075        { if (!IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); }        {
2076          if (!IS_NEWLINE(eptr))
2077            {
2078            if (md->partial != 0 &&
2079                eptr + 1 >= md->end_subject &&
2080                NLBLOCK->nltype == NLTYPE_FIXED &&
2081                NLBLOCK->nllen == 2 &&
2082                *eptr == NLBLOCK->nl[0])
2083              {
2084              md->hitend = TRUE;
2085              if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
2086              }
2087            RRETURN(MATCH_NOMATCH);
2088            }
2089          }
2090      else      else
2091        {        {
2092        if (md->noteol) RRETURN(MATCH_NOMATCH);        if (md->noteol) RRETURN(MATCH_NOMATCH);
# Line 2047  for (;;) Line 2118  for (;;)
2118      ASSERT_NL_OR_EOS:      ASSERT_NL_OR_EOS:
2119      if (eptr < md->end_subject &&      if (eptr < md->end_subject &&
2120          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))
2121          {
2122          if (md->partial != 0 &&
2123              eptr + 1 >= md->end_subject &&
2124              NLBLOCK->nltype == NLTYPE_FIXED &&
2125              NLBLOCK->nllen == 2 &&
2126              *eptr == NLBLOCK->nl[0])
2127            {
2128            md->hitend = TRUE;
2129            if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
2130            }
2131        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2132          }
2133    
2134      /* Either at end of string or \n before end. */      /* Either at end of string or \n before end. */
2135    
# Line 2066  for (;;) Line 2148  for (;;)
2148        be "non-word" characters. Remember the earliest consulted character for        be "non-word" characters. Remember the earliest consulted character for
2149        partial matching. */        partial matching. */
2150    
2151  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2152        if (utf8)        if (utf)
2153          {          {
2154          /* Get status of previous character */          /* Get status of previous character */
2155    
2156          if (eptr == md->start_subject) prev_is_word = FALSE; else          if (eptr == md->start_subject) prev_is_word = FALSE; else
2157            {            {
2158            USPTR lastptr = eptr - 1;            PCRE_PUCHAR lastptr = eptr - 1;
2159            while((*lastptr & 0xc0) == 0x80) lastptr--;            BACKCHAR(lastptr);
2160            if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;            if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;
2161            GETCHAR(c, lastptr);            GETCHAR(c, lastptr);
2162  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 2139  for (;;) Line 2221  for (;;)
2221              }              }
2222            else            else
2223  #endif  #endif
2224            prev_is_word = ((md->ctypes[eptr[-1]] & ctype_word) != 0);            prev_is_word = MAX_255(eptr[-1])
2225                && ((md->ctypes[eptr[-1]] & ctype_word) != 0);
2226            }            }
2227    
2228          /* Get status of next character */          /* Get status of next character */
# Line 2162  for (;;) Line 2245  for (;;)
2245            }            }
2246          else          else
2247  #endif  #endif
2248          cur_is_word = ((md->ctypes[*eptr] & ctype_word) != 0);          cur_is_word = MAX_255(*eptr)
2249              && ((md->ctypes[*eptr] & ctype_word) != 0);
2250          }          }
2251    
2252        /* Now see if the situation is what we want */        /* Now see if the situation is what we want */
# Line 2173  for (;;) Line 2257  for (;;)
2257        }        }
2258      break;      break;
2259    
2260      /* Match a single character type; inline for speed */      /* Match any single character type except newline; have to take care with
2261        CRLF newlines and partial matching. */
2262    
2263      case OP_ANY:      case OP_ANY:
2264      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
2265        if (md->partial != 0 &&
2266            eptr + 1 >= md->end_subject &&
2267            NLBLOCK->nltype == NLTYPE_FIXED &&
2268            NLBLOCK->nllen == 2 &&
2269            *eptr == NLBLOCK->nl[0])
2270          {
2271          md->hitend = TRUE;
2272          if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
2273          }
2274    
2275      /* Fall through */      /* Fall through */
2276    
2277        /* Match any single character whatsoever. */
2278    
2279      case OP_ALLANY:      case OP_ALLANY:
2280      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */
2281        {                            /* not be updated before SCHECK_PARTIAL. */        {                            /* not be updated before SCHECK_PARTIAL. */
# Line 2186  for (;;) Line 2283  for (;;)
2283        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2284        }        }
2285      eptr++;      eptr++;
2286      if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  #ifdef SUPPORT_UTF
2287        if (utf) ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
2288    #endif
2289      ecode++;      ecode++;
2290      break;      break;
2291    
# Line 2211  for (;;) Line 2310  for (;;)
2310        }        }
2311      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2312      if (      if (
2313  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2314         c < 256 &&         c < 256 &&
2315  #endif  #endif
2316         (md->ctypes[c] & ctype_digit) != 0         (md->ctypes[c] & ctype_digit) != 0
# Line 2228  for (;;) Line 2327  for (;;)
2327        }        }
2328      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2329      if (      if (
2330  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2331         c >= 256 ||         c > 255 ||
2332  #endif  #endif
2333         (md->ctypes[c] & ctype_digit) == 0         (md->ctypes[c] & ctype_digit) == 0
2334         )         )
# Line 2245  for (;;) Line 2344  for (;;)
2344        }        }
2345      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2346      if (      if (
2347  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2348         c < 256 &&         c < 256 &&
2349  #endif  #endif
2350         (md->ctypes[c] & ctype_space) != 0         (md->ctypes[c] & ctype_space) != 0
# Line 2262  for (;;) Line 2361  for (;;)
2361        }        }
2362      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2363      if (      if (
2364  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2365         c >= 256 ||         c > 255 ||
2366  #endif  #endif
2367         (md->ctypes[c] & ctype_space) == 0         (md->ctypes[c] & ctype_space) == 0
2368         )         )
# Line 2279  for (;;) Line 2378  for (;;)
2378        }        }
2379      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2380      if (      if (
2381  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2382         c < 256 &&         c < 256 &&
2383  #endif  #endif
2384         (md->ctypes[c] & ctype_word) != 0         (md->ctypes[c] & ctype_word) != 0
# Line 2296  for (;;) Line 2395  for (;;)
2395        }        }
2396      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2397      if (      if (
2398  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2399         c >= 256 ||         c > 255 ||
2400  #endif  #endif
2401         (md->ctypes[c] & ctype_word) == 0         (md->ctypes[c] & ctype_word) == 0
2402         )         )
# Line 2317  for (;;) Line 2416  for (;;)
2416        default: RRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
2417    
2418        case 0x000d:        case 0x000d:
2419        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr >= md->end_subject)
2420            {
2421            SCHECK_PARTIAL();
2422            }
2423          else if (*eptr == 0x0a) eptr++;
2424        break;        break;
2425    
2426        case 0x000a:        case 0x000a:
# Line 2475  for (;;) Line 2578  for (;;)
2578          break;          break;
2579    
2580          case PT_GC:          case PT_GC:
2581          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))          if ((ecode[2] != PRIV(ucp_gentype)[prop->chartype]) == (op == OP_PROP))
2582            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2583          break;          break;
2584    
# Line 2492  for (;;) Line 2595  for (;;)
2595          /* These are specials */          /* These are specials */
2596    
2597          case PT_ALNUM:          case PT_ALNUM:
2598          if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
2599               _pcre_ucp_gentype[prop->chartype] == ucp_N) == (op == OP_NOTPROP))               PRIV(ucp_gentype)[prop->chartype] == ucp_N) == (op == OP_NOTPROP))
2600            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2601          break;          break;
2602    
2603          case PT_SPACE:    /* Perl space */          case PT_SPACE:    /* Perl space */
2604          if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
2605               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
2606                 == (op == OP_NOTPROP))                 == (op == OP_NOTPROP))
2607            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2608          break;          break;
2609    
2610          case PT_PXSPACE:  /* POSIX space */          case PT_PXSPACE:  /* POSIX space */
2611          if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
2612               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
2613               c == CHAR_FF || c == CHAR_CR)               c == CHAR_FF || c == CHAR_CR)
2614                 == (op == OP_NOTPROP))                 == (op == OP_NOTPROP))
# Line 2513  for (;;) Line 2616  for (;;)
2616          break;          break;
2617    
2618          case PT_WORD:          case PT_WORD:
2619          if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
2620               _pcre_ucp_gentype[prop->chartype] == ucp_N ||               PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
2621               c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))               c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))
2622            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2623          break;          break;
# Line 2543  for (;;) Line 2646  for (;;)
2646      while (eptr < md->end_subject)      while (eptr < md->end_subject)
2647        {        {
2648        int len = 1;        int len = 1;
2649        if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }        if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
2650        if (UCD_CATEGORY(c) != ucp_M) break;        if (UCD_CATEGORY(c) != ucp_M) break;
2651        eptr += len;        eptr += len;
2652        }        }
2653        CHECK_PARTIAL();
2654      ecode++;      ecode++;
2655      break;      break;
2656  #endif  #endif
# Line 2564  for (;;) Line 2668  for (;;)
2668      case OP_REFI:      case OP_REFI:
2669      caseless = op == OP_REFI;      caseless = op == OP_REFI;
2670      offset = GET2(ecode, 1) << 1;               /* Doubled ref number */      offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
2671      ecode += 3;      ecode += 1 + IMM2_SIZE;
2672    
2673      /* If the reference is unset, there are two possibilities:      /* If the reference is unset, there are two possibilities:
2674    
# Line 2604  for (;;) Line 2708  for (;;)
2708        case OP_CRMINRANGE:        case OP_CRMINRANGE:
2709        minimize = (*ecode == OP_CRMINRANGE);        minimize = (*ecode == OP_CRMINRANGE);
2710        min = GET2(ecode, 1);        min = GET2(ecode, 1);
2711        max = GET2(ecode, 3);        max = GET2(ecode, 1 + IMM2_SIZE);
2712        if (max == 0) max = INT_MAX;        if (max == 0) max = INT_MAX;
2713        ecode += 5;        ecode += 1 + 2 * IMM2_SIZE;
2714        break;        break;
2715    
2716        default:               /* No repeat follows */        default:               /* No repeat follows */
2717        if ((length = match_ref(offset, eptr, length, md, caseless)) < 0)        if ((length = match_ref(offset, eptr, length, md, caseless)) < 0)
2718          {          {
2719            if (length == -2) eptr = md->end_subject;   /* Partial match */
2720          CHECK_PARTIAL();          CHECK_PARTIAL();
2721          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2722          }          }
# Line 2620  for (;;) Line 2725  for (;;)
2725        }        }
2726    
2727      /* Handle repeated back references. If the length of the reference is      /* Handle repeated back references. If the length of the reference is
2728      zero, just continue with the main loop. */      zero, just continue with the main loop. If the length is negative, it
2729        means the reference is unset in non-Java-compatible mode. If the minimum is
2730        zero, we can continue at the same level without recursion. For any other
2731        minimum, carrying on will result in NOMATCH. */
2732    
2733      if (length == 0) continue;      if (length == 0) continue;
2734        if (length < 0 && min == 0) continue;
2735    
2736      /* First, ensure the minimum number of matches are present. We get back      /* First, ensure the minimum number of matches are present. We get back
2737      the length of the reference string explicitly rather than passing the      the length of the reference string explicitly rather than passing the
# Line 2633  for (;;) Line 2742  for (;;)
2742        int slength;        int slength;
2743        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2744          {          {
2745            if (slength == -2) eptr = md->end_subject;   /* Partial match */
2746          CHECK_PARTIAL();          CHECK_PARTIAL();
2747          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2748          }          }
# Line 2656  for (;;) Line 2766  for (;;)
2766          if (fi >= max) RRETURN(MATCH_NOMATCH);          if (fi >= max) RRETURN(MATCH_NOMATCH);
2767          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2768            {            {
2769              if (slength == -2) eptr = md->end_subject;   /* Partial match */
2770            CHECK_PARTIAL();            CHECK_PARTIAL();
2771            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2772            }            }
# Line 2674  for (;;) Line 2785  for (;;)
2785          int slength;          int slength;
2786          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2787            {            {
2788            CHECK_PARTIAL();            /* Can't use CHECK_PARTIAL because we don't want to update eptr in
2789              the soft partial matching case. */
2790    
2791              if (slength == -2 && md->partial != 0 &&
2792                  md->end_subject > md->start_used_ptr)
2793                {
2794                md->hitend = TRUE;
2795                if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
2796                }
2797            break;            break;
2798            }            }
2799          eptr += slength;          eptr += slength;
2800          }          }
2801    
2802        while (eptr >= pp)        while (eptr >= pp)
2803          {          {
2804          RMATCH(eptr, ecode, offset_top, md, eptrb, RM15);          RMATCH(eptr, ecode, offset_top, md, eptrb, RM15);
# Line 2703  for (;;) Line 2823  for (;;)
2823      case OP_NCLASS:      case OP_NCLASS:
2824      case OP_CLASS:      case OP_CLASS:
2825        {        {
2826          /* The data variable is saved across frames, so the byte map needs to
