/[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 642 by ph10, Thu Jul 28 18:59:40 2011 UTC revision 888 by ph10, Tue Jan 17 14:43:23 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 82  negative to avoid the external error cod Line 82  negative to avoid the external error cod
82  #define MATCH_SKIP_ARG     (-993)  #define MATCH_SKIP_ARG     (-993)
83  #define MATCH_THEN         (-992)  #define MATCH_THEN         (-992)
84    
 /* This is a convenience macro for code that occurs many times. */  
   
 #define MRRETURN(ra) \  
   { \  
   md->mark = markptr; \  
   RRETURN(ra); \  
   }  
   
85  /* Maximum number of ints of offset to save on the stack for recursive calls.  /* Maximum number of ints of offset to save on the stack for recursive calls.
86  If the offset vector is bigger, malloc is used. This should be a multiple of 3,  If the offset vector is bigger, malloc is used. This should be a multiple of 3,
87  because the offset vector is always a multiple of 3 long. */  because the offset vector is always a multiple of 3 long. */
# Line 121  Returns: nothing Line 113  Returns: nothing
113  */  */
114    
115  static void  static void
116  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)
117  {  {
118  unsigned int c;  unsigned int c;
119  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 152  Returns: < 0 if not matched, otherw Line 144  Returns: < 0 if not matched, otherw
144  */  */
145    
146  static int  static int
147  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,
148    BOOL caseless)    BOOL caseless)
149  {  {
150  USPTR eptr_start = eptr;  PCRE_PUCHAR eptr_start = eptr;
151  register USPTR p = md->start_subject + md->offset_vector[offset];  register PCRE_PUCHAR p = md->start_subject + md->offset_vector[offset];
152    
153  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
154  if (eptr >= md->end_subject)  if (eptr >= md->end_subject)
# Line 181  ASCII characters. */ Line 173  ASCII characters. */
173    
174  if (caseless)  if (caseless)
175    {    {
176  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
177  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
178    if (md->utf8)    if (md->utf)
179      {      {
180      /* 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
181      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 193  if (caseless) Line 185  if (caseless)
185      the latter. It is important, therefore, to check the length along the      the latter. It is important, therefore, to check the length along the
186      reference, not along the subject (earlier code did this wrong). */      reference, not along the subject (earlier code did this wrong). */
187    
188      USPTR endptr = p + length;      PCRE_PUCHAR endptr = p + length;
189      while (p < endptr)      while (p < endptr)
190        {        {
191        int c, d;        int c, d;
# Line 212  if (caseless) Line 204  if (caseless)
204      {      {
205      if (eptr + length > md->end_subject) return -1;      if (eptr + length > md->end_subject) return -1;
206      while (length-- > 0)      while (length-- > 0)
207        { if (md->lcc[*p++] != md->lcc[*eptr++]) return -1; }        {
208          if (TABLE_GET(*p, md->lcc, *p) != TABLE_GET(*eptr, md->lcc, *eptr)) return -1;
209          p++;
210          eptr++;
211          }
212      }      }
213    }    }
214    
# Line 225  else Line 221  else
221    while (length-- > 0) if (*p++ != *eptr++) return -1;    while (length-- > 0) if (*p++ != *eptr++) return -1;
222    }    }
223    
224  return eptr - eptr_start;  return (int)(eptr - eptr_start);
225  }  }
226    
227    
# Line 277  enum { RM1=1, RM2, RM3, RM4, RM5, RM Line 273  enum { RM1=1, RM2, RM3, RM4, RM5, RM
273         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
274         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
275         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,
276         RM61,  RM62, RM63 };         RM61,  RM62, RM63, RM64, RM65, RM66 };
277    
278  /* These versions of the macros use the stack, as normal. There are debugging  /* These versions of the macros use the stack, as normal. There are debugging
279  versions and production versions. Note that the "rw" argument of RMATCH isn't  versions and production versions. Note that the "rw" argument of RMATCH isn't
# Line 290  actually used in this definition. */ Line 286  actually used in this definition. */
286  #define RMATCH(ra,rb,rc,rd,re,rw) \  #define RMATCH(ra,rb,rc,rd,re,rw) \
287    { \    { \
288    printf("match() called in line %d\n", __LINE__); \    printf("match() called in line %d\n", __LINE__); \
289    rrc = match(ra,rb,mstart,markptr,rc,rd,re,rdepth+1); \    rrc = match(ra,rb,mstart,rc,rd,re,rdepth+1); \
290    printf("to line %d\n", __LINE__); \    printf("to line %d\n", __LINE__); \
291    }    }
292  #define RRETURN(ra) \  #define RRETURN(ra) \
# Line 300  actually used in this definition. */ Line 296  actually used in this definition. */
296    }    }
297  #else  #else
298  #define RMATCH(ra,rb,rc,rd,re,rw) \  #define RMATCH(ra,rb,rc,rd,re,rw) \
299    rrc = match(ra,rb,mstart,markptr,rc,rd,re,rdepth+1)    rrc = match(ra,rb,mstart,rc,rd,re,rdepth+1)
300  #define RRETURN(ra) return ra  #define RRETURN(ra) return ra
301  #endif  #endif
302    
# Line 315  argument of match(), which never changes Line 311  argument of match(), which never changes
311    
312  #define RMATCH(ra,rb,rc,rd,re,rw)\  #define RMATCH(ra,rb,rc,rd,re,rw)\
313    {\    {\
314    heapframe *newframe = (heapframe *)(pcre_stack_malloc)(sizeof(heapframe));\    heapframe *newframe = (heapframe *)(PUBL(stack_malloc))(sizeof(heapframe));\
315    if (newframe == NULL) RRETURN(PCRE_ERROR_NOMEMORY);\    if (newframe == NULL) RRETURN(PCRE_ERROR_NOMEMORY);\
316    frame->Xwhere = rw; \    frame->Xwhere = rw; \
317    newframe->Xeptr = ra;\    newframe->Xeptr = ra;\
318    newframe->Xecode = rb;\    newframe->Xecode = rb;\
319    newframe->Xmstart = mstart;\    newframe->Xmstart = mstart;\
   newframe->Xmarkptr = markptr;\  
320    newframe->Xoffset_top = rc;\    newframe->Xoffset_top = rc;\
321    newframe->Xeptrb = re;\    newframe->Xeptrb = re;\
322    newframe->Xrdepth = frame->Xrdepth + 1;\    newframe->Xrdepth = frame->Xrdepth + 1;\
# Line 337  argument of match(), which never changes Line 332  argument of match(), which never changes
332    {\    {\
333    heapframe *oldframe = frame;\    heapframe *oldframe = frame;\
334    frame = oldframe->Xprevframe;\    frame = oldframe->Xprevframe;\
335    (pcre_stack_free)(oldframe);\    (PUBL(stack_free))(oldframe);\
336    if (frame != NULL)\    if (frame != NULL)\
337      {\      {\
338      rrc = ra;\      rrc = ra;\
# Line 354  typedef struct heapframe { Line 349  typedef struct heapframe {
349    
350    /* Function arguments that may change */    /* Function arguments that may change */
351    
352    USPTR Xeptr;    PCRE_PUCHAR Xeptr;
353    const uschar *Xecode;    const pcre_uchar *Xecode;
354    USPTR Xmstart;    PCRE_PUCHAR Xmstart;
   USPTR Xmarkptr;  
355    int Xoffset_top;    int Xoffset_top;
356    eptrblock *Xeptrb;    eptrblock *Xeptrb;
357    unsigned int Xrdepth;    unsigned int Xrdepth;
358    
359    /* Function local variables */    /* Function local variables */
360    
361    USPTR Xcallpat;    PCRE_PUCHAR Xcallpat;
362  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
363    USPTR Xcharptr;    PCRE_PUCHAR Xcharptr;
364  #endif  #endif
365    USPTR Xdata;    PCRE_PUCHAR Xdata;
366    USPTR Xnext;    PCRE_PUCHAR Xnext;
367    USPTR Xpp;    PCRE_PUCHAR Xpp;
368    USPTR Xprev;    PCRE_PUCHAR Xprev;
369    USPTR Xsaved_eptr;    PCRE_PUCHAR Xsaved_eptr;
370    
371    recursion_info Xnew_recursive;    recursion_info Xnew_recursive;
372    
# Line 385  typedef struct heapframe { Line 379  typedef struct heapframe {
379    int Xprop_value;    int Xprop_value;
380    int Xprop_fail_result;    int Xprop_fail_result;
381    int Xoclength;    int Xoclength;
382    uschar Xocchars[8];    pcre_uchar Xocchars[6];
383  #endif  #endif
384    
385    int Xcodelink;    int Xcodelink;
# Line 427  returns a negative (error) response, the Line 421  returns a negative (error) response, the
421  same response. */  same response. */
422    
423  /* These macros pack up tests that are used for partial matching, and which  /* These macros pack up tests that are used for partial matching, and which
424  appears several times in the code. We set the "hit end" flag if the pointer is  appear several times in the code. We set the "hit end" flag if the pointer is
425  at the end of the subject and also past the start of the subject (i.e.  at the end of the subject and also past the start of the subject (i.e.
426  something has been matched). For hard partial matching, we then return  something has been matched). For hard partial matching, we then return
427  immediately. The second one is used when we already know we are past the end of  immediately. The second one is used when we already know we are past the end of
# Line 438  the subject. */ Line 432  the subject. */
432        eptr > md->start_used_ptr) \        eptr > md->start_used_ptr) \
433      { \      { \
434      md->hitend = TRUE; \      md->hitend = TRUE; \
435      if (md->partial > 1) MRRETURN(PCRE_ERROR_PARTIAL); \      if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); \
436      }      }
437    
438  #define SCHECK_PARTIAL()\  #define SCHECK_PARTIAL()\
439    if (md->partial != 0 && eptr > md->start_used_ptr) \    if (md->partial != 0 && eptr > md->start_used_ptr) \
440      { \      { \
441      md->hitend = TRUE; \      md->hitend = TRUE; \
442      if (md->partial > 1) MRRETURN(PCRE_ERROR_PARTIAL); \      if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); \
443      }      }
444    
445    
446  /* Performance note: It might be tempting to extract commonly used fields from  /* Performance note: It might be tempting to extract commonly used fields from
447  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
448  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
449  made performance worse.  made performance worse.