2827          be stored there. */
2828    #define BYTE_MAP ((pcre_uint8 *)data)
2829        data = ecode + 1;                /* Save for matching */        data = ecode + 1;                /* Save for matching */
2830        ecode += 33;                     /* Advance past the item */        ecode += 1 + (32 / sizeof(pcre_uchar)); /* Advance past the item */
2831    
2832        switch (*ecode)        switch (*ecode)
2833          {          {
# Line 2725  for (;;) Line 2848  for (;;)
2848          case OP_CRMINRANGE:          case OP_CRMINRANGE:
2849          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
2850          min = GET2(ecode, 1);          min = GET2(ecode, 1);
2851          max = GET2(ecode, 3);          max = GET2(ecode, 1 + IMM2_SIZE);
2852          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
2853          ecode += 5;          ecode += 1 + 2 * IMM2_SIZE;
2854          break;          break;
2855    
2856          default:               /* No repeat follows */          default:               /* No repeat follows */
# Line 2737  for (;;) Line 2860  for (;;)
2860    
2861        /* First, ensure the minimum number of matches are present. */        /* First, ensure the minimum number of matches are present. */
2862    
2863  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2864        /* UTF-8 mode */        if (utf)
       if (utf8)  
2865          {          {
2866          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2867            {            {
# Line 2754  for (;;) Line 2876  for (;;)
2876              if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);              if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2877              }              }
2878            else            else
2879              {              if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
             if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);  
             }  
2880            }            }
2881          }          }
2882        else        else
2883  #endif  #endif
2884        /* Not UTF-8 mode */        /* Not UTF mode */
2885          {          {
2886          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2887            {            {
# Line 2771  for (;;) Line 2891  for (;;)
2891              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2892              }              }
2893            c = *eptr++;            c = *eptr++;
2894            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);  #ifndef COMPILE_PCRE8
2895              if (c > 255)
2896                {
2897                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2898                }
2899              else
2900    #endif
2901                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2902            }            }
2903          }          }
2904    
# Line 2785  for (;;) Line 2912  for (;;)
2912    
2913        if (minimize)        if (minimize)
2914          {          {
2915  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2916          /* UTF-8 mode */          if (utf)
         if (utf8)  
2917            {            {
2918            for (fi = min;; fi++)            for (fi = min;; fi++)
2919              {              {
# Line 2805  for (;;) Line 2931  for (;;)
2931                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2932                }                }
2933              else              else
2934                {                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
               if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);  
               }  
2935              }              }
2936            }            }
2937          else          else
2938  #endif  #endif
2939          /* Not UTF-8 mode */          /* Not UTF mode */
2940            {            {
2941            for (fi = min;; fi++)            for (fi = min;; fi++)
2942              {              {
# Line 2825  for (;;) Line 2949  for (;;)
2949                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2950                }                }
2951              c = *eptr++;              c = *eptr++;
2952              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);  #ifndef COMPILE_PCRE8
2953                if (c > 255)
2954                  {
2955                  if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2956                  }
2957                else
2958    #endif
2959                  if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2960              }              }
2961            }            }
2962          /* Control never gets here */          /* Control never gets here */
# Line 2837  for (;;) Line 2968  for (;;)
2968          {          {
2969          pp = eptr;          pp = eptr;
2970    
2971  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2972          /* UTF-8 mode */          if (utf)
         if (utf8)  
2973            {            {
2974            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2975              {              {
# Line 2855  for (;;) Line 2985  for (;;)
2985                if (op == OP_CLASS) break;                if (op == OP_CLASS) break;
2986                }                }
2987              else              else
2988                {                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
               if ((data[c/8] & (1 << (c&7))) == 0) break;  
               }  
2989              eptr += len;              eptr += len;
2990              }              }
2991            for (;;)            for (;;)
# Line 2870  for (;;) Line 2998  for (;;)
2998            }            }
2999          else          else
3000  #endif  #endif
3001            /* Not UTF-8 mode */            /* Not UTF mode */
3002            {            {
3003            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3004              {              {
# Line 2880  for (;;) Line 3008  for (;;)
3008                break;                break;
3009                }                }
3010              c = *eptr;              c = *eptr;
3011              if ((data[c/8] & (1 << (c&7))) == 0) break;  #ifndef COMPILE_PCRE8
3012                if (c > 255)
3013                  {
3014                  if (op == OP_CLASS) break;
3015                  }
3016                else
3017    #endif
3018                  if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
3019              eptr++;              eptr++;
3020              }              }
3021            while (eptr >= pp)            while (eptr >= pp)
# Line 2893  for (;;) Line 3028  for (;;)
3028    
3029          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3030          }          }
3031    #undef BYTE_MAP
3032        }        }
3033      /* Control never gets here */      /* Control never gets here */
3034    
# Line 2901  for (;;) Line 3037  for (;;)
3037      when UTF-8 mode mode is supported. Nevertheless, we may not be in UTF-8      when UTF-8 mode mode is supported. Nevertheless, we may not be in UTF-8
3038      mode, because Unicode properties are supported in non-UTF-8 mode. */      mode, because Unicode properties are supported in non-UTF-8 mode. */
3039    
3040  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
3041      case OP_XCLASS:      case OP_XCLASS:
3042        {        {
3043        data = ecode + 1 + LINK_SIZE;                /* Save for matching */        data = ecode + 1 + LINK_SIZE;                /* Save for matching */
# Line 2926  for (;;) Line 3062  for (;;)
3062          case OP_CRMINRANGE:          case OP_CRMINRANGE:
3063          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
3064          min = GET2(ecode, 1);          min = GET2(ecode, 1);
3065          max = GET2(ecode, 3);          max = GET2(ecode, 1 + IMM2_SIZE);
3066          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
3067          ecode += 5;          ecode += 1 + 2 * IMM2_SIZE;
3068          break;          break;
3069    
3070          default:               /* No repeat follows */          default:               /* No repeat follows */
# Line 2946  for (;;) Line 3082  for (;;)
3082            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3083            }            }
3084          GETCHARINCTEST(c, eptr);          GETCHARINCTEST(c, eptr);
3085          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);          if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH);
3086          }          }
3087    
3088        /* If max == min we can continue with the main loop without the        /* If max == min we can continue with the main loop without the
# Line 2970  for (;;) Line 3106  for (;;)
3106              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3107              }              }
3108            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3109            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);            if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH);
3110            }            }
3111          /* Control never gets here */          /* Control never gets here */
3112          }          }
# Line 2988  for (;;) Line 3124  for (;;)
3124              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3125              break;              break;
3126              }              }
3127    #ifdef SUPPORT_UTF
3128            GETCHARLENTEST(c, eptr, len);            GETCHARLENTEST(c, eptr, len);
3129            if (!_pcre_xclass(c, data)) break;  #else
3130              c = *eptr;
3131    #endif
3132              if (!PRIV(xclass)(c, data, utf)) break;
3133            eptr += len;            eptr += len;
3134            }            }
3135          for(;;)          for(;;)
# Line 2997  for (;;) Line 3137  for (;;)
3137            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);
3138            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3139            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
3140            if (utf8) BACKCHAR(eptr);  #ifdef SUPPORT_UTF
3141              if (utf) BACKCHAR(eptr);
3142    #endif
3143            }            }
3144          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3145          }          }
# Line 3009  for (;;) Line 3151  for (;;)
3151      /* Match a single character, casefully */      /* Match a single character, casefully */
3152    
3153      case OP_CHAR:      case OP_CHAR:
3154  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3155      if (utf8)      if (utf)
3156        {        {
3157        length = 1;        length = 1;
3158        ecode++;        ecode++;
# Line 3024  for (;;) Line 3166  for (;;)
3166        }        }
3167      else      else
3168  #endif  #endif
3169        /* Not UTF mode */
     /* Non-UTF-8 mode */  
3170        {        {
3171        if (md->end_subject - eptr < 1)        if (md->end_subject - eptr < 1)
3172          {          {
# Line 3037  for (;;) Line 3178  for (;;)
3178        }        }
3179      break;      break;
3180    
3181      /* Match a single character, caselessly. If we are at the end of the      /* Match a single character, caselessly. If we are at the end of the
3182      subject, give up immediately. */      subject, give up immediately. */
3183    
3184      case OP_CHARI:      case OP_CHARI:
3185      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
3186        {        {
3187        SCHECK_PARTIAL();        SCHECK_PARTIAL();
3188        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
3189        }        }
3190    
3191  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3192      if (utf8)      if (utf)
3193        {        {
3194        length = 1;        length = 1;
3195        ecode++;        ecode++;
3196        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
3197    
3198        /* If the pattern character's value is < 128, we have only one byte, and        /* If the pattern character's value is < 128, we have only one byte, and
3199        we know that its other case must also be one byte long, so we can use the        we know that its other case must also be one byte long, so we can use the
3200        fast lookup table. We know that there is at least one byte left in the        fast lookup table. We know that there is at least one byte left in the
3201        subject. */        subject. */
3202    
3203        if (fc < 128)        if (fc < 128)
3204          {          {
3205          if (md->lcc[*ecode++] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);          if (md->lcc[fc]
3206                != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH);
3207            ecode++;
3208            eptr++;
3209          }          }
3210    
3211        /* Otherwise we must pick up the subject character. Note that we cannot        /* Otherwise we must pick up the subject character. Note that we cannot
# Line 3087  for (;;) Line 3231  for (;;)
3231          }          }
3232        }        }
3233      else      else
3234  #endif   /* SUPPORT_UTF8 */  #endif   /* SUPPORT_UTF */
3235    
3236      /* Non-UTF-8 mode */      /* Not UTF mode */
3237        {        {
3238        if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);        if (TABLE_GET(ecode[1], md->lcc, ecode[1])
3239              != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH);
3240          eptr++;
3241        ecode += 2;        ecode += 2;
3242        }        }
3243      break;      break;
# Line 3101  for (;;) Line 3247  for (;;)
3247      case OP_EXACT:      case OP_EXACT:
3248      case OP_EXACTI:      case OP_EXACTI:
3249      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3250      ecode += 3;      ecode += 1 + IMM2_SIZE;
3251      goto REPEATCHAR;      goto REPEATCHAR;
3252    
3253      case OP_POSUPTO:      case OP_POSUPTO:
# Line 3116  for (;;) Line 3262  for (;;)
3262      min = 0;      min = 0;
3263      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3264      minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI;      minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI;
3265      ecode += 3;      ecode += 1 + IMM2_SIZE;
3266      goto REPEATCHAR;      goto REPEATCHAR;
3267    
3268      case OP_POSSTAR:      case OP_POSSTAR:
# Line 3164  for (;;) Line 3310  for (;;)
3310      /* Common code for all repeated single-character matches. */      /* Common code for all repeated single-character matches. */
3311    
3312      REPEATCHAR:      REPEATCHAR:
3313  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3314      if (utf8)      if (utf)
3315        {        {
3316        length = 1;        length = 1;
3317        charptr = ecode;        charptr = ecode;
# Line 3181  for (;;) Line 3327  for (;;)
3327          unsigned int othercase;          unsigned int othercase;
3328          if (op >= OP_STARI &&     /* Caseless */          if (op >= OP_STARI &&     /* Caseless */
3329              (othercase = UCD_OTHERCASE(fc)) != fc)              (othercase = UCD_OTHERCASE(fc)) != fc)
3330            oclength = _pcre_ord2utf8(othercase, occhars);            oclength = PRIV(ord2utf)(othercase, occhars);
3331          else oclength = 0;          else oclength = 0;
3332  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3333    
3334          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3335            {            {
3336            if (eptr <= md->end_subject - length &&            if (eptr <= md->end_subject - length &&