450    
# Line 459  Arguments: Line 453  Arguments:
453     ecode       pointer to current position in compiled code     ecode       pointer to current position in compiled code
454     mstart      pointer to the current match start position (can be modified     mstart      pointer to the current match start position (can be modified
455                   by encountering \K)                   by encountering \K)
    markptr     pointer to the most recent MARK name, or NULL  
456     offset_top  current top pointer     offset_top  current top pointer
457     md          pointer to "static" info for the match     md          pointer to "static" info for the match
458     eptrb       pointer to chain of blocks containing eptr at start of     eptrb       pointer to chain of blocks containing eptr at start of
# Line 474  Returns: MATCH_MATCH if matched Line 467  Returns: MATCH_MATCH if matched
467  */  */
468    
469  static int  static int
470  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, USPTR mstart,  match(REGISTER PCRE_PUCHAR eptr, REGISTER const pcre_uchar *ecode,
471    const uschar *markptr, int offset_top, match_data *md, eptrblock *eptrb,    PCRE_PUCHAR mstart, int offset_top, match_data *md, eptrblock *eptrb,
472    unsigned int rdepth)    unsigned int rdepth)
473  {  {
474  /* 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,
# Line 485  so they can be ordinary variables in all Line 478  so they can be ordinary variables in all
478  register int  rrc;         /* Returns from recursive calls */  register int  rrc;         /* Returns from recursive calls */
479  register int  i;           /* Used for loops not involving calls to RMATCH() */  register int  i;           /* Used for loops not involving calls to RMATCH() */
480  register unsigned int c;   /* Character values not kept over RMATCH() calls */  register unsigned int c;   /* Character values not kept over RMATCH() calls */
481  register BOOL utf8;        /* Local copy of UTF-8 flag for speed */  register BOOL utf;         /* Local copy of UTF flag for speed */
482    
483  BOOL minimize, possessive; /* Quantifier options */  BOOL minimize, possessive; /* Quantifier options */
484  BOOL caseless;  BOOL caseless;
# Line 497  heap storage. Set up the top-level frame Line 490  heap storage. Set up the top-level frame
490  heap whenever RMATCH() does a "recursion". See the macro definitions above. */  heap whenever RMATCH() does a "recursion". See the macro definitions above. */
491    
492  #ifdef NO_RECURSE  #ifdef NO_RECURSE
493  heapframe *frame = (heapframe *)(pcre_stack_malloc)(sizeof(heapframe));  heapframe *frame = (heapframe *)(PUBL(stack_malloc))(sizeof(heapframe));
494  if (frame == NULL) RRETURN(PCRE_ERROR_NOMEMORY);  if (frame == NULL) RRETURN(PCRE_ERROR_NOMEMORY);
495  frame->Xprevframe = NULL;            /* Marks the top level */  frame->Xprevframe = NULL;            /* Marks the top level */
496    
# Line 506  frame->Xprevframe = NULL; /* Line 499  frame->Xprevframe = NULL; /*
499  frame->Xeptr = eptr;  frame->Xeptr = eptr;
500  frame->Xecode = ecode;  frame->Xecode = ecode;
501  frame->Xmstart = mstart;  frame->Xmstart = mstart;
 frame->Xmarkptr = markptr;  
502  frame->Xoffset_top = offset_top;  frame->Xoffset_top = offset_top;
503  frame->Xeptrb = eptrb;  frame->Xeptrb = eptrb;
504  frame->Xrdepth = rdepth;  frame->Xrdepth = rdepth;
# Line 520  HEAP_RECURSE: Line 512  HEAP_RECURSE:
512  #define eptr               frame->Xeptr  #define eptr               frame->Xeptr
513  #define ecode              frame->Xecode  #define ecode              frame->Xecode
514  #define mstart             frame->Xmstart  #define mstart             frame->Xmstart
 #define markptr            frame->Xmarkptr  
515  #define offset_top         frame->Xoffset_top  #define offset_top         frame->Xoffset_top
516  #define eptrb              frame->Xeptrb  #define eptrb              frame->Xeptrb
517  #define rdepth             frame->Xrdepth  #define rdepth             frame->Xrdepth
518    
519  /* Ditto for the local variables */  /* Ditto for the local variables */
520    
521  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
522  #define charptr            frame->Xcharptr  #define charptr            frame->Xcharptr
523  #endif  #endif
524  #define callpat            frame->Xcallpat  #define callpat            frame->Xcallpat
# Line 585  declarations can be cut out in a block. Line 576  declarations can be cut out in a block.
576  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
577  to RMATCH(). */  to RMATCH(). */
578    
579  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
580  const uschar *charptr;  const pcre_uchar *charptr;
581  #endif  #endif
582  const uschar *callpat;  const pcre_uchar *callpat;
583  const uschar *data;  const pcre_uchar *data;
584  const uschar *next;  const pcre_uchar *next;
585  USPTR         pp;  PCRE_PUCHAR       pp;
586  const uschar *prev;  const pcre_uchar *prev;
587  USPTR         saved_eptr;  PCRE_PUCHAR       saved_eptr;
588    
589  recursion_info new_recursive;  recursion_info new_recursive;
590    
# Line 606  int prop_type; Line 597  int prop_type;
597  int prop_value;  int prop_value;
598  int prop_fail_result;  int prop_fail_result;
599  int oclength;  int oclength;
600  uschar occhars[8];  pcre_uchar occhars[6];
601  #endif  #endif
602    
603  int codelink;  int codelink;
# Line 634  the alternative names that are used. */ Line 625  the alternative names that are used. */
625  #define code_offset   codelink  #define code_offset   codelink
626  #define condassert    condition  #define condassert    condition
627  #define matched_once  prev_is_word  #define matched_once  prev_is_word
628    #define foc           number
629    #define save_mark     data
630    
631  /* These statements are here to stop the compiler complaining about unitialized  /* These statements are here to stop the compiler complaining about unitialized
632  variables. */  variables. */
# Line 659  defined). However, RMATCH isn't like a f Line 652  defined). However, RMATCH isn't like a f
652  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,
653  however, impact performance when true recursion is being used. */  however, impact performance when true recursion is being used. */
654    
655  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
656  utf8 = md->utf8;       /* Local copy of the flag */  utf = md->utf;       /* Local copy of the flag */
657  #else  #else
658  utf8 = FALSE;  utf = FALSE;
659  #endif  #endif
660    
661  /* 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 701  for (;;) Line 694  for (;;)
694    switch(op)    switch(op)
695      {      {
696      case OP_MARK:      case OP_MARK:
697      markptr = ecode + 2;      md->nomatch_mark = ecode + 2;
698      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      md->mark = NULL;    /* In case previously set by assertion */
699        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
700        eptrb, RM55);        eptrb, RM55);
701        if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
702             md->mark == NULL) md->mark = ecode + 2;
703    
704      /* A return of MATCH_SKIP_ARG means that matching failed at SKIP with an      /* A return of MATCH_SKIP_ARG means that matching failed at SKIP with an
705      argument, and we must check whether that argument matches this MARK's      argument, and we must check whether that argument matches this MARK's
# Line 712  for (;;) Line 708  for (;;)
708      position and return MATCH_SKIP. Otherwise, pass back the return code      position and return MATCH_SKIP. Otherwise, pass back the return code
709      unaltered. */      unaltered. */
710    
711      if (rrc == MATCH_SKIP_ARG &&      else if (rrc == MATCH_SKIP_ARG &&
712          strcmp((char *)markptr, (char *)(md->start_match_ptr)) == 0)          STRCMP_UC_UC(ecode + 2, md->start_match_ptr) == 0)
713        {        {
714        md->start_match_ptr = eptr;        md->start_match_ptr = eptr;
715        RRETURN(MATCH_SKIP);        RRETURN(MATCH_SKIP);
716        }        }
   
     if (md->mark == NULL) md->mark = markptr;  
717      RRETURN(rrc);      RRETURN(rrc);
718    
719      case OP_FAIL:      case OP_FAIL:
720      MRRETURN(MATCH_NOMATCH);      RRETURN(MATCH_NOMATCH);
721    
722      /* COMMIT overrides PRUNE, SKIP, and THEN */      /* COMMIT overrides PRUNE, SKIP, and THEN */
723    
724      case OP_COMMIT:      case OP_COMMIT:
725      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
726        eptrb, RM52);        eptrb, RM52);
727      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE &&      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE &&
728          rrc != MATCH_SKIP && rrc != MATCH_SKIP_ARG &&          rrc != MATCH_SKIP && rrc != MATCH_SKIP_ARG &&
729          rrc != MATCH_THEN)          rrc != MATCH_THEN)
730        RRETURN(rrc);        RRETURN(rrc);
731      MRRETURN(MATCH_COMMIT);      RRETURN(MATCH_COMMIT);
732    
733      /* PRUNE overrides THEN */      /* PRUNE overrides THEN */
734    
735      case OP_PRUNE:      case OP_PRUNE:
736      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
737        eptrb, RM51);        eptrb, RM51);
738      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
739      MRRETURN(MATCH_PRUNE);      RRETURN(MATCH_PRUNE);
740    
741      case OP_PRUNE_ARG:      case OP_PRUNE_ARG:
742      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      md->nomatch_mark = ecode + 2;
743        md->mark = NULL;    /* In case previously set by assertion */
744        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
745        eptrb, RM56);        eptrb, RM56);
746        if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
747             md->mark == NULL) md->mark = ecode + 2;
748      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
     md->mark = ecode + 2;  
749      RRETURN(MATCH_PRUNE);      RRETURN(MATCH_PRUNE);
750    
751      /* SKIP overrides PRUNE and THEN */      /* SKIP overrides PRUNE and THEN */
752    
753      case OP_SKIP:      case OP_SKIP:
754      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
755        eptrb, RM53);        eptrb, RM53);
756      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)
757        RRETURN(rrc);        RRETURN(rrc);
758      md->start_match_ptr = eptr;   /* Pass back current position */      md->start_match_ptr = eptr;   /* Pass back current position */
759      MRRETURN(MATCH_SKIP);      RRETURN(MATCH_SKIP);
760    
761        /* Note that, for Perl compatibility, SKIP with an argument does NOT set
762        nomatch_mark. There is a flag that disables this opcode when re-matching a
763        pattern that ended with a SKIP for which there was not a matching MARK. */
764    
765      case OP_SKIP_ARG:      case OP_SKIP_ARG:
766      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      if (md->ignore_skip_arg)
767          {
768          ecode += PRIV(OP_lengths)[*ecode] + ecode[1];
769          break;
770          }
771        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
772        eptrb, RM57);        eptrb, RM57);
773      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)
774        RRETURN(rrc);        RRETURN(rrc);
775    
776      /* 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
777      returning the special MATCH_SKIP_ARG return code. This will either be      returning the special MATCH_SKIP_ARG return code. This will either be
778      caught by a matching MARK, or get to the top, where it is treated the same      caught by a matching MARK, or get to the top, where it causes a rematch
779      as PRUNE. */      with the md->ignore_skip_arg flag set. */
780    
781      md->start_match_ptr = ecode + 2;      md->start_match_ptr = ecode + 2;
782      RRETURN(MATCH_SKIP_ARG);      RRETURN(MATCH_SKIP_ARG);
783    
784      /* For THEN (and THEN_ARG) we pass back the address of the bracket or      /* For THEN (and THEN_ARG) we pass back the address of the opcode, so that
785      the alt that is at the start of the current branch. This makes it possible      the branch in which it occurs can be determined. Overload the start of
786      to skip back past alternatives that precede the THEN within the current      match pointer to do this. */
     branch. */  
787    
788      case OP_THEN:      case OP_THEN:
789      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
790        eptrb, RM54);        eptrb, RM54);
791      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
792      md->start_match_ptr = ecode - GET(ecode, 1);      md->start_match_ptr = ecode;
793      MRRETURN(MATCH_THEN);      RRETURN(MATCH_THEN);
794    
795      case OP_THEN_ARG:      case OP_THEN_ARG:
796      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1+LINK_SIZE],      md->nomatch_mark = ecode + 2;
797        offset_top, md, eptrb, RM58);      md->mark = NULL;    /* In case previously set by assertion */
798        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top,
799          md, eptrb, RM58);
800        if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
801             md->mark == NULL) md->mark = ecode + 2;
802      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
803      md->start_match_ptr = ecode - GET(ecode, 1);      md->start_match_ptr = ecode;
     md->mark = ecode + LINK_SIZE + 2;  
804      RRETURN(MATCH_THEN);      RRETURN(MATCH_THEN);
805    
806        /* Handle an atomic group that does not contain any capturing parentheses.
807        This can be handled like an assertion. Prior to 8.13, all atomic groups
808        were handled this way. In 8.13, the code was changed as below for ONCE, so
809        that backups pass through the group and thereby reset captured values.
810        However, this uses a lot more stack, so in 8.20, atomic groups that do not
811        contain any captures generate OP_ONCE_NC, which can be handled in the old,
812        less stack intensive way.
813    
814        Check the alternative branches in turn - the matching won't pass the KET
815        for this kind of subpattern. If any one branch matches, we carry on as at
816        the end of a normal bracket, leaving the subject pointer, but resetting
817        the start-of-match value in case it was changed by \K. */
818    
819        case OP_ONCE_NC:
820        prev = ecode;
821        saved_eptr = eptr;
822        save_mark = md->mark;
823        do
824          {
825          RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM64);
826          if (rrc == MATCH_MATCH)  /* Note: _not_ MATCH_ACCEPT */
827            {
828            mstart = md->start_match_ptr;
829            break;
830            }
831          if (rrc == MATCH_THEN)
832            {
833            next = ecode + GET(ecode,1);
834            if (md->start_match_ptr < next &&
835                (*ecode == OP_ALT || *next == OP_ALT))
836              rrc = MATCH_NOMATCH;
837            }
838    
839          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
840          ecode += GET(ecode,1);
841          md->mark = save_mark;
842          }
843        while (*ecode == OP_ALT);
844    
845        /* If hit the end of the group (which could be repeated), fail */
846    
847        if (*ecode != OP_ONCE_NC && *ecode != OP_ALT) RRETURN(MATCH_NOMATCH);
848    
849        /* Continue as from after the group, updating the offsets high water
850        mark, since extracts may have been taken. */
851    
852        do ecode += GET(ecode, 1); while (*ecode == OP_ALT);
853    
854        offset_top = md->end_offset_top;
855        eptr = md->end_match_ptr;
856    
857        /* For a non-repeating ket, just continue at this level. This also
858        happens for a repeating ket if no characters were matched in the group.
859        This is the forcible breaking of infinite loops as implemented in Perl
860        5.005. */
861    
862        if (*ecode == OP_KET || eptr == saved_eptr)
863          {
864          ecode += 1+LINK_SIZE;
865          break;
866          }
867    
868        /* The repeating kets try the rest of the pattern or restart from the
869        preceding bracket, in the appropriate order. The second "call" of match()
870        uses tail recursion, to avoid using another stack frame. */
871    
872        if (*ecode == OP_KETRMIN)
873          {
874          RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM65);
875          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
876          ecode = prev;
877          goto TAIL_RECURSE;
878          }
879        else  /* OP_KETRMAX */
880          {
881          md->match_function_type = MATCH_CBEGROUP;
882          RMATCH(eptr, prev, offset_top, md, eptrb, RM66);
883          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
884          ecode += 1 + LINK_SIZE;
885          goto TAIL_RECURSE;
886          }
887        /* Control never gets here */
888    
889      /* Handle a capturing bracket, other than those that are possessive with an      /* Handle a capturing bracket, other than those that are possessive with an
890      unlimited repeat. If there is space in the offset vector, save the current      unlimited repeat. If there is space in the offset vector, save the current
891      subject position in the working slot at the top of the vector. We mustn't      subject position in the working slot at the top of the vector. We mustn't
# Line 827  for (;;) Line 918  for (;;)
918        save_offset2 = md->offset_vector[offset+1];        save_offset2 = md->offset_vector[offset+1];
919        save_offset3 = md->offset_vector[md->offset_end - number];        save_offset3 = md->offset_vector[md->offset_end - number];
920        save_capture_last = md->capture_last;        save_capture_last = md->capture_last;
921          save_mark = md->mark;
922    
923        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));
924        md->offset_vector[md->offset_end - number] =        md->offset_vector[md->offset_end - number] =
# Line 835  for (;;) Line 927  for (;;)
927        for (;;)        for (;;)
928          {          {
929          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
930          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
931            eptrb, RM1);            eptrb, RM1);
932          if (rrc == MATCH_ONCE) break;  /* Backing up through an atomic group */          if (rrc == MATCH_ONCE) break;  /* Backing up through an atomic group */
933          if (rrc != MATCH_NOMATCH &&  
934              (rrc != MATCH_THEN || md->start_match_ptr != ecode))          /* If we backed up to a THEN, check whether it is within the current
935            RRETURN(rrc);          branch by comparing the address of the THEN that is passed back with
936            the end of the branch. If it is within the current branch, and the
937            branch is one of two or more alternatives (it either starts or ends
938            with OP_ALT), we have reached the limit of THEN's action, so convert
939            the return code to NOMATCH, which will cause normal backtracking to
940            happen from now on. Otherwise, THEN is passed back to an outer
941            alternative. This implements Perl's treatment of parenthesized groups,
942            where a group not containing | does not affect the current alternative,
943            that is, (X) is NOT the same as (X|(*F)). */
944    
945            if (rrc == MATCH_THEN)
946              {
947              next = ecode + GET(ecode,1);
948              if (md->start_match_ptr < next &&
949                  (*ecode == OP_ALT || *next == OP_ALT))
950                rrc = MATCH_NOMATCH;
951              }
952    
953            /* Anything other than NOMATCH is passed back. */
954    
955            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
956          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
957          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
958            md->mark = save_mark;
959          if (*ecode != OP_ALT) break;          if (*ecode != OP_ALT) break;
960          }          }
961    
# Line 851  for (;;) Line 964  for (;;)
964        md->offset_vector[offset+1] = save_offset2;        md->offset_vector[offset+1] = save_offset2;
965        md->offset_vector[md->offset_end - number] = save_offset3;        md->offset_vector[md->offset_end - number] = save_offset3;
966    
967        /* At this point, rrc will be one of MATCH_ONCE, MATCH_NOMATCH, or        /* At this point, rrc will be one of MATCH_ONCE or MATCH_NOMATCH. */
       MATCH_THEN. */  
968    
969        if (rrc != MATCH_THEN && md->mark == NULL) md->mark = markptr;        RRETURN(rrc);
       RRETURN(((rrc == MATCH_ONCE)? MATCH_ONCE:MATCH_NOMATCH));  
970        }        }
971    
972      /* FALL THROUGH ... Insufficient room for saving captured contents. Treat      /* FALL THROUGH ... Insufficient room for saving captured contents. Treat
# Line 870  for (;;) Line 981  for (;;)
981      /* VVVVVVVVVVVVVVVVVVVVVVVVV */      /* VVVVVVVVVVVVVVVVVVVVVVVVV */
982    
983      /* Non-capturing or atomic group, except for possessive with unlimited      /* Non-capturing or atomic group, except for possessive with unlimited
984      repeat. Loop for all the alternatives. When we get to the final alternative      repeat and ONCE group with no captures. Loop for all the alternatives.
985      within the brackets, we used to return the result of a recursive call to  
986      match() whatever happened so it was possible to reduce stack usage by      When we get to the final alternative within the brackets, we used to return
987      turning this into a tail recursion, except in the case of a possibly empty      the result of a recursive call to match() whatever happened so it was
988      group. However, now that there is the possiblity of (*THEN) occurring in      possible to reduce stack usage by turning this into a tail recursion,
989      the final alternative, this optimization is no longer possible.      except in the case of a possibly empty group. However, now that there is
990        the possiblity of (*THEN) occurring in the final alternative, this
991        optimization is no longer always possible.
992    
993        We can optimize if we know there are no (*THEN)s in the pattern; at present
994        this is the best that can be done.
995    
996      MATCH_ONCE is returned when the end of an atomic group is successfully      MATCH_ONCE is returned when the end of an atomic group is successfully
997      reached, but subsequent matching fails. It passes back up the tree (causing      reached, but subsequent matching fails. It passes back up the tree (causing
# Line 892  for (;;) Line 1008  for (;;)
1008      for (;;)      for (;;)
1009        {        {
1010        if (op >= OP_SBRA || op == OP_ONCE) md->match_function_type = MATCH_CBEGROUP;        if (op >= OP_SBRA || op == OP_ONCE) md->match_function_type = MATCH_CBEGROUP;
1011        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, eptrb,  
1012          /* If this is not a possibly empty group, and there are no (*THEN)s in
1013          the pattern, and this is the final alternative, optimize as described
1014          above. */
1015    
1016          else if (!md->hasthen && ecode[GET(ecode, 1)] != OP_ALT)
1017            {
1018            ecode += PRIV(OP_lengths)[*ecode];
1019            goto TAIL_RECURSE;
1020            }
1021    
1022          /* In all other cases, we have to make another call to match(). */
1023    
1024          save_mark = md->mark;
1025          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb,
1026          RM2);          RM2);
1027        if (rrc != MATCH_NOMATCH &&  
1028            (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* See comment in the code for capturing groups above about handling
1029          THEN. */
1030    
1031          if (rrc == MATCH_THEN)
1032            {
1033            next = ecode + GET(ecode,1);
1034            if (md->start_match_ptr < next &&
1035                (*ecode == OP_ALT || *next == OP_ALT))
1036              rrc = MATCH_NOMATCH;
1037            }
1038    
1039          if (rrc != MATCH_NOMATCH)
1040          {          {
1041          if (rrc == MATCH_ONCE)          if (rrc == MATCH_ONCE)
1042            {            {
1043            const uschar *scode = ecode;            const pcre_uchar *scode = ecode;
1044            if (*scode != OP_ONCE)           /* If not at start, find it */            if (*scode != OP_ONCE)           /* If not at start, find it */
1045              {              {
1046              while (*scode == OP_ALT) scode += GET(scode, 1);              while (*scode == OP_ALT) scode += GET(scode, 1);
# Line 910  for (;;) Line 1051  for (;;)
1051          RRETURN(rrc);          RRETURN(rrc);
1052          }          }
1053        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1054          md->mark = save_mark;
1055        if (*ecode != OP_ALT) break;        if (*ecode != OP_ALT) break;
1056        }        }
1057      if (rrc != MATCH_THEN && md->mark == NULL) md->mark = markptr;  
1058      RRETURN(MATCH_NOMATCH);      RRETURN(MATCH_NOMATCH);
1059    
1060      /* Handle possessive capturing brackets with an unlimited repeat. We come      /* Handle possessive capturing brackets with an unlimited repeat. We come
# Line 941  for (;;) Line 1083  for (;;)
1083      if (offset < md->offset_max)      if (offset < md->offset_max)
1084        {        {
1085        matched_once = FALSE;        matched_once = FALSE;
1086        code_offset = ecode - md->start_code;        code_offset = (int)(ecode - md->start_code);
1087    
1088        save_offset1 = md->offset_vector[offset];        save_offset1 = md->offset_vector[offset];
1089        save_offset2 = md->offset_vector[offset+1];        save_offset2 = md->offset_vector[offset+1];
# Line 964  for (;;) Line 1106  for (;;)
1106          md->offset_vector[md->offset_end - number] =          md->offset_vector[md->offset_end - number] =
1107            (int)(eptr - md->start_subject);            (int)(eptr - md->start_subject);
1108          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
1109          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
1110            eptrb, RM63);            eptrb, RM63);
1111          if (rrc == MATCH_KETRPOS)          if (rrc == MATCH_KETRPOS)
1112            {            {
# Line 975  for (;;) Line 1117  for (;;)
1117            matched_once = TRUE;            matched_once = TRUE;
1118            continue;            continue;
1119            }            }
1120          if (rrc != MATCH_NOMATCH &&  
1121              (rrc != MATCH_THEN || md->start_match_ptr != ecode))          /* See comment in the code for capturing groups above about handling
1122            RRETURN(rrc);          THEN. */
1123    
1124            if (rrc == MATCH_THEN)
1125              {
1126              next = ecode + GET(ecode,1);
1127              if (md->start_match_ptr < next &&
1128                  (*ecode == OP_ALT || *next == OP_ALT))
1129                rrc = MATCH_NOMATCH;
1130              }
1131    
1132            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1133          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
1134          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
1135          if (*ecode != OP_ALT) break;          if (*ecode != OP_ALT) break;
# Line 990  for (;;) Line 1142  for (;;)
1142          md->offset_vector[md->offset_end - number] = save_offset3;          md->offset_vector[md->offset_end - number] = save_offset3;
1143          }          }
1144    
       if (rrc != MATCH_THEN && md->mark == NULL) md->mark = markptr;  
1145        if (allow_zero || matched_once)        if (allow_zero || matched_once)
1146          {          {
1147          ecode += 1 + LINK_SIZE;          ecode += 1 + LINK_SIZE;
# Line 1022  for (;;) Line 1173  for (;;)
1173    
1174      POSSESSIVE_NON_CAPTURE:      POSSESSIVE_NON_CAPTURE:
1175      matched_once = FALSE;      matched_once = FALSE;
1176      code_offset = ecode - md->start_code;      code_offset = (int)(ecode - md->start_code);
1177    
1178      for (;;)      for (;;)
1179        {        {
1180        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
1181        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
1182          eptrb, RM48);          eptrb, RM48);
1183        if (rrc == MATCH_KETRPOS)        if (rrc == MATCH_KETRPOS)
1184          {          {
# Line 1037  for (;;) Line 1188  for (;;)
1188          matched_once = TRUE;          matched_once = TRUE;
1189          continue;          continue;
1190          }          }
1191        if (rrc != MATCH_NOMATCH &&  
1192            (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* See comment in the code for capturing groups above about handling
1193          RRETURN(rrc);        THEN. */
1194    
1195          if (rrc == MATCH_THEN)
1196            {
1197            next = ecode + GET(ecode,1);
1198            if (md->start_match_ptr < next &&
1199                (*ecode == OP_ALT || *next == OP_ALT))
1200              rrc = MATCH_NOMATCH;
1201            }
1202    
1203          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1204        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1205        if (*ecode != OP_ALT) break;        if (*ecode != OP_ALT) break;
1206        }        }
# Line 1067  for (;;) Line 1228  for (;;)
1228    
1229      if (ecode[LINK_SIZE+1] == OP_CALLOUT)      if (ecode[LINK_SIZE+1] == OP_CALLOUT)
1230        {        {
1231        if (pcre_callout != NULL)        if (PUBL(callout) != NULL)
1232          {          {
1233          pcre_callout_block cb;          PUBL(callout_block) cb;
1234          cb.version          = 1;   /* Version 1 of the callout block */          cb.version          = 2;   /* Version 1 of the callout block */
1235          cb.callout_number   = ecode[LINK_SIZE+2];          cb.callout_number   = ecode[LINK_SIZE+2];
1236          cb.offset_vector    = md->offset_vector;          cb.offset_vector    = md->offset_vector;
1237    #ifdef COMPILE_PCRE8
1238          cb.subject          = (PCRE_SPTR)md->start_subject;          cb.subject          = (PCRE_SPTR)md->start_subject;
1239    #else
1240            cb.subject          = (PCRE_SPTR16)md->start_subject;
1241    #endif
1242          cb.subject_length   = (int)(md->end_subject - md->start_subject);          cb.subject_length   = (int)(md->end_subject - md->start_subject);
1243          cb.start_match      = (int)(mstart - md->start_subject);          cb.start_match      = (int)(mstart - md->start_subject);
1244          cb.current_position = (int)(eptr - md->start_subject);          cb.current_position = (int)(eptr - md->start_subject);
# Line 1082  for (;;) Line 1247  for (;;)
1247          cb.capture_top      = offset_top/2;          cb.capture_top      = offset_top/2;
1248          cb.capture_last     = md->capture_last;          cb.capture_last     = md->capture_last;
1249          cb.callout_data     = md->callout_data;          cb.callout_data     = md->callout_data;
1250          if ((rrc = (*pcre_callout)(&cb)) > 0) MRRETURN(MATCH_NOMATCH);          cb.mark             = md->nomatch_mark;
1251            if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1252          if (rrc < 0) RRETURN(rrc);          if (rrc < 0) RRETURN(rrc);
1253          }          }
1254        ecode += _pcre_OP_lengths[OP_CALLOUT];        ecode += PRIV(OP_lengths)[OP_CALLOUT];
1255        }        }
1256    
1257      condcode = ecode[LINK_SIZE+1];      condcode = ecode[LINK_SIZE+1];
# Line 1102  for (;;) Line 1268  for (;;)
1268        else        else
1269          {          {
1270          int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/          int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/
1271          condition =  (recno == RREF_ANY || recno == md->recursive->group_num);          condition = (recno == RREF_ANY || recno == md->recursive->group_num);
1272    
1273          /* If the test is for recursion into a specific subpattern, and it is          /* If the test is for recursion into a specific subpattern, and it is
1274          false, but the test was set up by name, scan the table to see if the          false, but the test was set up by name, scan the table to see if the
1275          name refers to any other numbers, and test them. The condition is true          name refers to any other numbers, and test them. The condition is true
1276          if any one is set. */          if any one is set. */
1277    
1278          if (!condition && condcode == OP_NRREF && recno != RREF_ANY)          if (!condition && condcode == OP_NRREF)
1279            {            {
1280            uschar *slotA = md->name_table;            pcre_uchar *slotA = md->name_table;
1281            for (i = 0; i < md->name_count; i++)            for (i = 0; i < md->name_count; i++)
1282              {              {
1283              if (GET2(slotA, 0) == recno) break;              if (GET2(slotA, 0) == recno) break;
# Line 1124  for (;;) Line 1290  for (;;)
1290    
1291            if (i < md->name_count)            if (i < md->name_count)
1292              {              {
1293              uschar *slotB = slotA;              pcre_uchar *slotB = slotA;
1294              while (slotB > md->name_table)              while (slotB > md->name_table)
1295                {                {
1296                slotB -= md->name_entry_size;                slotB -= md->name_entry_size;
1297                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1298                  {                  {
1299                  condition = GET2(slotB, 0) == md->recursive->group_num;                  condition = GET2(slotB, 0) == md->recursive->group_num;
1300                  if (condition) break;                  if (condition) break;
# Line 1144  for (;;) Line 1310  for (;;)
1310                for (i++; i < md->name_count; i++)                for (i++; i < md->name_count; i++)
1311                  {                  {
1312                  slotB += md->name_entry_size;                  slotB += md->name_entry_size;
1313                  if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                  if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1314                    {                    {
1315                    condition = GET2(slotB, 0) == md->recursive->group_num;                    condition = GET2(slotB, 0) == md->recursive->group_num;
1316                    if (condition) break;                    if (condition) break;
# Line 1157  for (;;) Line 1323  for (;;)
1323    
1324          /* Chose branch according to the condition */          /* Chose branch according to the condition */
1325    
1326          ecode += condition? 3 : GET(ecode, 1);          ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
1327          }          }
1328        }        }
1329    
# Line 1174  for (;;) Line 1340  for (;;)
1340        if (!condition && condcode == OP_NCREF)        if (!condition && condcode == OP_NCREF)
1341          {          {
1342          int refno = offset >> 1;          int refno = offset >> 1;
1343          uschar *slotA = md->name_table;          pcre_uchar *slotA = md->name_table;
1344    
1345          for (i = 0; i < md->name_count; i++)          for (i = 0; i < md->name_count; i++)
1346            {            {
# Line 1188  for (;;) Line 1354  for (;;)
1354    
1355          if (i < md->name_count)          if (i < md->name_count)
1356            {            {
1357            uschar *slotB = slotA;            pcre_uchar *slotB = slotA;
1358            while (slotB > md->name_table)            while (slotB > md->name_table)
1359              {              {
1360              slotB -= md->name_entry_size;              slotB -= md->name_entry_size;
1361              if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)              if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1362                {                {
1363                offset = GET2(slotB, 0) << 1;                offset = GET2(slotB, 0) << 1;
1364                condition = offset < offset_top &&                condition = offset < offset_top &&
# Line 1210  for (;;) Line 1376  for (;;)
1376              for (i++; i < md->name_count; i++)              for (i++; i < md->name_count; i++)
1377                {                {
1378                slotB += md->name_entry_size;                slotB += md->name_entry_size;
1379                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1380                  {                  {
1381                  offset = GET2(slotB, 0) << 1;                  offset = GET2(slotB, 0) << 1;
1382                  condition = offset < offset_top &&                  condition = offset < offset_top &&
# Line 1225  for (;;) Line 1391  for (;;)
1391    
1392        /* Chose branch according to the condition */        /* Chose branch according to the condition */
1393    
1394        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
1395        }        }
1396    
1397      else if (condcode == OP_DEF)     /* DEFINE - always false */      else if (condcode == OP_DEF)     /* DEFINE - always false */
# Line 1250  for (;;) Line 1416  for (;;)
1416          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
1417          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
1418          }          }
1419        else if (rrc != MATCH_NOMATCH &&  
1420                (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* PCRE doesn't allow the effect of (*THEN) to escape beyond an
1421          assertion; it is therefore treated as NOMATCH. */
1422    
1423          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1424          {          {
1425          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
1426          }          }
# Line 1262  for (;;) Line 1431  for (;;)
1431          }          }
1432        }        }
1433    
1434      /* We are now at the branch that is to be obeyed. As there is only one,      /* We are now at the branch that is to be obeyed. As there is only one, can
1435      we used to use tail recursion to avoid using another stack frame, except      use tail recursion to avoid using another stack frame, except when there is
1436      when there was unlimited repeat of a possibly empty group. However, that      unlimited repeat of a possibly empty group. In the latter case, a recursive
1437      strategy no longer works because of the possibilty of (*THEN) being      call to match() is always required, unless the second alternative doesn't
1438      encountered in the branch. A recursive call to match() is always required,      exist, in which case we can just plough on. Note that, for compatibility
1439      unless the second alternative doesn't exist, in which case we can just      with Perl, the | in a conditional group is NOT treated as creating two
1440      plough on. */      alternatives. If a THEN is encountered in the branch, it propagates out to
1441        the enclosing alternative (unless nested in a deeper set of alternatives,
1442        of course). */
1443    
1444      if (condition || *ecode == OP_ALT)      if (condition || *ecode == OP_ALT)
1445        {        {
1446        if (op == OP_SCOND) md->match_function_type = MATCH_CBEGROUP;        if (op != OP_SCOND)
1447            {
1448            ecode += 1 + LINK_SIZE;
1449            goto TAIL_RECURSE;
1450            }
1451    
1452          md->match_function_type = MATCH_CBEGROUP;
1453        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM49);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM49);
       if (rrc == MATCH_THEN && md->start_match_ptr == ecode)  
         rrc = MATCH_NOMATCH;  
1454        RRETURN(rrc);        RRETURN(rrc);
1455        }        }
1456      else                         /* Condition false & no alternative */  
1457         /* Condition false & no alternative; continue after the group. */
1458    
1459        else
1460        {        {
1461        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1462        }        }
# Line 1305  for (;;) Line 1483  for (;;)
1483        md->offset_vector[offset+1] = (int)(eptr - md->start_subject);        md->offset_vector[offset+1] = (int)(eptr - md->start_subject);
1484        if (offset_top <= offset) offset_top = offset + 2;        if (offset_top <= offset) offset_top = offset + 2;
1485        }        }
1486      ecode += 3;      ecode += 1 + IMM2_SIZE;
1487      break;      break;
1488    
1489    
# Line 1325  for (;;) Line 1503  for (;;)
1503           (md->notempty ||           (md->notempty ||
1504             (md->notempty_atstart &&             (md->notempty_atstart &&
1505               mstart == md->start_subject + md->start_offset)))               mstart == md->start_subject + md->start_offset)))
1506        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
1507    
1508      /* Otherwise, we have a match. */      /* Otherwise, we have a match. */
1509    
# Line 1334  for (;;) Line 1512  for (;;)
1512      md->start_match_ptr = mstart;       /* and the start (\K can modify) */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
1513    
1514      /* For some reason, the macros don't work properly if an expression is      /* For some reason, the macros don't work properly if an expression is
1515      given as the argument to MRRETURN when the heap is in use. */      given as the argument to RRETURN when the heap is in use. */
1516    
1517      rrc = (op == OP_END)? MATCH_MATCH : MATCH_ACCEPT;      rrc = (op == OP_END)? MATCH_MATCH : MATCH_ACCEPT;
1518      MRRETURN(rrc);      RRETURN(rrc);
1519    
1520      /* Assertion brackets. Check the alternative branches in turn - the      /* Assertion brackets. Check the alternative branches in turn - the
1521      matching won't pass the KET for an assertion. If any one branch matches,      matching won't pass the KET for an assertion. If any one branch matches,
# Line 1352  for (;;) Line 1530  for (;;)
1530    
1531      case OP_ASSERT:      case OP_ASSERT:
1532      case OP_ASSERTBACK:      case OP_ASSERTBACK:
1533        save_mark = md->mark;
1534      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1535        {        {
1536        condassert = TRUE;        condassert = TRUE;
# Line 1365  for (;;) Line 1544  for (;;)
1544        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
1545          {          {
1546          mstart = md->start_match_ptr;   /* In case \K reset it */          mstart = md->start_match_ptr;   /* In case \K reset it */
         markptr = md->mark;  
1547          break;          break;
1548          }          }
1549        if (rrc != MATCH_NOMATCH &&  
1550            (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* PCRE does not allow THEN to escape beyond an assertion; it is treated
1551          RRETURN(rrc);        as NOMATCH. */
1552    
1553          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1554        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1555          md->mark = save_mark;
1556        }        }
1557      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
1558    
1559      if (*ecode == OP_KET) MRRETURN(MATCH_NOMATCH);      if (*ecode == OP_KET) RRETURN(MATCH_NOMATCH);
1560    
1561      /* If checking an assertion for a condition, return MATCH_MATCH. */      /* If checking an assertion for a condition, return MATCH_MATCH. */
1562    
# Line 1395  for (;;) Line 1576  for (;;)
1576    
1577      case OP_ASSERT_NOT:      case OP_ASSERT_NOT:
1578      case OP_ASSERTBACK_NOT:      case OP_ASSERTBACK_NOT:
1579        save_mark = md->mark;
1580      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1581        {        {
1582        condassert = TRUE;        condassert = TRUE;
# Line 1405  for (;;) Line 1587  for (;;)
1587      do      do
1588        {        {
1589        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);
1590        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) MRRETURN(MATCH_NOMATCH);        md->mark = save_mark;
1591          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) RRETURN(MATCH_NOMATCH);
1592        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)
1593          {          {
1594          do ecode += GET(ecode,1); while (*ecode == OP_ALT);          do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1595          break;          break;
1596          }          }
1597        if (rrc != MATCH_NOMATCH &&  
1598            (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* PCRE does not allow THEN to escape beyond an assertion; it is treated
1599          RRETURN(rrc);        as NOMATCH. */
1600    
1601          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1602        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1603        }        }
1604      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 1429  for (;;) Line 1614  for (;;)
1614      back a number of characters, not bytes. */      back a number of characters, not bytes. */
1615    
1616      case OP_REVERSE:      case OP_REVERSE:
1617  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
1618      if (utf8)      if (utf)
1619        {        {
1620        i = GET(ecode, 1);        i = GET(ecode, 1);
1621        while (i-- > 0)        while (i-- > 0)
1622          {          {
1623          eptr--;          eptr--;
1624          if (eptr < md->start_subject) MRRETURN(MATCH_NOMATCH);          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1625          BACKCHAR(eptr);          BACKCHAR(eptr);
1626          }          }
1627        }        }
# Line 1447  for (;;) Line 1632  for (;;)
1632    
1633        {        {
1634        eptr -= GET(ecode, 1);        eptr -= GET(ecode, 1);
1635        if (eptr < md->start_subject) MRRETURN(MATCH_NOMATCH);        if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1636        }        }
1637    
1638      /* Save the earliest consulted character, then skip to next op code */      /* Save the earliest consulted character, then skip to next op code */
# Line 1461  for (;;) Line 1646  for (;;)
1646      function is able to force a failure. */      function is able to force a failure. */
1647    
1648      case OP_CALLOUT:      case OP_CALLOUT:
1649      if (pcre_callout != NULL)      if (PUBL(callout) != NULL)
1650        {        {
1651        pcre_callout_block cb;        PUBL(callout_block) cb;
1652        cb.version          = 1;   /* Version 1 of the callout block */        cb.version          = 2;   /* Version 1 of the callout block */
1653        cb.callout_number   = ecode[1];        cb.callout_number   = ecode[1];
1654        cb.offset_vector    = md->offset_vector;        cb.offset_vector    = md->offset_vector;
1655    #ifdef COMPILE_PCRE8
1656        cb.subject          = (PCRE_SPTR)md->start_subject;        cb.subject          = (PCRE_SPTR)md->start_subject;
1657    #else
1658          cb.subject          = (PCRE_SPTR16)md->start_subject;
1659    #endif
1660        cb.subject_length   = (int)(md->end_subject - md->start_subject);        cb.subject_length   = (int)(md->end_subject - md->start_subject);
1661        cb.start_match      = (int)(mstart - md->start_subject);        cb.start_match      = (int)(mstart - md->start_subject);
1662        cb.current_position = (int)(eptr - md->start_subject);        cb.current_position = (int)(eptr - md->start_subject);
# Line 1476  for (;;) Line 1665  for (;;)
1665        cb.capture_top      = offset_top/2;        cb.capture_top      = offset_top/2;
1666        cb.capture_last     = md->capture_last;        cb.capture_last     = md->capture_last;
1667        cb.callout_data     = md->callout_data;        cb.callout_data     = md->callout_data;
1668        if ((rrc = (*pcre_callout)(&cb)) > 0) MRRETURN(MATCH_NOMATCH);        cb.mark             = md->nomatch_mark;
1669          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1670        if (rrc < 0) RRETURN(rrc);        if (rrc < 0) RRETURN(rrc);
1671        }        }
1672      ecode += 2 + 2*LINK_SIZE;      ecode += 2 + 2*LINK_SIZE;
# Line 1503  for (;;) Line 1693  for (;;)
1693        {        {
1694        recursion_info *ri;        recursion_info *ri;
1695        int recno;        int recno;
1696    
1697        callpat = md->start_code + GET(ecode, 1);        callpat = md->start_code + GET(ecode, 1);
1698        recno = (callpat == md->start_code)? 0 :        recno = (callpat == md->start_code)? 0 :
1699          GET2(callpat, 1 + LINK_SIZE);          GET2(callpat, 1 + LINK_SIZE);
1700    
1701        /* Check for repeating a recursion without advancing the subject pointer.        /* Check for repeating a recursion without advancing the subject pointer.
1702        This should catch convoluted mutual recursions. (Some simple cases are        This should catch convoluted mutual recursions. (Some simple cases are
1703        caught at compile time.) */        caught at compile time.) */
1704    
1705        for (ri = md->recursive; ri != NULL; ri = ri->prevrec)        for (ri = md->recursive; ri != NULL; ri = ri->prevrec)
1706          if (recno == ri->group_num && eptr == ri->subject_position)          if (recno == ri->group_num && eptr == ri->subject_position)
1707            RRETURN(PCRE_ERROR_RECURSELOOP);            RRETURN(PCRE_ERROR_RECURSELOOP);
1708    
1709        /* Add to "recursing stack" */        /* Add to "recursing stack" */
# Line 1535  for (;;) Line 1725  for (;;)
1725        else        else
1726          {          {
1727          new_recursive.offset_save =          new_recursive.offset_save =
1728            (int *)(pcre_malloc)(new_recursive.saved_max * sizeof(int));            (int *)(PUBL(malloc))(new_recursive.saved_max * sizeof(int));
1729          if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY);          if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY);
1730          }          }
1731        memcpy(new_recursive.offset_save, md->offset_vector,        memcpy(new_recursive.offset_save, md->offset_vector,
# Line 1550  for (;;) Line 1740  for (;;)
1740        do        do
1741          {          {
1742          if (cbegroup) md->match_function_type = MATCH_CBEGROUP;          if (cbegroup) md->match_function_type = MATCH_CBEGROUP;
1743          RMATCH(eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,          RMATCH(eptr, callpat + PRIV(OP_lengths)[*callpat], offset_top,
1744            md, eptrb, RM6);            md, eptrb, RM6);
1745          memcpy(md->offset_vector, new_recursive.offset_save,          memcpy(md->offset_vector, new_recursive.offset_save,
1746              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
1747            md->recursive = new_recursive.prevrec;
1748          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
1749            {            {
1750            DPRINTF(("Recursion matched\n"));            DPRINTF(("Recursion matched\n"));
           md->recursive = new_recursive.prevrec;  
1751            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
1752              (pcre_free)(new_recursive.offset_save);              (PUBL(free))(new_recursive.offset_save);
1753    
1754            /* 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
1755            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 1569  for (;;) Line 1759  for (;;)
1759            mstart = md->start_match_ptr;            mstart = md->start_match_ptr;
1760            goto RECURSION_MATCHED;        /* Exit loop; end processing */            goto RECURSION_MATCHED;        /* Exit loop; end processing */
1761            }            }
1762          else if (rrc != MATCH_NOMATCH &&  
1763                  (rrc != MATCH_THEN || md->start_match_ptr != ecode))          /* PCRE does not allow THEN to escape beyond a recursion; it is treated
1764            as NOMATCH. */
1765    
1766            else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1767            {            {
1768            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1769            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
1770              (pcre_free)(new_recursive.offset_save);              (PUBL(free))(new_recursive.offset_save);
1771            RRETURN(rrc);            RRETURN(rrc);
1772            }            }
1773    
# Line 1586  for (;;) Line 1779  for (;;)
1779        DPRINTF(("Recursion didn't match\n"));        DPRINTF(("Recursion didn't match\n"));
1780        md->recursive = new_recursive.prevrec;        md->recursive = new_recursive.prevrec;
1781        if (new_recursive.offset_save != stacksave)        if (new_recursive.offset_save != stacksave)
1782          (pcre_free)(new_recursive.offset_save);          (PUBL(free))(new_recursive.offset_save);
1783        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
1784        }        }
1785    
1786      RECURSION_MATCHED:      RECURSION_MATCHED:
# Line 1656  for (;;) Line 1849  for (;;)
1849        }        }
1850      else saved_eptr = NULL;      else saved_eptr = NULL;
1851    
1852      /* If we are at the end of an assertion group, stop matching and return      /* If we are at the end of an assertion group or a non-capturing atomic
1853      MATCH_MATCH, but record the current high water mark for use by positive      group, stop matching and return MATCH_MATCH, but record the current high
1854      assertions. We also need to record the match start in case it was changed      water mark for use by positive assertions. We also need to record the match
1855      by \K. */      start in case it was changed by \K. */
1856    
1857      if (*prev == OP_ASSERT || *prev == OP_ASSERT_NOT ||      if ((*prev >= OP_ASSERT && *prev <= OP_ASSERTBACK_NOT) ||
1858          *prev == OP_ASSERTBACK || *prev == OP_ASSERTBACK_NOT)           *prev == OP_ONCE_NC)
1859        {        {
1860        md->end_match_ptr = eptr;      /* For ONCE */        md->end_match_ptr = eptr;      /* For ONCE_NC */
1861        md->end_offset_top = offset_top;        md->end_offset_top = offset_top;
1862        md->start_match_ptr = mstart;        md->start_match_ptr = mstart;
1863        MRRETURN(MATCH_MATCH);         /* Sets md->mark */        RRETURN(MATCH_MATCH);         /* Sets md->mark */
1864        }        }
1865    
1866      /* For capturing groups we have to check the group number back at the start      /* For capturing groups we have to check the group number back at the start
# Line 1732  for (;;) Line 1925  for (;;)
1925      /* For an ordinary non-repeating ket, just continue at this level. This      /* For an ordinary non-repeating ket, just continue at this level. This
1926      also happens for a repeating ket if no characters were matched in the      also happens for a repeating ket if no characters were matched in the
1927      group. This is the forcible breaking of infinite loops as implemented in      group. This is the forcible breaking of infinite loops as implemented in
1928      Perl 5.005. For a non-repeating atomic group, establish a backup point by      Perl 5.005. For a non-repeating atomic group that includes captures,
1929      processing the rest of the pattern at a lower level. If this results in a      establish a backup point by processing the rest of the pattern at a lower
1930      NOMATCH return, pass MATCH_ONCE back to the original OP_ONCE level, thereby      level. If this results in a NOMATCH return, pass MATCH_ONCE back to the
1931      bypassing intermediate backup points, but resetting any captures that      original OP_ONCE level, thereby bypassing intermediate backup points, but
1932      happened along the way. */      resetting any captures that happened along the way. */
1933    
1934      if (*ecode == OP_KET || eptr == saved_eptr)      if (*ecode == OP_KET || eptr == saved_eptr)
1935        {        {
# Line 1809  for (;;) Line 2002  for (;;)
2002      /* Not multiline mode: start of subject assertion, unless notbol. */      /* Not multiline mode: start of subject assertion, unless notbol. */
2003    
2004      case OP_CIRC:      case OP_CIRC:
2005      if (md->notbol && eptr == md->start_subject) MRRETURN(MATCH_NOMATCH);      if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH);
2006    
2007      /* Start of subject assertion */      /* Start of subject assertion */
2008    