3337              memcmp(eptr, charptr, length) == 0) eptr += length;              memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3338  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3339            else if (oclength > 0 &&            else if (oclength > 0 &&
3340                     eptr <= md->end_subject - oclength &&                     eptr <= md->end_subject - oclength &&
3341                     memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                     memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3342  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3343            else            else
3344              {              {
# Line 3211  for (;;) Line 3357  for (;;)
3357              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3358              if (fi >= max) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3359              if (eptr <= md->end_subject - length &&              if (eptr <= md->end_subject - length &&
3360                memcmp(eptr, charptr, length) == 0) eptr += length;                memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3361  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3362              else if (oclength > 0 &&              else if (oclength > 0 &&
3363                       eptr <= md->end_subject - oclength &&                       eptr <= md->end_subject - oclength &&
3364                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                       memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3365  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3366              else              else
3367                {                {
# Line 3232  for (;;) Line 3378  for (;;)
3378            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3379              {              {
3380              if (eptr <= md->end_subject - length &&              if (eptr <= md->end_subject - length &&
3381                  memcmp(eptr, charptr, length) == 0) eptr += length;                  memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3382  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3383              else if (oclength > 0 &&              else if (oclength > 0 &&
3384                       eptr <= md->end_subject - oclength &&                       eptr <= md->end_subject - oclength &&
3385                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                       memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3386  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3387              else              else
3388                {                {
# Line 3268  for (;;) Line 3414  for (;;)
3414        value of fc will always be < 128. */        value of fc will always be < 128. */
3415        }        }
3416      else      else
3417  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
3418          /* When not in UTF-8 mode, load a single-byte character. */
3419      /* When not in UTF-8 mode, load a single-byte character. */        fc = *ecode++;
3420    
3421      fc = *ecode++;      /* The value of fc at this point is always one character, though we may
3422        or may not be in UTF mode. The code is duplicated for the caseless and
     /* The value of fc at this point is always less than 256, though we may or  
     may not be in UTF-8 mode. The code is duplicated for the caseless and  
3423      caseful cases, for speed, since matching characters is likely to be quite      caseful cases, for speed, since matching characters is likely to be quite
3424      common. First, ensure the minimum number of matches are present. If min =      common. First, ensure the minimum number of matches are present. If min =
3425      max, continue at the same level without recursing. Otherwise, if      max, continue at the same level without recursing. Otherwise, if
# Line 3284  for (;;) Line 3428  for (;;)
3428      maximizing, find the maximum number of characters and work backwards. */      maximizing, find the maximum number of characters and work backwards. */
3429    
3430      DPRINTF(("matching %c{%d,%d} against subject %.*s\n", fc, min, max,      DPRINTF(("matching %c{%d,%d} against subject %.*s\n", fc, min, max,
3431        max, eptr));        max, (char *)eptr));
3432    
3433      if (op >= OP_STARI)  /* Caseless */      if (op >= OP_STARI)  /* Caseless */
3434        {        {
3435        fc = md->lcc[fc];  #ifdef COMPILE_PCRE8
3436          /* fc must be < 128 if UTF is enabled. */
3437          foc = md->fcc[fc];
3438    #else
3439    #ifdef SUPPORT_UTF
3440    #ifdef SUPPORT_UCP
3441          if (utf && fc > 127)
3442            foc = UCD_OTHERCASE(fc);
3443    #else
3444          if (utf && fc > 127)
3445            foc = fc;
3446    #endif /* SUPPORT_UCP */
3447          else
3448    #endif /* SUPPORT_UTF */
3449            foc = TABLE_GET(fc, md->fcc, fc);
3450    #endif /* COMPILE_PCRE8 */
3451    
3452        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
3453          {          {
3454          if (eptr >= md->end_subject)          if (eptr >= md->end_subject)
# Line 3296  for (;;) Line 3456  for (;;)
3456            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3457            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3458            }            }
3459          if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);          if (fc != *eptr && foc != *eptr) RRETURN(MATCH_NOMATCH);
3460            eptr++;
3461          }          }
3462        if (min == max) continue;        if (min == max) continue;
3463        if (minimize)        if (minimize)
# Line 3311  for (;;) Line 3472  for (;;)
3472              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3473              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3474              }              }
3475            if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);            if (fc != *eptr && foc != *eptr) RRETURN(MATCH_NOMATCH);
3476              eptr++;
3477            }            }
3478          /* Control never gets here */          /* Control never gets here */
3479          }          }
# Line 3325  for (;;) Line 3487  for (;;)
3487              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3488              break;              break;
3489              }              }
3490            if (fc != md->lcc[*eptr]) break;            if (fc != *eptr && foc != *eptr) break;
3491            eptr++;            eptr++;
3492            }            }
3493    
# Line 3410  for (;;) Line 3572  for (;;)
3572        SCHECK_PARTIAL();        SCHECK_PARTIAL();
3573        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
3574        }        }
3575      ecode++;  #ifdef SUPPORT_UTF
3576      GETCHARINCTEST(c, eptr);      if (utf)
     if (op == OP_NOTI)         /* The caseless case */  
3577        {        {
3578  #ifdef SUPPORT_UTF8        register unsigned int ch, och;
3579        if (c < 256)  
3580  #endif        ecode++;
3581        c = md->lcc[c];        GETCHARINC(ch, ecode);
3582        if (md->lcc[*ecode++] == c) RRETURN(MATCH_NOMATCH);        GETCHARINC(c, eptr);
3583    
3584          if (op == OP_NOT)
3585            {
3586            if (ch == c) RRETURN(MATCH_NOMATCH);
3587            }
3588          else
3589            {
3590    #ifdef SUPPORT_UCP
3591            if (ch > 127)
3592              och = UCD_OTHERCASE(ch);
3593    #else
3594            if (ch > 127)
3595              och = ch;
3596    #endif /* SUPPORT_UCP */
3597            else
3598              och = TABLE_GET(ch, md->fcc, ch);
3599            if (ch == c || och == c) RRETURN(MATCH_NOMATCH);
3600            }
3601        }        }
3602      else    /* Caseful */      else
3603    #endif
3604        {        {
3605        if (*ecode++ == c) RRETURN(MATCH_NOMATCH);        register unsigned int ch = ecode[1];
3606          c = *eptr++;
3607          if (ch == c || (op == OP_NOTI && TABLE_GET(ch, md->fcc, ch) == c))
3608            RRETURN(MATCH_NOMATCH);
3609          ecode += 2;
3610        }        }
3611      break;      break;
3612    
# Line 3436  for (;;) Line 3620  for (;;)
3620      case OP_NOTEXACT:      case OP_NOTEXACT:
3621      case OP_NOTEXACTI:      case OP_NOTEXACTI:
3622      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3623      ecode += 3;      ecode += 1 + IMM2_SIZE;
3624      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3625    
3626      case OP_NOTUPTO:      case OP_NOTUPTO:
# Line 3446  for (;;) Line 3630  for (;;)
3630      min = 0;      min = 0;
3631      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3632      minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI;      minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI;
3633      ecode += 3;      ecode += 1 + IMM2_SIZE;
3634      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3635    
3636      case OP_NOTPOSSTAR:      case OP_NOTPOSSTAR:
# Line 3478  for (;;) Line 3662  for (;;)
3662      possessive = TRUE;      possessive = TRUE;
3663      min = 0;      min = 0;
3664      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3665      ecode += 3;      ecode += 1 + IMM2_SIZE;
3666      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3667    
3668      case OP_NOTSTAR:      case OP_NOTSTAR:
# Line 3502  for (;;) Line 3686  for (;;)
3686      /* Common code for all repeated single-byte matches. */      /* Common code for all repeated single-byte matches. */
3687    
3688      REPEATNOTCHAR:      REPEATNOTCHAR:
3689      fc = *ecode++;      GETCHARINCTEST(fc, ecode);
3690    
3691      /* The code is duplicated for the caseless and caseful cases, for speed,      /* The code is duplicated for the caseless and caseful cases, for speed,
3692      since matching characters is likely to be quite common. First, ensure the      since matching characters is likely to be quite common. First, ensure the
# Line 3513  for (;;) Line 3697  for (;;)
3697      characters and work backwards. */      characters and work backwards. */
3698    
3699      DPRINTF(("negative matching %c{%d,%d} against subject %.*s\n", fc, min, max,      DPRINTF(("negative matching %c{%d,%d} against subject %.*s\n", fc, min, max,
3700        max, eptr));        max, (char *)eptr));
3701    
3702      if (op >= OP_NOTSTARI)     /* Caseless */      if (op >= OP_NOTSTARI)     /* Caseless */
3703        {        {
3704        fc = md->lcc[fc];  #ifdef SUPPORT_UTF
3705    #ifdef SUPPORT_UCP
3706          if (utf && fc > 127)
3707            foc = UCD_OTHERCASE(fc);
3708    #else
3709          if (utf && fc > 127)
3710            foc = fc;
3711    #endif /* SUPPORT_UCP */
3712          else
3713    #endif /* SUPPORT_UTF */
3714            foc = TABLE_GET(fc, md->fcc, fc);
3715    
3716  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3717        /* UTF-8 mode */        if (utf)
       if (utf8)  
3718          {          {
3719          register unsigned int d;          register unsigned int d;
3720          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 3532  for (;;) Line 3725  for (;;)
3725              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3726              }              }
3727            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3728            if (d < 256) d = md->lcc[d];            if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH);
           if (fc == d) RRETURN(MATCH_NOMATCH);  
3729            }            }
3730          }          }
3731        else        else
3732  #endif  #endif
3733          /* Not UTF mode */
       /* Not UTF-8 mode */  
3734          {          {
3735          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3736            {            {
# Line 3548  for (;;) Line 3739  for (;;)
3739              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3740              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3741              }              }
3742            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);            if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
3743              eptr++;
3744            }            }
3745          }          }
3746    
# Line 3556  for (;;) Line 3748  for (;;)
3748    
3749        if (minimize)        if (minimize)
3750          {          {
3751  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3752          /* UTF-8 mode */          if (utf)
         if (utf8)  
3753            {            {
3754            register unsigned int d;            register unsigned int d;
3755            for (fi = min;; fi++)            for (fi = min;; fi++)
# Line 3572  for (;;) Line 3763  for (;;)
3763                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3764                }                }
3765              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3766              if (d < 256) d = md->lcc[d];              if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH);
             if (fc == d) RRETURN(MATCH_NOMATCH);  
3767              }              }
3768            }            }
3769          else          else
3770  #endif  #endif
3771          /* Not UTF-8 mode */          /* Not UTF mode */
3772            {            {
3773            for (fi = min;; fi++)            for (fi = min;; fi++)
3774              {              {
# Line 3590  for (;;) Line 3780  for (;;)
3780                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3781                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3782                }                }
3783              if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);              if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
3784                eptr++;
3785              }              }
3786            }            }
3787          /* Control never gets here */          /* Control never gets here */
# Line 3602  for (;;) Line 3793  for (;;)
3793          {          {
3794          pp = eptr;          pp = eptr;
3795    
3796  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3797          /* UTF-8 mode */          if (utf)
         if (utf8)  
3798            {            {
3799            register unsigned int d;            register unsigned int d;
3800            for (i = min; i < max; i++)            for (i = min; i < max; i++)
# Line 3616  for (;;) Line 3806  for (;;)
3806                break;                break;
3807                }                }
3808              GETCHARLEN(d, eptr, len);              GETCHARLEN(d, eptr, len);
3809              if (d < 256) d = md->lcc[d];              if (fc == d || (unsigned int)foc == d) break;
             if (fc == d) break;  
3810              eptr += len;              eptr += len;
3811              }              }
3812          if (possessive) continue;            if (possessive) continue;
3813          for(;;)            for(;;)
3814              {              {
3815              RMATCH(eptr, ecode, offset_top, md, eptrb, RM30);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM30);
3816              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
# Line 3631  for (;;) Line 3820  for (;;)
3820            }            }
3821          else          else
3822  #endif  #endif
3823          /* Not UTF-8 mode */          /* Not UTF mode */
3824            {            {
3825            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3826              {              {
# Line 3640  for (;;) Line 3829  for (;;)
3829                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3830                break;                break;