2009      case OP_SOD:      case OP_SOD:
2010      if (eptr != md->start_subject) MRRETURN(MATCH_NOMATCH);      if (eptr != md->start_subject) RRETURN(MATCH_NOMATCH);
2011      ecode++;      ecode++;
2012      break;      break;
2013    
2014      /* Multiline mode: start of subject unless notbol, or after any newline. */      /* Multiline mode: start of subject unless notbol, or after any newline. */
2015    
2016      case OP_CIRCM:      case OP_CIRCM:
2017      if (md->notbol && eptr == md->start_subject) MRRETURN(MATCH_NOMATCH);      if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH);
2018      if (eptr != md->start_subject &&      if (eptr != md->start_subject &&
2019          (eptr == md->end_subject || !WAS_NEWLINE(eptr)))          (eptr == md->end_subject || !WAS_NEWLINE(eptr)))
2020        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2021      ecode++;      ecode++;
2022      break;      break;
2023    
2024      /* Start of match assertion */      /* Start of match assertion */
2025    
2026      case OP_SOM:      case OP_SOM:
2027      if (eptr != md->start_subject + md->start_offset) MRRETURN(MATCH_NOMATCH);      if (eptr != md->start_subject + md->start_offset) RRETURN(MATCH_NOMATCH);
2028      ecode++;      ecode++;
2029      break;      break;
2030    
# Line 1847  for (;;) Line 2040  for (;;)
2040    
2041      case OP_DOLLM:      case OP_DOLLM:
2042      if (eptr < md->end_subject)      if (eptr < md->end_subject)
2043        { if (!IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH); }        { if (!IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); }
2044      else      else
2045        {        {
2046        if (md->noteol) MRRETURN(MATCH_NOMATCH);        if (md->noteol) RRETURN(MATCH_NOMATCH);
2047        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2048        }        }
2049      ecode++;      ecode++;
# Line 1860  for (;;) Line 2053  for (;;)
2053      subject unless noteol is set. */      subject unless noteol is set. */
2054    
2055      case OP_DOLL:      case OP_DOLL:
2056      if (md->noteol) MRRETURN(MATCH_NOMATCH);      if (md->noteol) RRETURN(MATCH_NOMATCH);
2057      if (!md->endonly) goto ASSERT_NL_OR_EOS;      if (!md->endonly) goto ASSERT_NL_OR_EOS;
2058    
2059      /* ... else fall through for endonly */      /* ... else fall through for endonly */
# Line 1868  for (;;) Line 2061  for (;;)
2061      /* End of subject assertion (\z) */      /* End of subject assertion (\z) */
2062    
2063      case OP_EOD:      case OP_EOD:
2064      if (eptr < md->end_subject) MRRETURN(MATCH_NOMATCH);      if (eptr < md->end_subject) RRETURN(MATCH_NOMATCH);
2065      SCHECK_PARTIAL();      SCHECK_PARTIAL();
2066      ecode++;      ecode++;
2067      break;      break;
# Line 1879  for (;;) Line 2072  for (;;)
2072      ASSERT_NL_OR_EOS:      ASSERT_NL_OR_EOS:
2073      if (eptr < md->end_subject &&      if (eptr < md->end_subject &&
2074          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))
2075        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2076    
2077      /* Either at end of string or \n before end. */      /* Either at end of string or \n before end. */
2078    
# Line 1898  for (;;) Line 2091  for (;;)
2091        be "non-word" characters. Remember the earliest consulted character for        be "non-word" characters. Remember the earliest consulted character for
2092        partial matching. */        partial matching. */
2093    
2094  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2095        if (utf8)        if (utf)
2096          {          {
2097          /* Get status of previous character */          /* Get status of previous character */
2098    
2099          if (eptr == md->start_subject) prev_is_word = FALSE; else          if (eptr == md->start_subject) prev_is_word = FALSE; else
2100            {            {
2101            USPTR lastptr = eptr - 1;            PCRE_PUCHAR lastptr = eptr - 1;
2102            while((*lastptr & 0xc0) == 0x80) lastptr--;            BACKCHAR(lastptr);
2103            if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;            if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;
2104            GETCHAR(c, lastptr);            GETCHAR(c, lastptr);
2105  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 1971  for (;;) Line 2164  for (;;)
2164              }              }
2165            else            else
2166  #endif  #endif
2167            prev_is_word = ((md->ctypes[eptr[-1]] & ctype_word) != 0);            prev_is_word = MAX_255(eptr[-1])
2168                && ((md->ctypes[eptr[-1]] & ctype_word) != 0);
2169            }            }
2170    
2171          /* Get status of next character */          /* Get status of next character */
# Line 1994  for (;;) Line 2188  for (;;)
2188            }            }
2189          else          else
2190  #endif  #endif
2191          cur_is_word = ((md->ctypes[*eptr] & ctype_word) != 0);          cur_is_word = MAX_255(*eptr)
2192              && ((md->ctypes[*eptr] & ctype_word) != 0);
2193          }          }
2194    
2195        /* Now see if the situation is what we want */        /* Now see if the situation is what we want */
2196    
2197        if ((*ecode++ == OP_WORD_BOUNDARY)?        if ((*ecode++ == OP_WORD_BOUNDARY)?
2198             cur_is_word == prev_is_word : cur_is_word != prev_is_word)             cur_is_word == prev_is_word : cur_is_word != prev_is_word)
2199          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2200        }        }
2201      break;      break;
2202    
2203      /* Match a single character type; inline for speed */      /* Match a single character type; inline for speed */
2204    
2205      case OP_ANY:      case OP_ANY:
2206      if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
2207      /* Fall through */      /* Fall through */
2208    
2209      case OP_ALLANY:      case OP_ALLANY:
2210      if (eptr++ >= md->end_subject)      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */
2211        {        {                            /* not be updated before SCHECK_PARTIAL. */
2212        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2213        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2214        }        }
2215      if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;      eptr++;
2216    #ifdef SUPPORT_UTF
2217        if (utf) ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
2218    #endif
2219      ecode++;      ecode++;
2220      break;      break;
2221    
# Line 2025  for (;;) Line 2223  for (;;)
2223      any byte, even newline, independent of the setting of PCRE_DOTALL. */      any byte, even newline, independent of the setting of PCRE_DOTALL. */
2224    
2225      case OP_ANYBYTE:      case OP_ANYBYTE:
2226      if (eptr++ >= md->end_subject)      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */
2227        {        {                            /* not be updated before SCHECK_PARTIAL. */
2228        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2229        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2230        }        }
2231        eptr++;
2232      ecode++;      ecode++;
2233      break;      break;
2234    
# Line 2037  for (;;) Line 2236  for (;;)
2236      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2237        {        {
2238        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2239        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2240        }        }
2241      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2242      if (      if (
2243  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2244         c < 256 &&         c < 256 &&
2245  #endif  #endif
2246         (md->ctypes[c] & ctype_digit) != 0         (md->ctypes[c] & ctype_digit) != 0
2247         )         )
2248        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2249      ecode++;      ecode++;
2250      break;      break;
2251    
# Line 2054  for (;;) Line 2253  for (;;)
2253      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2254        {        {
2255        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2256        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2257        }        }
2258      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2259      if (      if (
2260  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2261         c >= 256 ||         c > 255 ||
2262  #endif  #endif
2263         (md->ctypes[c] & ctype_digit) == 0         (md->ctypes[c] & ctype_digit) == 0
2264         )         )
2265        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2266      ecode++;      ecode++;
2267      break;      break;
2268    
# Line 2071  for (;;) Line 2270  for (;;)
2270      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2271        {        {
2272        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2273        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2274        }        }
2275      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2276      if (      if (
2277  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2278         c < 256 &&         c < 256 &&
2279  #endif  #endif
2280         (md->ctypes[c] & ctype_space) != 0         (md->ctypes[c] & ctype_space) != 0
2281         )         )
2282        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2283      ecode++;      ecode++;
2284      break;      break;
2285    
# Line 2088  for (;;) Line 2287  for (;;)
2287      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2288        {        {
2289        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2290        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2291        }        }
2292      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2293      if (      if (
2294  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2295         c >= 256 ||         c > 255 ||
2296  #endif  #endif
2297         (md->ctypes[c] & ctype_space) == 0         (md->ctypes[c] & ctype_space) == 0
2298         )         )
2299        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2300      ecode++;      ecode++;
2301      break;      break;
2302    
# Line 2105  for (;;) Line 2304  for (;;)
2304      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2305        {        {
2306        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2307        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2308        }        }
2309      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2310      if (      if (
2311  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2312         c < 256 &&         c < 256 &&
2313  #endif  #endif
2314         (md->ctypes[c] & ctype_word) != 0         (md->ctypes[c] & ctype_word) != 0
2315         )         )
2316        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2317      ecode++;      ecode++;
2318      break;      break;
2319    
# Line 2122  for (;;) Line 2321  for (;;)
2321      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2322        {        {
2323        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2324        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2325        }        }
2326      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2327      if (      if (
2328  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2329         c >= 256 ||         c > 255 ||
2330  #endif  #endif
2331         (md->ctypes[c] & ctype_word) == 0         (md->ctypes[c] & ctype_word) == 0
2332         )         )
2333        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2334      ecode++;      ecode++;
2335      break;      break;
2336    
# Line 2139  for (;;) Line 2338  for (;;)
2338      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2339        {        {
2340        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2341        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2342        }        }
2343      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2344      switch(c)      switch(c)
2345        {        {
2346        default: MRRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
2347    
2348        case 0x000d:        case 0x000d:
2349        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
# Line 2158  for (;;) Line 2357  for (;;)
2357        case 0x0085:        case 0x0085:
2358        case 0x2028:        case 0x2028:
2359        case 0x2029:        case 0x2029:
2360        if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);        if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
2361        break;        break;
2362        }        }
2363      ecode++;      ecode++;
# Line 2168  for (;;) Line 2367  for (;;)
2367      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2368        {        {
2369        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2370        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2371        }        }
2372      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2373      switch(c)      switch(c)
# Line 2193  for (;;) Line 2392  for (;;)
2392        case 0x202f:    /* NARROW NO-BREAK SPACE */        case 0x202f:    /* NARROW NO-BREAK SPACE */
2393        case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */        case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
2394        case 0x3000:    /* IDEOGRAPHIC SPACE */        case 0x3000:    /* IDEOGRAPHIC SPACE */
2395        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2396        }        }
2397      ecode++;      ecode++;
2398      break;      break;
# Line 2202  for (;;) Line 2401  for (;;)
2401      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2402        {        {
2403        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2404        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2405        }        }
2406      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2407      switch(c)      switch(c)
2408        {        {
2409        default: MRRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
2410        case 0x09:      /* HT */        case 0x09:      /* HT */
2411        case 0x20:      /* SPACE */        case 0x20:      /* SPACE */