3831                }                }
3832              if (fc == md->lcc[*eptr]) break;              if (fc == *eptr || foc == *eptr) break;
3833              eptr++;              eptr++;
3834              }              }
3835            if (possessive) continue;            if (possessive) continue;
# Line 3661  for (;;) Line 3850  for (;;)
3850    
3851      else      else
3852        {        {
3853  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3854        /* UTF-8 mode */        if (utf)
       if (utf8)  
3855          {          {
3856          register unsigned int d;          register unsigned int d;
3857          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 3679  for (;;) Line 3867  for (;;)
3867          }          }
3868        else        else
3869  #endif  #endif
3870        /* Not UTF-8 mode */        /* Not UTF mode */
3871          {          {
3872          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3873            {            {
# Line 3696  for (;;) Line 3884  for (;;)
3884    
3885        if (minimize)        if (minimize)
3886          {          {
3887  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3888          /* UTF-8 mode */          if (utf)
         if (utf8)  
3889            {            {
3890            register unsigned int d;            register unsigned int d;
3891            for (fi = min;; fi++)            for (fi = min;; fi++)
# Line 3717  for (;;) Line 3904  for (;;)
3904            }            }
3905          else          else
3906  #endif  #endif
3907          /* Not UTF-8 mode */          /* Not UTF mode */
3908            {            {
3909            for (fi = min;; fi++)            for (fi = min;; fi++)
3910              {              {
# Line 3741  for (;;) Line 3928  for (;;)
3928          {          {
3929          pp = eptr;          pp = eptr;
3930    
3931  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3932          /* UTF-8 mode */          if (utf)
         if (utf8)  
3933            {            {
3934            register unsigned int d;            register unsigned int d;
3935            for (i = min; i < max; i++)            for (i = min; i < max; i++)
# Line 3769  for (;;) Line 3955  for (;;)
3955            }            }
3956          else          else
3957  #endif  #endif
3958          /* Not UTF-8 mode */          /* Not UTF mode */
3959            {            {
3960            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3961              {              {
# Line 3802  for (;;) Line 3988  for (;;)
3988      case OP_TYPEEXACT:      case OP_TYPEEXACT:
3989      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3990      minimize = TRUE;      minimize = TRUE;
3991      ecode += 3;      ecode += 1 + IMM2_SIZE;
3992      goto REPEATTYPE;      goto REPEATTYPE;
3993    
3994      case OP_TYPEUPTO:      case OP_TYPEUPTO:
# Line 3810  for (;;) Line 3996  for (;;)
3996      min = 0;      min = 0;
3997      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3998      minimize = *ecode == OP_TYPEMINUPTO;      minimize = *ecode == OP_TYPEMINUPTO;
3999      ecode += 3;      ecode += 1 + IMM2_SIZE;
4000      goto REPEATTYPE;      goto REPEATTYPE;
4001    
4002      case OP_TYPEPOSSTAR:      case OP_TYPEPOSSTAR:
# Line 3838  for (;;) Line 4024  for (;;)
4024      possessive = TRUE;      possessive = TRUE;
4025      min = 0;      min = 0;
4026      max = GET2(ecode, 1);      max = GET2(ecode, 1);
4027      ecode += 3;      ecode += 1 + IMM2_SIZE;
4028      goto REPEATTYPE;      goto REPEATTYPE;
4029    
4030      case OP_TYPESTAR:      case OP_TYPESTAR:
# Line 4045  for (;;) Line 4231  for (;;)
4231            while (eptr < md->end_subject)            while (eptr < md->end_subject)
4232              {              {
4233              int len = 1;              int len = 1;
4234              if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
4235              if (UCD_CATEGORY(c) != ucp_M) break;              if (UCD_CATEGORY(c) != ucp_M) break;
4236              eptr += len;              eptr += len;
4237              }              }
4238              CHECK_PARTIAL();
4239            }            }
4240          }          }
4241    
# Line 4057  for (;;) Line 4244  for (;;)
4244    
4245  /* Handle all other cases when the coding is UTF-8 */  /* Handle all other cases when the coding is UTF-8 */
4246    
4247  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
4248        if (utf8) switch(ctype)        if (utf) switch(ctype)
4249          {          {
4250          case OP_ANY:          case OP_ANY:
4251          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 4069  for (;;) Line 4256  for (;;)
4256              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4257              }              }
4258            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
4259              if (md->partial != 0 &&
4260                  eptr + 1 >= md->end_subject &&
4261                  NLBLOCK->nltype == NLTYPE_FIXED &&
4262                  NLBLOCK->nllen == 2 &&
4263                  *eptr == NLBLOCK->nl[0])
4264                {
4265                md->hitend = TRUE;
4266                if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
4267                }
4268            eptr++;            eptr++;
4269            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4270            }            }
4271          break;          break;
4272    
# Line 4083  for (;;) Line 4279  for (;;)
4279              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4280              }              }
4281            eptr++;            eptr++;
4282            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4283            }            }
4284          break;          break;
4285    
# Line 4265  for (;;) Line 4461  for (;;)
4461              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4462              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4463              }              }
4464            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_digit) == 0)
4465              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4466              eptr++;
4467            /* No need to skip more bytes - we know it's a 1-byte character */            /* No need to skip more bytes - we know it's a 1-byte character */
4468            }            }
4469          break;          break;
# Line 4281  for (;;) Line 4478  for (;;)
4478              }              }
4479            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)
4480              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4481            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);            eptr++;
4482              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4483            }            }
4484          break;          break;
4485    
# Line 4293  for (;;) Line 4491  for (;;)
4491              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4492              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4493              }              }
4494            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_space) == 0)
4495              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4496              eptr++;
4497            /* No need to skip more bytes - we know it's a 1-byte character */            /* No need to skip more bytes - we know it's a 1-byte character */
4498            }            }
4499          break;          break;
# Line 4309  for (;;) Line 4508  for (;;)
4508              }              }
4509            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0)            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0)
4510              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4511            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);            eptr++;
4512              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4513            }            }
4514          break;          break;
4515    
# Line 4321  for (;;) Line 4521  for (;;)
4521              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4522              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4523              }              }
4524            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_word) == 0)
4525              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4526              eptr++;
4527            /* No need to skip more bytes - we know it's a 1-byte character */            /* No need to skip more bytes - we know it's a 1-byte character */
4528            }            }
4529          break;          break;
# Line 4332  for (;;) Line 4533  for (;;)
4533          }  /* End switch(ctype) */          }  /* End switch(ctype) */
4534    
4535        else        else
4536  #endif     /* SUPPORT_UTF8 */  #endif     /* SUPPORT_UTF */
4537    
4538        /* Code for the non-UTF-8 case for minimum matching of operators other        /* Code for the non-UTF-8 case for minimum matching of operators other
4539        than OP_PROP and OP_NOTPROP. */        than OP_PROP and OP_NOTPROP. */
# Line 4348  for (;;) Line 4549  for (;;)
4549              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4550              }              }
4551            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
4552              if (md->partial != 0 &&
4553                  eptr + 1 >= md->end_subject &&
4554                  NLBLOCK->nltype == NLTYPE_FIXED &&
4555                  NLBLOCK->nllen == 2 &&
4556                  *eptr == NLBLOCK->nl[0])
4557                {
4558                md->hitend = TRUE;
4559                if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
4560                }
4561            eptr++;            eptr++;
4562            }            }
4563          break;          break;
# Line 4392  for (;;) Line 4602  for (;;)
4602              case 0x000b:              case 0x000b:
4603              case 0x000c:              case 0x000c:
4604              case 0x0085:              case 0x0085:
4605    #ifdef COMPILE_PCRE16
4606                case 0x2028:
4607                case 0x2029:
4608    #endif
4609              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4610              break;              break;
4611              }              }
# Line 4412  for (;;) Line 4626  for (;;)
4626              case 0x09:      /* HT */              case 0x09:      /* HT */
4627              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4628              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
4629    #ifdef COMPILE_PCRE16
4630                case 0x1680:    /* OGHAM SPACE MARK */
4631                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
4632                case 0x2000:    /* EN QUAD */
4633                case 0x2001:    /* EM QUAD */
4634                case 0x2002:    /* EN SPACE */
4635                case 0x2003:    /* EM SPACE */
4636                case 0x2004:    /* THREE-PER-EM SPACE */
4637                case 0x2005:    /* FOUR-PER-EM SPACE */
4638                case 0x2006:    /* SIX-PER-EM SPACE */
4639                case 0x2007:    /* FIGURE SPACE */
4640                case 0x2008:    /* PUNCTUATION SPACE */
4641                case 0x2009:    /* THIN SPACE */
4642                case 0x200A:    /* HAIR SPACE */
4643                case 0x202f:    /* NARROW NO-BREAK SPACE */
4644                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4645                case 0x3000:    /* IDEOGRAPHIC SPACE */
4646    #endif
4647              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4648              }              }
4649            }            }
# Line 4431  for (;;) Line 4663  for (;;)
4663              case 0x09:      /* HT */              case 0x09:      /* HT */
4664              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4665              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
4666    #ifdef COMPILE_PCRE16
4667                case 0x1680:    /* OGHAM SPACE MARK */
4668                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
4669                case 0x2000:    /* EN QUAD */
4670                case 0x2001:    /* EM QUAD */
4671                case 0x2002:    /* EN SPACE */
4672                case 0x2003:    /* EM SPACE */
4673                case 0x2004:    /* THREE-PER-EM SPACE */
4674                case 0x2005:    /* FOUR-PER-EM SPACE */
4675                case 0x2006:    /* SIX-PER-EM SPACE */
4676                case 0x2007:    /* FIGURE SPACE */
4677                case 0x2008:    /* PUNCTUATION SPACE */
4678                case 0x2009:    /* THIN SPACE */
4679                case 0x200A:    /* HAIR SPACE */
4680                case 0x202f:    /* NARROW NO-BREAK SPACE */
4681                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4682                case 0x3000:    /* IDEOGRAPHIC SPACE */
4683    #endif
4684              break;              break;
4685              }              }
4686            }            }
# Line 4452  for (;;) Line 4702  for (;;)
4702              case 0x0c:      /* FF */              case 0x0c:      /* FF */
4703              case 0x0d:      /* CR */              case 0x0d:      /* CR */
4704              case 0x85:      /* NEL */              case 0x85:      /* NEL */
4705    #ifdef COMPILE_PCRE16
4706                case 0x2028:    /* LINE SEPARATOR */
4707                case 0x2029:    /* PARAGRAPH SEPARATOR */
4708    #endif
4709              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4710              }              }
4711            }            }
# Line 4473  for (;;) Line 4727  for (;;)
4727              case 0x0c:      /* FF */              case 0x0c:      /* FF */
4728              case 0x0d:      /* CR */              case 0x0d:      /* CR */
4729              case 0x85:      /* NEL */              case 0x85:      /* NEL */
4730    #ifdef COMPILE_PCRE16
4731                case 0x2028:    /* LINE SEPARATOR */
4732                case 0x2029:    /* PARAGRAPH SEPARATOR */
4733    #endif
4734              break;              break;
4735              }              }
4736            }            }
# Line 4486  for (;;) Line 4744  for (;;)
4744              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4745              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4746              }              }
4747            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_digit) != 0)
4748                RRETURN(MATCH_NOMATCH);
4749              eptr++;
4750            }            }
4751          break;          break;
4752    
# Line 4498  for (;;) Line 4758  for (;;)
4758              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4759              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4760              }              }
4761            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_digit) == 0)
4762                RRETURN(MATCH_NOMATCH);
4763              eptr++;
4764            }            }
4765          break;          break;
4766    
# Line 4510  for (;;) Line 4772  for (;;)
4772              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4773              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4774              }              }
4775            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_space) != 0)
4776                RRETURN(MATCH_NOMATCH);
4777              eptr++;
4778            }            }
4779          break;          break;
4780    
# Line 4522  for (;;) Line 4786  for (;;)
4786              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4787              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4788              }              }
4789            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_space) == 0)
4790                RRETURN(MATCH_NOMATCH);
4791              eptr++;
4792            }            }
4793          break;          break;
4794    
# Line 4534  for (;;) Line 4800  for (;;)
4800              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4801              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4802              }              }
4803            if ((md->ctypes[*eptr++] & ctype_word) != 0)            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_word) != 0)
4804              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4805              eptr++;
4806            }            }
4807          break;          break;
4808    
# Line 4547  for (;;) Line 4814  for (;;)
4814              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4815              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4816              }              }
4817            if ((md->ctypes[*eptr++] & ctype_word) == 0)            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_word) == 0)
4818              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4819              eptr++;
4820            }            }
4821          break;          break;
4822    
# Line 4766  for (;;) Line 5034  for (;;)
5034            while (eptr < md->end_subject)            while (eptr < md->end_subject)
5035              {              {
5036              int len = 1;              int len = 1;
5037              if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
5038              if (UCD_CATEGORY(c) != ucp_M) break;              if (UCD_CATEGORY(c) != ucp_M) break;
5039              eptr += len;              eptr += len;
5040              }              }
5041              CHECK_PARTIAL();
5042            }            }
5043          }          }
5044        else        else
5045  #endif     /* SUPPORT_UCP */  #endif     /* SUPPORT_UCP */
5046    
5047  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
5048        /* UTF-8 mode */        if (utf)
       if (utf8)  
5049          {          {
5050          for (fi = min;; fi++)          for (fi = min;; fi++)
5051            {            {
# Line 4794  for (;;) Line 5062  for (;;)
5062            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
5063            switch(ctype)            switch(ctype)
5064              {              {
5065              case OP_ANY:        /* This is the non-NL case */              case OP_ANY:               /* This is the non-NL case */
5066                if (md->partial != 0 &&    /* Take care with CRLF partial */
5067                    eptr >= md->end_subject &&
5068                    NLBLOCK->nltype == NLTYPE_FIXED &&
5069                    NLBLOCK->nllen == 2 &&
5070                    c == NLBLOCK->nl[0])
5071                  {
5072                  md->hitend = TRUE;
5073                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5074                  }
5075                break;
5076    
5077              case OP_ALLANY:              case OP_ALLANY:
5078              case OP_ANYBYTE:              case OP_ANYBYTE:
5079              break;              break;
# Line 4919  for (;;) Line 5198  for (;;)
5198              break;              break;
5199    
5200              case OP_WHITESPACE:              case OP_WHITESPACE:
5201              if  (c >= 256 || (md->ctypes[c] & ctype_space) == 0)              if (c >= 256 || (md->ctypes[c] & ctype_space) == 0)
5202                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
5203              break;              break;
5204    
# Line 4940  for (;;) Line 5219  for (;;)
5219          }          }
5220        else        else
5221  #endif  #endif
5222        /* Not UTF-8 mode */        /* Not UTF mode */
5223          {          {
5224          for (fi = min;; fi++)          for (fi = min;; fi++)
5225            {            {
# Line 4957  for (;;) Line 5236  for (;;)
5236            c = *eptr++;            c = *eptr++;
5237            switch(ctype)            switch(ctype)
5238              {              {
5239              case OP_ANY:     /* This is the non-NL case */              case OP_ANY:               /* This is the non-NL case */
5240                if (md->partial != 0 &&    /* Take care with CRLF partial */
5241                    eptr >= md->end_subject &&
5242                    NLBLOCK->nltype == NLTYPE_FIXED &&
5243                    NLBLOCK->nllen == 2 &&
5244                    c == NLBLOCK->nl[0])
5245                  {
5246                  md->hitend = TRUE;
5247                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5248                  }
5249                break;
5250    
5251              case OP_ALLANY:              case OP_ALLANY:
5252              case OP_ANYBYTE:              case OP_ANYBYTE:
5253              break;              break;
# Line 4976  for (;;) Line 5266  for (;;)
5266                case 0x000b:                case 0x000b:
5267                case 0x000c:                case 0x000c:
5268                case 0x0085:                case 0x0085:
5269    #ifdef COMPILE_PCRE16
5270                  case 0x2028:
5271                  case 0x2029:
5272    #endif
5273                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
5274                break;                break;
5275                }                }
# Line 4988  for (;;) Line 5282  for (;;)
5282                case 0x09:      /* HT */                case 0x09:      /* HT */
5283                case 0x20:      /* SPACE */                case 0x20:      /* SPACE */
5284                case 0xa0:      /* NBSP */                case 0xa0:      /* NBSP */
5285    #ifdef COMPILE_PCRE16
5286                  case 0x1680:    /* OGHAM SPACE MARK */
5287                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
5288                  case 0x2000:    /* EN QUAD */
5289                  case 0x2001:    /* EM QUAD */
5290                  case 0x2002:    /* EN SPACE */
5291                  case 0x2003:    /* EM SPACE */
5292                  case 0x2004:    /* THREE-PER-EM SPACE */
5293                  case 0x2005:    /* FOUR-PER-EM SPACE */
5294                  case 0x2006:    /* SIX-PER-EM SPACE */
5295                  case 0x2007:    /* FIGURE SPACE */
5296                  case 0x2008:    /* PUNCTUATION SPACE */
5297                  case 0x2009:    /* THIN SPACE */
5298                  case 0x200A:    /* HAIR SPACE */
5299                  case 0x202f:    /* NARROW NO-BREAK SPACE */
5300                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
5301                  case 0x3000:    /* IDEOGRAPHIC SPACE */
5302    #endif
5303                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
5304                }                }
5305              break;              break;
# Line 4999  for (;;) Line 5311  for (;;)
5311                case 0x09:      /* HT */                case 0x09:      /* HT */
5312                case 0x20:      /* SPACE */                case 0x20:      /* SPACE */
5313                case 0xa0:      /* NBSP */                case 0xa0:      /* NBSP */
5314    #ifdef COMPILE_PCRE16
5315                  case 0x1680:    /* OGHAM SPACE MARK */
5316                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
5317                  case 0x2000:    /* EN QUAD */
5318                  case 0x2001:    /* EM QUAD */
5319                  case 0x2002:    /* EN SPACE */
5320                  case 0x2003:    /* EM SPACE */
5321                  case 0x2004:    /* THREE-PER-EM SPACE */
5322                  case 0x2005:    /* FOUR-PER-EM SPACE */
5323                  case 0x2006:    /* SIX-PER-EM SPACE */
5324                  case 0x2007:    /* FIGURE SPACE */
5325                  case 0x2008:    /* PUNCTUATION SPACE */
5326                  case 0x2009:    /* THIN SPACE */
5327                  case 0x200A:    /* HAIR SPACE */
5328                  case 0x202f:    /* NARROW NO-BREAK SPACE */
5329                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
5330                  case 0x3000:    /* IDEOGRAPHIC SPACE */
5331    #endif
5332                break;                break;
5333                }                }
5334              break;              break;
# Line 5012  for (;;) Line 5342  for (;;)
5342                case 0x0c:      /* FF */                case 0x0c:      /* FF */
5343                case 0x0d:      /* CR */                case 0x0d:      /* CR */
5344                case 0x85:      /* NEL */                case 0x85:      /* NEL */
5345    #ifdef COMPILE_PCRE16
5346                  case 0x2028:    /* LINE SEPARATOR */
5347                  case 0x2029:    /* PARAGRAPH SEPARATOR */
5348    #endif
5349                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
5350                }                }
5351              break;              break;
# Line 5025  for (;;) Line 5359  for (;;)
5359                case 0x0c:      /* FF */                case 0x0c:      /* FF */
5360                case 0x0d:      /* CR */                case 0x0d:      /* CR */
5361                case 0x85:      /* NEL */                case 0x85:      /* NEL */
5362    #ifdef COMPILE_PCRE16
5363                  case 0x2028:    /* LINE SEPARATOR */
5364                  case 0x2029:    /* PARAGRAPH SEPARATOR */
5365    #endif
5366                break;                break;
5367                }                }
5368              break;              break;
5369    
5370              case OP_NOT_DIGIT:              case OP_NOT_DIGIT:
5371              if ((md->ctypes[c] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);              if (MAX_255(c) && (md->ctypes[c] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);
5372              break;              break;
5373    
5374              case OP_DIGIT:              case OP_DIGIT:
5375              if ((md->ctypes[c] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);              if (!MAX_255(c) || (md->ctypes[c] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);
5376              break;              break;
5377    
5378              case OP_NOT_WHITESPACE:              case OP_NOT_WHITESPACE:
5379              if ((md->ctypes[c] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);              if (MAX_255(c) && (md->ctypes[c] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);
5380              break;              break;
5381    
5382              case OP_WHITESPACE:              case OP_WHITESPACE:
5383              if  ((md->ctypes[c] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);              if (!MAX_255(c) || (md->ctypes[c] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);
5384              break;              break;
5385    
5386              case OP_NOT_WORDCHAR:              case OP_NOT_WORDCHAR:
5387              if ((md->ctypes[c] & ctype_word) != 0) RRETURN(MATCH_NOMATCH);              if (MAX_255(c) && (md->ctypes[c] & ctype_word) != 0) RRETURN(MATCH_NOMATCH);
5388              break;              break;
5389    
5390              case OP_WORDCHAR:              case OP_WORDCHAR:
5391              if ((md->ctypes[c] & ctype_word) == 0) RRETURN(MATCH_NOMATCH);              if (!MAX_255(c) || (md->ctypes[c] & ctype_word) == 0) RRETURN(MATCH_NOMATCH);
5392              break;              break;
5393    
5394              default:              default:
# Line 5239  for (;;) Line 5577  for (;;)
5577            RMATCH(eptr, ecode, offset_top, md, eptrb, RM44);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM44);
5578            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
5579            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
5580            if (utf8) BACKCHAR(eptr);            if (utf) BACKCHAR(eptr);
5581            }            }
5582          }          }
5583    
# Line 5256  for (;;) Line 5594  for (;;)
5594              SCHECK_PARTIAL();              SCHECK_PARTIAL();
5595              break;              break;
5596              }              }
5597            if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }            if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
5598            if (UCD_CATEGORY(c) == ucp_M) break;            if (UCD_CATEGORY(c) == ucp_M) break;
5599            eptr += len;            eptr += len;
5600            while (eptr < md->end_subject)            while (eptr < md->end_subject)
5601              {              {
5602              len = 1;              len = 1;
5603              if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
5604              if (UCD_CATEGORY(c) != ucp_M) break;              if (UCD_CATEGORY(c) != ucp_M) break;
5605              eptr += len;              eptr += len;
5606              }              }
5607              CHECK_PARTIAL();
5608            }            }
5609    
5610          /* eptr is now past the end of the maximum run */          /* eptr is now past the end of the maximum run */
# Line 5279  for (;;) Line 5618  for (;;)
5618            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
5619            for (;;)                        /* Move back over one extended */            for (;;)                        /* Move back over one extended */
5620              {              {
5621              if (!utf8) c = *eptr; else              if (!utf) c = *eptr; else
5622                {                {
5623                BACKCHAR(eptr);                BACKCHAR(eptr);
5624                GETCHAR(c, eptr);                GETCHAR(c, eptr);
# Line 5293  for (;;) Line 5632  for (;;)
5632        else        else
5633  #endif   /* SUPPORT_UCP */  #endif   /* SUPPORT_UCP */
5634    
5635  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
5636        /* UTF-8 mode */        if (utf)
   
       if (utf8)  
5637          {          {
5638          switch(ctype)          switch(ctype)
5639            {            {
# Line 5311  for (;;) Line 5648  for (;;)
5648                  break;                  break;
5649                  }                  }
5650                if (IS_NEWLINE(eptr)) break;                if (IS_NEWLINE(eptr)) break;
5651                  if (md->partial != 0 &&    /* Take care with CRLF partial */
5652                      eptr + 1 >= md->end_subject &&
5653                      NLBLOCK->nltype == NLTYPE_FIXED &&
5654                      NLBLOCK->nllen == 2 &&
5655                      *eptr == NLBLOCK->nl[0])
5656                    {
5657                    md->hitend = TRUE;
5658                    if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5659                    }
5660                eptr++;                eptr++;
5661                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;                ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
5662                }                }
5663              }              }
5664    
# Line 5328  for (;;) Line 5674  for (;;)
5674                  break;                  break;
5675                  }                  }
5676                if (IS_NEWLINE(eptr)) break;                if (IS_NEWLINE(eptr)) break;
5677                  if (md->partial != 0 &&    /* Take care with CRLF partial */