2412        case 0xa0:      /* NBSP */        case 0xa0:      /* NBSP */
# Line 2236  for (;;) Line 2435  for (;;)
2435      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2436        {        {
2437        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2438        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2439        }        }
2440      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2441      switch(c)      switch(c)
# Line 2249  for (;;) Line 2448  for (;;)
2448        case 0x85:      /* NEL */        case 0x85:      /* NEL */
2449        case 0x2028:    /* LINE SEPARATOR */        case 0x2028:    /* LINE SEPARATOR */
2450        case 0x2029:    /* PARAGRAPH SEPARATOR */        case 0x2029:    /* PARAGRAPH SEPARATOR */
2451        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2452        }        }
2453      ecode++;      ecode++;
2454      break;      break;
# Line 2258  for (;;) Line 2457  for (;;)
2457      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2458        {        {
2459        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2460        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2461        }        }
2462      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2463      switch(c)      switch(c)
2464        {        {
2465        default: MRRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
2466        case 0x0a:      /* LF */        case 0x0a:      /* LF */
2467        case 0x0b:      /* VT */        case 0x0b:      /* VT */
2468        case 0x0c:      /* FF */        case 0x0c:      /* FF */
# Line 2285  for (;;) Line 2484  for (;;)
2484      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2485        {        {
2486        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2487        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2488        }        }
2489      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2490        {        {
# Line 2294  for (;;) Line 2493  for (;;)
2493        switch(ecode[1])        switch(ecode[1])
2494          {          {
2495          case PT_ANY:          case PT_ANY:
2496          if (op == OP_NOTPROP) MRRETURN(MATCH_NOMATCH);          if (op == OP_NOTPROP) RRETURN(MATCH_NOMATCH);
2497          break;          break;
2498    
2499          case PT_LAMP:          case PT_LAMP:
2500          if ((prop->chartype == ucp_Lu ||          if ((prop->chartype == ucp_Lu ||
2501               prop->chartype == ucp_Ll ||               prop->chartype == ucp_Ll ||
2502               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))
2503            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2504          break;          break;
2505    
2506          case PT_GC:          case PT_GC:
2507          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))          if ((ecode[2] != PRIV(ucp_gentype)[prop->chartype]) == (op == OP_PROP))
2508            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2509          break;          break;
2510    
2511          case PT_PC:          case PT_PC:
2512          if ((ecode[2] != prop->chartype) == (op == OP_PROP))          if ((ecode[2] != prop->chartype) == (op == OP_PROP))
2513            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2514          break;          break;
2515    
2516          case PT_SC:          case PT_SC:
2517          if ((ecode[2] != prop->script) == (op == OP_PROP))          if ((ecode[2] != prop->script) == (op == OP_PROP))
2518            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2519          break;          break;
2520    
2521          /* These are specials */          /* These are specials */
2522    
2523          case PT_ALNUM:          case PT_ALNUM:
2524          if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
2525               _pcre_ucp_gentype[prop->chartype] == ucp_N) == (op == OP_NOTPROP))               PRIV(ucp_gentype)[prop->chartype] == ucp_N) == (op == OP_NOTPROP))
2526            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2527          break;          break;
2528    
2529          case PT_SPACE:    /* Perl space */          case PT_SPACE:    /* Perl space */
2530          if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
2531               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
2532                 == (op == OP_NOTPROP))                 == (op == OP_NOTPROP))
2533            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2534          break;          break;
2535    
2536          case PT_PXSPACE:  /* POSIX space */          case PT_PXSPACE:  /* POSIX space */
2537          if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
2538               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
2539               c == CHAR_FF || c == CHAR_CR)               c == CHAR_FF || c == CHAR_CR)
2540                 == (op == OP_NOTPROP))                 == (op == OP_NOTPROP))
2541            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2542          break;          break;
2543    
2544          case PT_WORD:          case PT_WORD:
2545          if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
2546               _pcre_ucp_gentype[prop->chartype] == ucp_N ||               PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
2547               c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))               c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))
2548            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2549          break;          break;
2550    
2551          /* This should never occur */          /* This should never occur */
# Line 2366  for (;;) Line 2565  for (;;)
2565      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2566        {        {
2567        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2568        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2569        }        }
2570      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2571      if (UCD_CATEGORY(c) == ucp_M) MRRETURN(MATCH_NOMATCH);      if (UCD_CATEGORY(c) == ucp_M) RRETURN(MATCH_NOMATCH);
2572      while (eptr < md->end_subject)      while (eptr < md->end_subject)
2573        {        {
2574        int len = 1;        int len = 1;
2575        if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }        if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
2576        if (UCD_CATEGORY(c) != ucp_M) break;        if (UCD_CATEGORY(c) != ucp_M) break;
2577        eptr += len;        eptr += len;
2578        }        }
# Line 2394  for (;;) Line 2593  for (;;)
2593      case OP_REFI:      case OP_REFI:
2594      caseless = op == OP_REFI;      caseless = op == OP_REFI;
2595      offset = GET2(ecode, 1) << 1;               /* Doubled ref number */      offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
2596      ecode += 3;      ecode += 1 + IMM2_SIZE;
2597    
2598      /* If the reference is unset, there are two possibilities:      /* If the reference is unset, there are two possibilities:
2599    
# Line 2434  for (;;) Line 2633  for (;;)
2633        case OP_CRMINRANGE:        case OP_CRMINRANGE:
2634        minimize = (*ecode == OP_CRMINRANGE);        minimize = (*ecode == OP_CRMINRANGE);
2635        min = GET2(ecode, 1);        min = GET2(ecode, 1);
2636        max = GET2(ecode, 3);        max = GET2(ecode, 1 + IMM2_SIZE);
2637        if (max == 0) max = INT_MAX;        if (max == 0) max = INT_MAX;
2638        ecode += 5;        ecode += 1 + 2 * IMM2_SIZE;
2639        break;        break;
2640    
2641        default:               /* No repeat follows */        default:               /* No repeat follows */
2642        if ((length = match_ref(offset, eptr, length, md, caseless)) < 0)        if ((length = match_ref(offset, eptr, length, md, caseless)) < 0)
2643          {          {
2644          CHECK_PARTIAL();          CHECK_PARTIAL();
2645          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2646          }          }
2647        eptr += length;        eptr += length;
2648        continue;              /* With the main loop */        continue;              /* With the main loop */
2649        }        }
2650    
2651      /* Handle repeated back references. If the length of the reference is      /* Handle repeated back references. If the length of the reference is
2652      zero, just continue with the main loop. */      zero, just continue with the main loop. If the length is negative, it
2653        means the reference is unset in non-Java-compatible mode. If the minimum is
2654        zero, we can continue at the same level without recursion. For any other
2655        minimum, carrying on will result in NOMATCH. */
2656    
2657      if (length == 0) continue;      if (length == 0) continue;
2658        if (length < 0 && min == 0) continue;
2659    
2660      /* First, ensure the minimum number of matches are present. We get back      /* First, ensure the minimum number of matches are present. We get back
2661      the length of the reference string explicitly rather than passing the      the length of the reference string explicitly rather than passing the
# Line 2464  for (;;) Line 2667  for (;;)
2667        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2668          {          {
2669          CHECK_PARTIAL();          CHECK_PARTIAL();
2670          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2671          }          }
2672        eptr += slength;        eptr += slength;
2673        }        }
# Line 2483  for (;;) Line 2686  for (;;)
2686          int slength;          int slength;
2687          RMATCH(eptr, ecode, offset_top, md, eptrb, RM14);          RMATCH(eptr, ecode, offset_top, md, eptrb, RM14);
2688          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2689          if (fi >= max) MRRETURN(MATCH_NOMATCH);          if (fi >= max) RRETURN(MATCH_NOMATCH);
2690          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2691            {            {
2692            CHECK_PARTIAL();            CHECK_PARTIAL();
2693            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2694            }            }
2695          eptr += slength;          eptr += slength;
2696          }          }
# Line 2515  for (;;) Line 2718  for (;;)
2718          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2719          eptr -= length;          eptr -= length;
2720          }          }
2721        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2722        }        }
2723      /* Control never gets here */      /* Control never gets here */
2724    
# Line 2533  for (;;) Line 2736  for (;;)
2736      case OP_NCLASS:      case OP_NCLASS:
2737      case OP_CLASS:      case OP_CLASS:
2738        {        {
2739          /* The data variable is saved across frames, so the byte map needs to
2740          be stored there. */
2741    #define BYTE_MAP ((pcre_uint8 *)data)
2742        data = ecode + 1;                /* Save for matching */        data = ecode + 1;                /* Save for matching */
2743        ecode += 33;                     /* Advance past the item */        ecode += 1 + (32 / sizeof(pcre_uchar)); /* Advance past the item */
2744    
2745        switch (*ecode)        switch (*ecode)
2746          {          {
# Line 2555  for (;;) Line 2761  for (;;)
2761          case OP_CRMINRANGE:          case OP_CRMINRANGE:
2762          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
2763          min = GET2(ecode, 1);          min = GET2(ecode, 1);
2764          max = GET2(ecode, 3);          max = GET2(ecode, 1 + IMM2_SIZE);
2765          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
2766          ecode += 5;          ecode += 1 + 2 * IMM2_SIZE;
2767          break;          break;
2768    
2769          default:               /* No repeat follows */          default:               /* No repeat follows */
# Line 2567  for (;;) Line 2773  for (;;)
2773    
2774        /* First, ensure the minimum number of matches are present. */        /* First, ensure the minimum number of matches are present. */
2775    
2776  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2777        /* UTF-8 mode */        if (utf)
       if (utf8)  
2778          {          {
2779          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2780            {            {
2781            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
2782              {              {
2783              SCHECK_PARTIAL();              SCHECK_PARTIAL();
2784              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2785              }              }
2786            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
2787            if (c > 255)            if (c > 255)
2788              {              {
2789              if (op == OP_CLASS) MRRETURN(MATCH_NOMATCH);              if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2790              }              }
2791            else            else
2792              {              if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
             if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  
             }  
2793            }            }
2794          }          }
2795        else        else
2796  #endif  #endif
2797        /* Not UTF-8 mode */        /* Not UTF mode */
2798          {          {
2799          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2800            {            {
2801            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
2802              {              {
2803              SCHECK_PARTIAL();              SCHECK_PARTIAL();
2804              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2805              }              }
2806            c = *eptr++;            c = *eptr++;
2807            if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  #ifndef COMPILE_PCRE8
2808              if (c > 255)
2809                {
2810                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2811                }
2812              else
2813    #endif
2814                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2815            }            }
2816          }          }
2817    
# Line 2615  for (;;) Line 2825  for (;;)
2825    