5678                      eptr + 1 >= md->end_subject &&
5679                      NLBLOCK->nltype == NLTYPE_FIXED &&
5680                      NLBLOCK->nllen == 2 &&
5681                      *eptr == NLBLOCK->nl[0])
5682                    {
5683                    md->hitend = TRUE;
5684                    if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5685                    }
5686                eptr++;                eptr++;
5687                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;                ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
5688                }                }
5689              }              }
5690            break;            break;
# Line 5345  for (;;) Line 5700  for (;;)
5700                  break;                  break;
5701                  }                  }
5702                eptr++;                eptr++;
5703                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;                ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
5704                }                }
5705              }              }
5706            else            else
# Line 5578  for (;;) Line 5933  for (;;)
5933            }            }
5934          }          }
5935        else        else
5936  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
5937          /* Not UTF mode */
       /* Not UTF-8 mode */  
5938          {          {
5939          switch(ctype)          switch(ctype)
5940            {            {
# Line 5593  for (;;) Line 5947  for (;;)
5947                break;                break;
5948                }                }
5949              if (IS_NEWLINE(eptr)) break;              if (IS_NEWLINE(eptr)) break;
5950                if (md->partial != 0 &&    /* Take care with CRLF partial */
5951                    eptr + 1 >= md->end_subject &&
5952                    NLBLOCK->nltype == NLTYPE_FIXED &&
5953                    NLBLOCK->nllen == 2 &&
5954                    *eptr == NLBLOCK->nl[0])
5955                  {
5956                  md->hitend = TRUE;
5957                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5958                  }
5959              eptr++;              eptr++;
5960              }              }
5961            break;            break;
# Line 5624  for (;;) Line 5987  for (;;)
5987                }                }
5988              else              else
5989                {                {
5990                if (c != 0x000a &&                if (c != 0x000a && (md->bsr_anycrlf ||
5991                    (md->bsr_anycrlf ||                  (c != 0x000b && c != 0x000c && c != 0x0085
5992                      (c != 0x000b && c != 0x000c && c != 0x0085)))  #ifdef COMPILE_PCRE16
5993                  break;                  && c != 0x2028 && c != 0x2029
5994    #endif
5995                    ))) break;
5996                eptr++;                eptr++;
5997                }                }
5998              }              }
# Line 5642  for (;;) Line 6007  for (;;)
6007                break;                break;
6008                }                }
6009              c = *eptr;              c = *eptr;
6010              if (c == 0x09 || c == 0x20 || c == 0xa0) break;              if (c == 0x09 || c == 0x20 || c == 0xa0
6011    #ifdef COMPILE_PCRE16
6012                  || c == 0x1680 || c == 0x180e || (c >= 0x2000 && c <= 0x200A)
6013                  || c == 0x202f || c == 0x205f || c == 0x3000
6014    #endif
6015                  ) break;
6016              eptr++;              eptr++;
6017              }              }
6018            break;            break;
# Line 5656  for (;;) Line 6026  for (;;)
6026                break;                break;
6027                }                }
6028              c = *eptr;              c = *eptr;
6029              if (c != 0x09 && c != 0x20 && c != 0xa0) break;              if (c != 0x09 && c != 0x20 && c != 0xa0
6030    #ifdef COMPILE_PCRE16
6031                  && c != 0x1680 && c != 0x180e && (c < 0x2000 || c > 0x200A)
6032                  && c != 0x202f && c != 0x205f && c != 0x3000
6033    #endif
6034                  ) break;
6035              eptr++;              eptr++;
6036              }              }
6037            break;            break;
# Line 5670  for (;;) Line 6045  for (;;)
6045                break;                break;
6046                }                }
6047              c = *eptr;              c = *eptr;
6048              if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)              if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85
6049                break;  #ifdef COMPILE_PCRE16
6050                  || c == 0x2028 || c == 0x2029
6051    #endif
6052                  ) break;
6053              eptr++;              eptr++;
6054              }              }
6055            break;            break;
# Line 5685  for (;;) Line 6063  for (;;)
6063                break;                break;
6064                }                }
6065              c = *eptr;              c = *eptr;
6066              if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85)              if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85
6067                break;  #ifdef COMPILE_PCRE16
6068                  && c != 0x2028 && c != 0x2029
6069    #endif
6070                  ) break;
6071              eptr++;              eptr++;
6072              }              }
6073            break;            break;
# Line 5699  for (;;) Line 6080  for (;;)
6080                SCHECK_PARTIAL();                SCHECK_PARTIAL();
6081                break;                break;
6082                }                }
6083              if ((md->ctypes[*eptr] & ctype_digit) != 0) break;              if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_digit) != 0) break;
6084              eptr++;              eptr++;
6085              }              }
6086            break;            break;
# Line 5712  for (;;) Line 6093  for (;;)
6093                SCHECK_PARTIAL();                SCHECK_PARTIAL();
6094                break;                break;
6095                }                }
6096              if ((md->ctypes[*eptr] & ctype_digit) == 0) break;              if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_digit) == 0) break;
6097              eptr++;              eptr++;
6098              }              }
6099            break;            break;
# Line 5725  for (;;) Line 6106  for (;;)
6106                SCHECK_PARTIAL();                SCHECK_PARTIAL();
6107                break;                break;
6108                }                }
6109              if ((md->ctypes[*eptr] & ctype_space) != 0) break;              if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_space) != 0) break;
6110              eptr++;              eptr++;
6111              }              }
6112            break;            break;
# Line 5738  for (;;) Line 6119  for (;;)
6119                SCHECK_PARTIAL();                SCHECK_PARTIAL();
6120                break;                break;
6121                }                }
6122              if ((md->ctypes[*eptr] & ctype_space) == 0) break;              if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_space) == 0) break;
6123              eptr++;              eptr++;
6124              }              }
6125            break;            break;
# Line 5751  for (;;) Line 6132  for (;;)
6132                SCHECK_PARTIAL();                SCHECK_PARTIAL();
6133                break;                break;
6134                }                }
6135              if ((md->ctypes[*eptr] & ctype_word) != 0) break;              if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_word) != 0) break;
6136              eptr++;              eptr++;
6137              }              }
6138            break;            break;
# Line 5764  for (;;) Line 6145  for (;;)
6145                SCHECK_PARTIAL();                SCHECK_PARTIAL();
6146                break;                break;
6147                }                }
6148              if ((md->ctypes[*eptr] & ctype_word) == 0) break;              if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_word) == 0) break;
6149              eptr++;              eptr++;
6150              }              }
6151            break;            break;
# Line 5827  switch (frame->Xwhere) Line 6208  switch (frame->Xwhere)
6208    LBL(35) LBL(43) LBL(47) LBL(48) LBL(49) LBL(50) LBL(51) LBL(52)    LBL(35) LBL(43) LBL(47) LBL(48) LBL(49) LBL(50) LBL(51) LBL(52)
6209    LBL(53) LBL(54) LBL(55) LBL(56) LBL(57) LBL(58) LBL(63) LBL(64)    LBL(53) LBL(54) LBL(55) LBL(56) LBL(57) LBL(58) LBL(63) LBL(64)
6210    LBL(65) LBL(66)    LBL(65) LBL(66)
6211  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
6212    LBL(16) LBL(18) LBL(20) LBL(21) LBL(22) LBL(23) LBL(28) LBL(30)    LBL(21)
6213    #endif
6214    #ifdef SUPPORT_UTF
6215      LBL(16) LBL(18) LBL(20)
6216      LBL(22) LBL(23) LBL(28) LBL(30)
6217    LBL(32) LBL(34) LBL(42) LBL(46)    LBL(32) LBL(34) LBL(42) LBL(46)
6218  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
6219    LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)    LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
6220    LBL(59) LBL(60) LBL(61) LBL(62)    LBL(59) LBL(60) LBL(61) LBL(62)
6221  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
6222  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
6223    default:    default:
6224    DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));    DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));
6225    
6226    printf("+++jump error in pcre match: label %d non-existent\n", frame->Xwhere);
6227    
6228    return PCRE_ERROR_INTERNAL;    return PCRE_ERROR_INTERNAL;
6229    }    }
6230  #undef LBL  #undef LBL
# Line 5898  Undefine all the macros that were define Line 6286  Undefine all the macros that were define
6286  ***************************************************************************/  ***************************************************************************/
6287    
6288    
6289    #ifdef NO_RECURSE
6290    /*************************************************
6291    *          Release allocated heap frames         *
6292    *************************************************/
6293    
6294    /* This function releases all the allocated frames. The base frame is on the
6295    machine stack, and so must not be freed.
6296    
6297    Argument: the address of the base frame
6298    Returns:  nothing
6299    */
6300    
6301    static void
6302    release_match_heapframes (heapframe *frame_base)
6303    {
6304    heapframe *nextframe = frame_base->Xnextframe;
6305    while (nextframe != NULL)
6306      {
6307      heapframe *oldframe = nextframe;
6308      nextframe = nextframe->Xnextframe;
6309      (PUBL(stack_free))(oldframe);
6310      }
6311    }
6312    #endif
6313    
6314    
6315  /*************************************************  /*************************************************
6316  *         Execute a Regular Expression           *  *         Execute a Regular Expression           *
# Line 5923  Returns:          > 0 => success; value Line 6336  Returns:          > 0 => success; value
6336                   < -1 => some kind of unexpected problem                   < -1 => some kind of unexpected problem
6337  */  */
6338    
6339    #ifdef COMPILE_PCRE8
6340  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
6341  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,
6342    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,
6343    int offsetcount)    int offsetcount)
6344    #else
6345    PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
6346    pcre16_exec(const pcre16 *argument_re, const pcre16_extra *extra_data,
6347      PCRE_SPTR16 subject, int length, int start_offset, int options, int *offsets,
6348      int offsetcount)
6349    #endif
6350  {  {
6351  int rc, ocount, arg_offset_max;  int rc, ocount, arg_offset_max;
 int first_byte = -1;  
 int req_byte = -1;  
 int req_byte2 = -1;  
6352  int newline;  int newline;
6353  BOOL using_temporary_offsets = FALSE;  BOOL using_temporary_offsets = FALSE;
6354  BOOL anchored;  BOOL anchored;
6355  BOOL startline;  BOOL startline;
6356  BOOL firstline;  BOOL firstline;
6357  BOOL first_byte_caseless = FALSE;  BOOL utf;
6358  BOOL req_byte_caseless = FALSE;  BOOL has_first_char = FALSE;
6359  BOOL utf8;  BOOL has_req_char = FALSE;
6360    pcre_uchar first_char = 0;
6361    pcre_uchar first_char2 = 0;
6362    pcre_uchar req_char = 0;
6363    pcre_uchar req_char2 = 0;
6364  match_data match_block;  match_data match_block;
6365  match_data *md = &match_block;  match_data *md = &match_block;
6366  const uschar *tables;  const pcre_uint8 *tables;
6367  const uschar *start_bits = NULL;  const pcre_uint8 *start_bits = NULL;
6368  USPTR start_match = (USPTR)subject + start_offset;  PCRE_PUCHAR start_match = (PCRE_PUCHAR)subject + start_offset;
6369  USPTR end_subject;  PCRE_PUCHAR end_subject;
6370  USPTR start_partial = NULL;  PCRE_PUCHAR start_partial = NULL;
6371  USPTR req_byte_ptr = start_match - 1;  PCRE_PUCHAR req_char_ptr = start_match - 1;
6372    
 pcre_study_data internal_study;  
6373  const pcre_study_data *study;  const pcre_study_data *study;
6374    const REAL_PCRE *re = (const REAL_PCRE *)argument_re;
6375    
6376  real_pcre internal_re;  #ifdef NO_RECURSE
6377  const real_pcre *external_re = (const real_pcre *)argument_re;  heapframe frame_zero;
6378  const real_pcre *re = external_re;  frame_zero.Xprevframe = NULL;            /* Marks the top level */
6379    frame_zero.Xnextframe = NULL;            /* None are allocated yet */
6380    md->match_frames_base = &frame_zero;
6381    #endif
6382    
6383    /* Check for the special magic call that measures the size of the stack used
6384    per recursive call of match(). Without the funny casting for sizeof, a Windows
6385    compiler gave this error: "unary minus operator applied to unsigned type,
6386    result still unsigned". Hopefully the cast fixes that. */
6387    
6388    if (re == NULL && extra_data == NULL && subject == NULL && length == -999 &&
6389        start_offset == -999)
6390    #ifdef NO_RECURSE
6391      return -((int)sizeof(heapframe));
6392    #else
6393      return match(NULL, NULL, NULL, 0, NULL, NULL, 0);
6394    #endif
6395    
6396  /* Plausibility checks */  /* Plausibility checks */
6397    
6398  if ((options & ~PUBLIC_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION;  if ((options & ~PUBLIC_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION;
6399  if (re == NULL || subject == NULL ||  if (re == NULL || subject == NULL || (offsets == NULL && offsetcount > 0))
6400     (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL;    return PCRE_ERROR_NULL;
6401  if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;  if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;
6402  if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET;  if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET;
6403    
6404    /* Check that the first field in the block is the magic number. If it is not,
6405    return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to
6406    REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which
6407    means that the pattern is likely compiled with different endianness. */
6408    
6409    if (re->magic_number != MAGIC_NUMBER)
6410      return re->magic_number == REVERSED_MAGIC_NUMBER?
6411        PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC;
6412    if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
6413    
6414  /* These two settings are used in the code for checking a UTF-8 string that  /* These two settings are used in the code for checking a UTF-8 string that
6415  follows immediately afterwards. Other values in the md block are used only  follows immediately afterwards. Other values in the md block are used only
6416  during "normal" pcre_exec() processing, not when the JIT support is in use,  during "normal" pcre_exec() processing, not when the JIT support is in use,
6417  so they are set up later. */  so they are set up later. */
6418    
6419  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;  /* PCRE_UTF16 has the same value as PCRE_UTF8. */
6420    utf = md->utf = (re->options & PCRE_UTF8) != 0;
6421  md->partial = ((options & PCRE_PARTIAL_HARD) != 0)? 2 :  md->partial = ((options & PCRE_PARTIAL_HARD) != 0)? 2 :
6422                ((options & PCRE_PARTIAL_SOFT) != 0)? 1 : 0;                ((options & PCRE_PARTIAL_SOFT) != 0)? 1 : 0;
6423    
6424  /* Check a UTF-8 string if required. Pass back the character offset and error  /* Check a UTF-8 string if required. Pass back the character offset and error
6425  code for an invalid string if a results vector is available. */  code for an invalid string if a results vector is available. */
6426    
6427  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
6428  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)  if (utf && (options & PCRE_NO_UTF8_CHECK) == 0)
6429    {    {
6430    int erroroffset;    int erroroffset;
6431    int errorcode = _pcre_valid_utf8((USPTR)subject, length, &erroroffset);    int errorcode = PRIV(valid_utf)((PCRE_PUCHAR)subject, length, &erroroffset);
6432    if (errorcode != 0)    if (errorcode != 0)
6433      {      {
6434      if (offsetcount >= 2)      if (offsetcount >= 2)
# Line 5988  if (utf8 && (options & PCRE_NO_UTF8_CHEC Line 6436  if (utf8 && (options & PCRE_NO_UTF8_CHEC
6436        offsets[0] = erroroffset;        offsets[0] = erroroffset;
6437        offsets[1] = errorcode;        offsets[1] = errorcode;
6438        }        }
6439    #ifdef COMPILE_PCRE16
6440        return (errorcode <= PCRE_UTF16_ERR1 && md->partial > 1)?
6441          PCRE_ERROR_SHORTUTF16 : PCRE_ERROR_BADUTF16;
6442    #else
6443      return (errorcode <= PCRE_UTF8_ERR5 && md->partial > 1)?      return (errorcode <= PCRE_UTF8_ERR5 && md->partial > 1)?
6444        PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8;        PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8;
6445    #endif
6446      }      }
6447    
6448    /* Check that a start_offset points to the start of a UTF-8 character. */    /* Check that a start_offset points to the start of a UTF character. */
6449    if (start_offset > 0 && start_offset < length &&    if (start_offset > 0 && start_offset < length &&
6450        (((USPTR)subject)[start_offset] & 0xc0) == 0x80)        NOT_FIRSTCHAR(((PCRE_PUCHAR)subject)[start_offset]))
6451      return PCRE_ERROR_BADUTF8_OFFSET;      return PCRE_ERROR_BADUTF8_OFFSET;
6452    }    }
6453  #endif  #endif
# Line 6002  if (utf8 && (options & PCRE_NO_UTF8_CHEC Line 6455  if (utf8 && (options & PCRE_NO_UTF8_CHEC
6455  /* If the pattern was successfully studied with JIT support, run the JIT  /* If the pattern was successfully studied with JIT support, run the JIT
6456  executable instead of the rest of this function. Most options must be set at  executable instead of the rest of this function. Most options must be set at
6457  compile time for the JIT code to be usable. Fallback to the normal code path if  compile time for the JIT code to be usable. Fallback to the normal code path if
6458  an unsupported flag is set. In particular, JIT does not support partial  an unsupported flag is set. */
 matching. */  
6459    
6460  #ifdef SUPPORT_JIT  #ifdef SUPPORT_JIT
6461  if (extra_data != NULL  if (extra_data != NULL
6462      && (extra_data->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0      && (extra_data->flags & (PCRE_EXTRA_EXECUTABLE_JIT |
6463                                 PCRE_EXTRA_TABLES)) == PCRE_EXTRA_EXECUTABLE_JIT
6464      && extra_data->executable_jit != NULL      && extra_data->executable_jit != NULL
     && (extra_data->flags & PCRE_EXTRA_TABLES) == 0  
6465      && (options & ~(PCRE_NO_UTF8_CHECK | PCRE_NOTBOL | PCRE_NOTEOL |      && (options & ~(PCRE_NO_UTF8_CHECK | PCRE_NOTBOL | PCRE_NOTEOL |
6466                      PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART)) == 0)                      PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART |
6467    return _pcre_jit_exec(re, extra_data->executable_jit, subject, length,                      PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD)) == 0)
6468      start_offset, options, ((extra_data->flags & PCRE_EXTRA_MATCH_LIMIT) == 0)    {
6469      ? MATCH_LIMIT : extra_data->match_limit, offsets, offsetcount);    rc = PRIV(jit_exec)(re, extra_data, (const pcre_uchar *)subject, length,
6470           start_offset, options, offsets, offsetcount);
6471    
6472      /* PCRE_ERROR_NULL means that the selected normal or partial matching
6473      mode is not compiled. In this case we simply fallback to interpreter. */
6474    
6475      if (rc != PCRE_ERROR_NULL) return rc;
6476      }
6477  #endif  #endif
6478    
6479  /* Carry on with non-JIT matching. This information is for finding all the  /* Carry on with non-JIT matching. This information is for finding all the
6480  numbers associated with a given name, for condition testing. */  numbers associated with a given name, for condition testing. */
6481    
6482  md->name_table = (uschar *)re + re->name_table_offset;  md->name_table = (pcre_uchar *)re + re->name_table_offset;
6483  md->name_count = re->name_count;  md->name_count = re->name_count;
6484  md->name_entry_size = re->name_entry_size;  md->name_entry_size = re->name_entry_size;
6485    
# Line 6034  md->callout_data = NULL; Line 6493  md->callout_data = NULL;
6493    
6494  /* The table pointer is always in native byte order. */  /* The table pointer is always in native byte order. */
6495    
6496  tables = external_re->tables;  tables = re->tables;
6497    
6498  if (extra_data != NULL)  if (extra_data != NULL)
6499    {    {
# Line 6054  if (extra_data != NULL) Line 6513  if (extra_data != NULL)
6513  is a feature that makes it possible to save compiled regex and re-use them  is a feature that makes it possible to save compiled regex and re-use them
6514  in other programs later. */  in other programs later. */
6515    
6516  if (tables == NULL) tables = _pcre_default_tables;  if (tables == NULL) tables = PRIV(default_tables);
   
 /* Check that the first field in the block is the magic number. If it is not,  
 test for a regex that was compiled on a host of opposite endianness. If this is  
 the case, flipped values are put in internal_re and internal_study if there was  
 study data too. */  
   
 if (re->magic_number != MAGIC_NUMBER)  
   {  
   re = _pcre_try_flipped(re, &internal_re, study, &internal_study);  
   if (re == NULL) return PCRE_ERROR_BADMAGIC;  
   if (study != NULL) study = &internal_study;  
   }  
6517    
6518  /* Set up other data */  /* Set up other data */
6519    
# Line 6076  firstline = (re->options & PCRE_FIRSTLIN Line 6523  firstline = (re->options & PCRE_FIRSTLIN
6523    
6524  /* The code starts after the real_pcre block and the capture name table. */  /* The code starts after the real_pcre block and the capture name table. */
6525    
6526  md->start_code = (const uschar *)external_re + re->name_table_offset +  md->start_code = (const pcre_uchar *)re + re->name_table_offset +
6527    re->name_count * re->name_entry_size;    re->name_count * re->name_entry_size;
6528    
6529  md->start_subject = (USPTR)subject;  md->start_subject = (PCRE_PUCHAR)subject;
6530  md->start_offset = start_offset;  md->start_offset = start_offset;
6531  md->end_subject = md->start_subject + length;  md->end_subject = md->start_subject + length;
6532  end_subject = md->end_subject;  end_subject = md->end_subject;
# Line 6104  md->recursive = NULL; Line 6551  md->recursive = NULL;
6551  md->hasthen = (re->flags & PCRE_HASTHEN) != 0;  md->hasthen = (re->flags & PCRE_HASTHEN) != 0;
6552    
6553  md->lcc = tables + lcc_offset;  md->lcc = tables + lcc_offset;
6554    md->fcc = tables + fcc_offset;
6555  md->ctypes = tables + ctypes_offset;  md->ctypes = tables + ctypes_offset;
6556    
6557  /* Handle different \R options. */  /* Handle different \R options. */
# Line 6190  arg_offset_max = (2*ocount)/3; Line 6638  arg_offset_max = (2*ocount)/3;
6638  if (re->top_backref > 0 && re->top_backref >= ocount/3)  if (re->top_backref > 0 && re->top_backref >= ocount/3)
6639    {    {
6640    ocount = re->top_backref * 3 + 3;    ocount = re->top_backref * 3 + 3;
6641    md->offset_vector = (int *)(pcre_malloc)(ocount * sizeof(int));    md->offset_vector = (int *)(PUBL(malloc))(ocount * sizeof(int));
6642    if (md->offset_vector == NULL) return PCRE_ERROR_NOMEMORY;    if (md->offset_vector == NULL) return PCRE_ERROR_NOMEMORY;
6643    using_temporary_offsets = TRUE;    using_temporary_offsets = TRUE;
6644    DPRINTF(("Got memory to hold back references\n"));    DPRINTF(("Got memory to hold back references\n"));
# Line 6217  if (md->offset_vector != NULL) Line 6665  if (md->offset_vector != NULL)
6665    md->offset_vector[0] = md->offset_vector[1] = -1;    md->offset_vector[0] = md->offset_vector[1] = -1;
6666    }    }
6667    
6668  /* Set up the first character to match, if available. The first_byte value is  /* Set up the first character to match, if available. The first_char value is
6669  never set for an anchored regular expression, but the anchoring may be forced  never set for an anchored regular expression, but the anchoring may be forced
6670  at run time, so we have to test for anchoring. The first char may be unset for  at run time, so we have to test for anchoring. The first char may be unset for
6671  an unanchored pattern, of course. If there's no first char and the pattern was  an unanchored pattern, of course. If there's no first char and the pattern was
# Line 6227  if (!anchored) Line 6675  if (!anchored)
6675    {    {
6676    if ((re->flags & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
6677      {      {
6678      first_byte = re->first_byte & 255;      has_first_char = TRUE;
6679      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)      first_char = first_char2 = (pcre_uchar)(re->first_char);
6680        first_byte = md->lcc[first_byte];      if ((re->flags & PCRE_FCH_CASELESS) != 0)
6681          {
6682          first_char2 = TABLE_GET(first_char, md->fcc, first_char);
6683    #if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
6684          if (utf && first_char > 127)
6685            first_char2 = UCD_OTHERCASE(first_char);
6686    #endif
6687          }
6688      }      }
6689    else    else
6690      if (!startline && study != NULL &&      if (!startline && study != NULL &&
# Line 6242  character" set. */ Line 6697  character" set. */
6697    
6698  if ((re->flags & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
6699    {    {
6700    req_byte = re->req_byte & 255;    has_req_char = TRUE;
6701    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;    req_char = req_char2 = (pcre_uchar)(re->req_char);
6702    req_byte2 = (tables + fcc_offset)[req_byte];  /* case flipped */    if ((re->flags & PCRE_RCH_CASELESS) != 0)
6703        {
6704        req_char2 = TABLE_GET(req_char, md->fcc, req_char);
6705    #if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
6706        if (utf && req_char > 127)
6707          req_char2 = UCD_OTHERCASE(req_char);
6708    #endif
6709        }
6710    }    }
6711    
6712    
   
   
6713  /* ==========================================================================*/  /* ==========================================================================*/
6714    
6715  /* Loop for handling unanchored repeated matching attempts; for anchored regexs  /* Loop for handling unanchored repeated matching attempts; for anchored regexs
# Line 6257  the loop runs just once. */ Line 6717  the loop runs just once. */
6717    
6718  for(;;)  for(;;)
6719    {    {
6720    USPTR save_end_subject = end_subject;    PCRE_PUCHAR save_end_subject = end_subject;
6721    USPTR new_start_match;    PCRE_PUCHAR new_start_match;
6722    
6723    /* If firstline is TRUE, the start of the match is constrained to the first    /* If firstline is TRUE, the start of the match is constrained to the first
6724    line of a multiline string. That is, the match must be before or at the first    line of a multiline string. That is, the match must be before or at the first
# Line 6268  for(;;) Line 6728  for(;;)
6728    
6729    if (firstline)    if (firstline)
6730      {      {
6731      USPTR t = start_match;      PCRE_PUCHAR t = start_match;
6732  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
6733      if (utf8)      if (utf)
6734        {        {
6735        while (t < md->end_subject && !IS_NEWLINE(t))        while (t < md->end_subject && !IS_NEWLINE(t))
6736          {          {
6737          t++;          t++;
6738          while (t < end_subject && (*t & 0xc0) == 0x80) t++;          ACROSSCHAR(t < end_subject, *t, t++);
6739          }          }
6740        }        }
6741      else      else
# Line 6292  for(;;) Line 6752  for(;;)
6752    
6753    if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0)    if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0)
6754      {      {
6755      /* Advance to a unique first byte if there is one. */      /* Advance to a unique first char if there is one. */
6756    
6757      if (first_byte >= 0)      if (has_first_char)
6758        {        {
6759        if (first_byte_caseless)        if (first_char != first_char2)
6760          while (start_match < end_subject && md->lcc[*start_match] != first_byte)          while (start_match < end_subject &&
6761                *start_match != first_char && *start_match != first_char2)
6762            start_match++;            start_match++;
6763        else        else
6764          while (start_match < end_subject && *start_match != first_byte)          while (start_match < end_subject && *start_match != first_char)
6765            start_match++;            start_match++;
6766        }        }
6767    
# Line 6310  for(;;) Line 6771  for(;;)
6771        {        {
6772        if (start_match > md->start_subject + start_offset)        if (start_match > md->start_subject + start_offset)
6773          {          {
6774  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
6775          if (utf8)          if (utf)
6776            {            {
6777            while (start_match < end_subject && !WAS_NEWLINE(start_match))            while (start_match < end_subject && !WAS_NEWLINE(start_match))
6778              {              {
6779              start_match++;              start_match++;
6780              while(start_match < end_subject && (*start_match & 0xc0) == 0x80)              ACROSSCHAR(start_match < end_subject, *start_match,
6781                start_match++;                start_match++);
6782              }              }
6783            }            }
6784          else          else
# Line 6344  for(;;) Line 6805  for(;;)
6805        while (start_match < end_subject)        while (start_match < end_subject)
6806          {          {
6807          register unsigned int c = *start_match;          register unsigned int c = *start_match;
6808    #ifndef COMPILE_PCRE8
6809            if (c > 255) c = 255;
6810    #endif
6811          if ((start_bits[c/8] & (1 << (c&7))) == 0)          if ((start_bits[c/8] & (1 << (c&7))) == 0)
6812            {            {
6813            start_match++;            start_match++;
6814  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF && defined COMPILE_PCRE8
6815            if (utf8)            /* In non 8-bit mode, the iteration will stop for
6816              while(start_match < end_subject && (*start_match & 0xc0) == 0x80)            characters > 255 at the beginning or not stop at all. */
6817                start_match++;            if (utf)
6818                ACROSSCHAR(start_match < end_subject, *start_match,
6819                  start_match++);
6820  #endif  #endif
6821            }            }
6822          else break;          else break;
# Line 6379  for(;;) Line 6845  for(;;)
6845        break;        break;
6846        }        }
6847    
6848      /* If req_byte is set, we know that that character must appear in the      /* If req_char is set, we know that that character must appear in the
6849      subject for the match to succeed. If the first character is set, req_byte      subject for the match to succeed. If the first character is set, req_char
6850      must be later in the subject; otherwise the test starts at the match point.      must be later in the subject; otherwise the test starts at the match point.
6851      This optimization can save a huge amount of backtracking in patterns with      This optimization can save a huge amount of backtracking in patterns with
6852      nested unlimited repeats that aren't going to match. Writing separate code      nested unlimited repeats that aren't going to match. Writing separate code
# Line 6393  for(;;) Line 6859  for(;;)
6859      32-megabyte string... so we don't do this when the string is sufficiently      32-megabyte string... so we don't do this when the string is sufficiently
6860      long. */      long. */
6861    
6862      if (req_byte >= 0 && end_subject - start_match < REQ_BYTE_MAX)      if (has_req_char && end_subject - start_match < REQ_BYTE_MAX)
6863        {        {
6864        register USPTR p = start_match + ((first_byte >= 0)? 1 : 0);        register PCRE_PUCHAR p = start_match + (has_first_char? 1:0);
6865    
6866        /* We don't need to repeat the search if we haven't yet reached the        /* We don't need to repeat the search if we haven't yet reached the
6867        place we found it at last time. */        place we found it at last time. */
6868    
6869        if (p > req_byte_ptr)        if (p > req_char_ptr)
6870          {          {
6871          if (req_byte_caseless)          if (req_char != req_char2)
6872            {            {
6873            while (p < end_subject)            while (p < end_subject)
6874              {              {
6875              register int pp = *p++;              register int pp = *p++;
6876              if (pp == req_byte || pp == req_byte2) { p--; break; }              if (pp == req_char || pp == req_char2) { p--; break; }
6877              }              }
6878            }            }
6879          else          else
6880            {            {
6881            while (p < end_subject)            while (p < end_subject)
6882              {              {
6883              if (*p++ == req_byte) { p--; break; }              if (*p++ == req_char) { p--; break; }
6884              }              }
6885            }            }
6886    
# Line 6431  for(;;) Line 6897  for(;;)
6897          found it, so that we don't search again next time round the loop if          found it, so that we don't search again next time round the loop if
6898          the start hasn't passed this character yet. */          the start hasn't passed this character yet. */
6899    
6900          req_byte_ptr = p;          req_char_ptr = p;
6901          }          }
6902        }        }
6903      }      }
# Line 6456  for(;;) Line 6922  for(;;)
6922    switch(rc)    switch(rc)
6923      {      {
6924      /* If MATCH_SKIP_ARG reaches this level it means that a MARK that matched      /* If MATCH_SKIP_ARG reaches this level it means that a MARK that matched
6925      the SKIP's arg was not found. In this circumstance, Perl ignores the SKIP      the SKIP's arg was not found. In this circumstance, Perl ignores the SKIP
6926      entirely. The only way we can do that is to re-do the match at the same      entirely. The only way we can do that is to re-do the match at the same
6927      point, with a flag to force SKIP with an argument to be ignored. Just      point, with a flag to force SKIP with an argument to be ignored. Just
6928      treating this case as NOMATCH does not work because it does not check other      treating this case as NOMATCH does not work because it does not check other
6929      alternatives in patterns such as A(*SKIP:A)B|AC when the subject is AC. */      alternatives in patterns such as A(*SKIP:A)B|AC when the subject is AC. */
6930    
6931      case MATCH_SKIP_ARG:      case MATCH_SKIP_ARG:
6932      new_start_match = start_match;      new_start_match = start_match;
6933      md->ignore_skip_arg = TRUE;      md->ignore_skip_arg = TRUE;
6934      break;      break;
6935    
6936      /* SKIP passes back the next starting point explicitly, but if it is the      /* SKIP passes back the next starting point explicitly, but if it is the
6937      same as the match we have just done, treat it as NOMATCH. */      same as the match we have just done, treat it as NOMATCH. */
# Line 6486  for(;;) Line 6952  for(;;)
6952      case MATCH_THEN:      case MATCH_THEN:
6953      md->ignore_skip_arg = FALSE;      md->ignore_skip_arg = FALSE;
6954      new_start_match = start_match + 1;      new_start_match = start_match + 1;
6955  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
6956      if (utf8)      if (utf)
6957        while(new_start_match < end_subject && (*new_start_match & 0xc0) == 0x80)        ACROSSCHAR(new_start_match < end_subject, *new_start_match,
6958          new_start_match++;          new_start_match++);
6959  #endif  #endif
6960      break;      break;
6961    
# Line 6527  for(;;) Line 6993  for(;;)
6993    
6994    /* If we have just passed a CR and we are now at a LF, and the pattern does    /* If we have just passed a CR and we are now at a LF, and the pattern does
6995    not contain any explicit matches for \r or \n, and the newline option is CRLF    not contain any explicit matches for \r or \n, and the newline option is CRLF
6996    or ANY or ANYCRLF, advance the match position by one more character. */    or ANY or ANYCRLF, advance the match position by one more character. In
6997      normal matching start_match will aways be greater than the first position at
6998      this stage, but a failed *SKIP can cause a return at the same point, which is
6999      why the first test exists. */
7000    
7001    if (start_match[-1] == CHAR_CR &&    if (start_match > (PCRE_PUCHAR)subject + start_offset &&
7002          start_match[-1] == CHAR_CR &&
7003        start_match < end_subject &&        start_match < end_subject &&
7004        *start_match == CHAR_NL &&        *start_match == CHAR_NL &&
7005        (re->flags & PCRE_HASCRORLF) == 0 &&        (re->flags & PCRE_HASCRORLF) == 0 &&
# Line 6575  if (rc == MATCH_MATCH || rc == MATCH_ACC Line 7045  if (rc == MATCH_MATCH || rc == MATCH_ACC
7045        }        }
7046      if (md->end_offset_top > arg_offset_max) md->offset_overflow = TRUE;      if (md->end_offset_top > arg_offset_max) md->offset_overflow = TRUE;
7047      DPRINTF(("Freeing temporary memory\n"));      DPRINTF(("Freeing temporary memory\n"));
7048      (pcre_free)(md->offset_vector);      (PUBL(free))(md->offset_vector);
7049      }      }
7050    
7051    /* Set the return code to the number of captured strings, or 0 if there were    /* Set the return code to the number of captured strings, or 0 if there were
# Line 6597  if (rc == MATCH_MATCH || rc == MATCH_ACC Line 7067  if (rc == MATCH_MATCH || rc == MATCH_ACC
7067      {      {
7068      register int *iptr, *iend;      register int *iptr, *iend;
7069      int resetcount = 2 + re->top_bracket * 2;      int resetcount = 2 + re->top_bracket * 2;
7070      if (resetcount > offsetcount) resetcount = ocount;      if (resetcount > offsetcount) resetcount = offsetcount;
7071      iptr = offsets + md->end_offset_top;      iptr = offsets + md->end_offset_top;
7072      iend = offsets + resetcount;      iend = offsets + resetcount;
7073      while (iptr < iend) *iptr++ = -1;      while (iptr < iend) *iptr++ = -1;
# Line 6614  if (rc == MATCH_MATCH || rc == MATCH_ACC Line 7084  if (rc == MATCH_MATCH || rc == MATCH_ACC
7084      }      }
7085    
7086    /* Return MARK data if requested */    /* Return MARK data if requested */
7087    
7088    if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)    if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)
7089      *(extra_data->mark) = (unsigned char *)(md->mark);      *(extra_data->mark) = (pcre_uchar *)md->mark;
7090    DPRINTF((">>>> returning %d\n", rc));    DPRINTF((">>>> returning %d\n", rc));
7091    #ifdef NO_RECURSE
7092      release_match_heapframes(&frame_zero);
7093    #endif
7094    return rc;    return rc;
7095    }    }
7096    
# Line 6627  attempt has failed at all permitted star Line 7100  attempt has failed at all permitted star
7100  if (using_temporary_offsets)  if (using_temporary_offsets)
7101    {    {
7102    DPRINTF(("Freeing temporary memory\n"));    DPRINTF(("Freeing temporary memory\n"));
7103    (pcre_free)(md->offset_vector);    (PUBL(free))(md->offset_vector);
7104    }    }
7105    
7106  /* For anything other than nomatch or partial match, just return the code. */  /* For anything other than nomatch or partial match, just return the code. */
# Line 6635  if (using_temporary_offsets) Line 7108  if (using_temporary_offsets)
7108  if (rc != MATCH_NOMATCH && rc != PCRE_ERROR_PARTIAL)  if (rc != MATCH_NOMATCH && rc != PCRE_ERROR_PARTIAL)
7109    {    {
7110    DPRINTF((">>>> error: returning %d\n", rc));    DPRINTF((">>>> error: returning %d\n", rc));
7111    #ifdef NO_RECURSE
7112      release_match_heapframes(&frame_zero);
7113    #endif
7114    return rc;    return rc;
7115    }    }
7116    
# Line 6646  if (start_partial != NULL) Line 7122  if (start_partial != NULL)
7122    md->mark = NULL;    md->mark = NULL;
7123    if (offsetcount > 1)    if (offsetcount > 1)
7124      {      {
7125      offsets[0] = (int)(start_partial - (USPTR)subject);      offsets[0] = (int)(start_partial - (PCRE_PUCHAR)subject);
7126      offsets[1] = (int)(end_subject - (USPTR)subject);      offsets[1] = (int)(end_subject - (PCRE_PUCHAR)subject);
7127      }      }
7128    rc = PCRE_ERROR_PARTIAL;    rc = PCRE_ERROR_PARTIAL;
7129    }    }
# Line 6663  else Line 7139  else
7139  /* Return the MARK data if it has been requested. */  /* Return the MARK data if it has been requested. */
7140    
7141  if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)  if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)
7142    *(extra_data->mark) = (unsigned char *)(md->nomatch_mark);    *(extra_data->mark) = (pcre_uchar *)md->nomatch_mark;
7143    #ifdef NO_RECURSE
7144      release_match_heapframes(&frame_zero);
7145    #endif
7146  return rc;  return rc;
7147  }  }
7148    

Legend:
Removed from v.778  
changed lines
  Added in v.976

  ViewVC Help
Powered by ViewVC 1.1.5