2826        if (minimize)        if (minimize)
2827          {          {
2828  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2829          /* UTF-8 mode */          if (utf)
         if (utf8)  
2830            {            {
2831            for (fi = min;; fi++)            for (fi = min;; fi++)
2832              {              {
2833              RMATCH(eptr, ecode, offset_top, md, eptrb, RM16);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM16);
2834              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2835              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2836              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
2837                {                {
2838                SCHECK_PARTIAL();                SCHECK_PARTIAL();
2839                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2840                }                }
2841              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
2842              if (c > 255)              if (c > 255)
2843                {                {
2844                if (op == OP_CLASS) MRRETURN(MATCH_NOMATCH);                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2845                }                }
2846              else              else
2847                {                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
               if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  
               }  
2848              }              }
2849            }            }
2850          else          else
2851  #endif  #endif
2852          /* Not UTF-8 mode */          /* Not UTF mode */
2853            {            {
2854            for (fi = min;; fi++)            for (fi = min;; fi++)
2855              {              {
2856              RMATCH(eptr, ecode, offset_top, md, eptrb, RM17);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM17);
2857              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2858              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2859              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
2860                {                {
2861                SCHECK_PARTIAL();                SCHECK_PARTIAL();
2862                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2863                }                }
2864              c = *eptr++;              c = *eptr++;
2865              if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  #ifndef COMPILE_PCRE8
2866                if (c > 255)
2867                  {
2868                  if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2869                  }
2870                else
2871    #endif
2872                  if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2873              }              }
2874            }            }
2875          /* Control never gets here */          /* Control never gets here */
# Line 2667  for (;;) Line 2881  for (;;)
2881          {          {
2882          pp = eptr;          pp = eptr;
2883    
2884  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2885          /* UTF-8 mode */          if (utf)
         if (utf8)  
2886            {            {
2887            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2888              {              {
# Line 2685  for (;;) Line 2898  for (;;)
2898                if (op == OP_CLASS) break;                if (op == OP_CLASS) break;
2899                }                }
2900              else              else
2901                {                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
               if ((data[c/8] & (1 << (c&7))) == 0) break;  
               }  
2902              eptr += len;              eptr += len;
2903              }              }
2904            for (;;)            for (;;)
# Line 2700  for (;;) Line 2911  for (;;)
2911            }            }
2912          else          else
2913  #endif  #endif
2914            /* Not UTF-8 mode */            /* Not UTF mode */
2915            {            {
2916            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2917              {              {
# Line 2710  for (;;) Line 2921  for (;;)
2921                break;                break;
2922                }                }
2923              c = *eptr;              c = *eptr;
2924              if ((data[c/8] & (1 << (c&7))) == 0) break;  #ifndef COMPILE_PCRE8
2925                if (c > 255)
2926                  {
2927                  if (op == OP_CLASS) break;
2928                  }
2929                else
2930    #endif
2931                  if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
2932              eptr++;              eptr++;
2933              }              }
2934            while (eptr >= pp)            while (eptr >= pp)
# Line 2721  for (;;) Line 2939  for (;;)
2939              }              }
2940            }            }
2941    
2942          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2943          }          }
2944    #undef BYTE_MAP
2945        }        }
2946      /* Control never gets here */      /* Control never gets here */
2947    
# Line 2731  for (;;) Line 2950  for (;;)
2950      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
2951      mode, because Unicode properties are supported in non-UTF-8 mode. */      mode, because Unicode properties are supported in non-UTF-8 mode. */
2952    
2953  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
2954      case OP_XCLASS:      case OP_XCLASS:
2955        {        {
2956        data = ecode + 1 + LINK_SIZE;                /* Save for matching */        data = ecode + 1 + LINK_SIZE;                /* Save for matching */
# Line 2756  for (;;) Line 2975  for (;;)
2975          case OP_CRMINRANGE:          case OP_CRMINRANGE:
2976          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
2977          min = GET2(ecode, 1);          min = GET2(ecode, 1);
2978          max = GET2(ecode, 3);          max = GET2(ecode, 1 + IMM2_SIZE);
2979          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
2980          ecode += 5;          ecode += 1 + 2 * IMM2_SIZE;
2981          break;          break;
2982    
2983          default:               /* No repeat follows */          default:               /* No repeat follows */
# Line 2773  for (;;) Line 2992  for (;;)
2992          if (eptr >= md->end_subject)          if (eptr >= md->end_subject)
2993            {            {
2994            SCHECK_PARTIAL();            SCHECK_PARTIAL();
2995            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2996            }            }
2997          GETCHARINCTEST(c, eptr);          GETCHARINCTEST(c, eptr);
2998          if (!_pcre_xclass(c, data)) MRRETURN(MATCH_NOMATCH);          if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH);
2999          }          }
3000    
3001        /* 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 2793  for (;;) Line 3012  for (;;)
3012            {            {
3013            RMATCH(eptr, ecode, offset_top, md, eptrb, RM20);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM20);
3014            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3015            if (fi >= max) MRRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3016            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3017              {              {
3018              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3019              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3020              }              }
3021            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3022            if (!_pcre_xclass(c, data)) MRRETURN(MATCH_NOMATCH);            if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH);
3023            }            }
3024          /* Control never gets here */          /* Control never gets here */
3025          }          }
# Line 2818  for (;;) Line 3037  for (;;)
3037              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3038              break;              break;
3039              }              }
3040    #ifdef SUPPORT_UTF
3041            GETCHARLENTEST(c, eptr, len);            GETCHARLENTEST(c, eptr, len);
3042            if (!_pcre_xclass(c, data)) break;  #else
3043              c = *eptr;
3044    #endif
3045              if (!PRIV(xclass)(c, data, utf)) break;
3046            eptr += len;            eptr += len;
3047            }            }
3048          for(;;)          for(;;)
# Line 2827  for (;;) Line 3050  for (;;)
3050            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);
3051            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3052            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
3053            if (utf8) BACKCHAR(eptr);  #ifdef SUPPORT_UTF
3054              if (utf) BACKCHAR(eptr);
3055    #endif
3056            }            }
3057          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3058          }          }
3059    
3060        /* Control never gets here */        /* Control never gets here */
# Line 2839  for (;;) Line 3064  for (;;)
3064      /* Match a single character, casefully */      /* Match a single character, casefully */
3065    
3066      case OP_CHAR:      case OP_CHAR:
3067  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3068      if (utf8)      if (utf)
3069        {        {
3070        length = 1;        length = 1;
3071        ecode++;        ecode++;
# Line 2848  for (;;) Line 3073  for (;;)
3073        if (length > md->end_subject - eptr)        if (length > md->end_subject - eptr)
3074          {          {
3075          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
3076          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3077          }          }
3078        while (length-- > 0) if (*ecode++ != *eptr++) MRRETURN(MATCH_NOMATCH);        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);
3079        }        }
3080      else      else
3081  #endif  #endif
3082        /* Not UTF mode */
     /* Non-UTF-8 mode */  
3083        {        {
3084        if (md->end_subject - eptr < 1)        if (md->end_subject - eptr < 1)
3085          {          {
3086          SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */          SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
3087          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3088          }          }
3089        if (ecode[1] != *eptr++) MRRETURN(MATCH_NOMATCH);        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);
3090        ecode += 2;        ecode += 2;
3091        }        }
3092      break;      break;
3093    
3094      /* Match a single character, caselessly */      /* Match a single character, caselessly. If we are at the end of the
3095        subject, give up immediately. */
3096    
3097      case OP_CHARI:      case OP_CHARI:
3098  #ifdef SUPPORT_UTF8      if (eptr >= md->end_subject)
3099      if (utf8)        {
3100          SCHECK_PARTIAL();
3101          RRETURN(MATCH_NOMATCH);
3102          }
3103    
3104    #ifdef SUPPORT_UTF
3105        if (utf)
3106        {        {
3107        length = 1;        length = 1;
3108        ecode++;        ecode++;
3109        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
3110    
       if (length > md->end_subject - eptr)  
         {  
         CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */  
         MRRETURN(MATCH_NOMATCH);  
         }  
   
3111        /* 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
3112        can use the fast lookup table. */        we know that its other case must also be one byte long, so we can use the
3113          fast lookup table. We know that there is at least one byte left in the
3114          subject. */
3115    
3116        if (fc < 128)        if (fc < 128)
3117          {          {
3118          if (md->lcc[*ecode++] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);          if (md->lcc[fc]
3119                != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH);
3120            ecode++;
3121            eptr++;
3122          }          }
3123    
3124        /* Otherwise we must pick up the subject character */        /* Otherwise we must pick up the subject character. Note that we cannot
3125          use the value of "length" to check for sufficient bytes left, because the
3126          other case of the character may have more or fewer bytes.  */
3127    
3128        else        else
3129          {          {
# Line 2907  for (;;) Line 3139  for (;;)
3139  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3140            if (dc != UCD_OTHERCASE(fc))            if (dc != UCD_OTHERCASE(fc))
3141  #endif  #endif
3142              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3143            }            }
3144          }          }
3145        }        }
3146      else      else
3147  #endif   /* SUPPORT_UTF8 */  #endif   /* SUPPORT_UTF */
3148    
3149      /* Non-UTF-8 mode */      /* Not UTF mode */
3150        {        {
3151        if (md->end_subject - eptr < 1)        if (TABLE_GET(ecode[1], md->lcc, ecode[1])
3152          {            != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH);
3153          SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */        eptr++;
         MRRETURN(MATCH_NOMATCH);  
         }  
       if (md->lcc[ecode[1]] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);  
3154        ecode += 2;        ecode += 2;
3155        }        }
3156      break;      break;
# Line 2931  for (;;) Line 3160  for (;;)
3160      case OP_EXACT:      case OP_EXACT:
3161      case OP_EXACTI:      case OP_EXACTI:
3162      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3163      ecode += 3;      ecode += 1 + IMM2_SIZE;
3164      goto REPEATCHAR;      goto REPEATCHAR;
3165    
3166      case OP_POSUPTO:      case OP_POSUPTO:
# Line 2946  for (;;) Line 3175  for (;;)
3175      min = 0;      min = 0;
3176      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3177      minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI;      minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI;
3178      ecode += 3;      ecode += 1 + IMM2_SIZE;
3179      goto REPEATCHAR;      goto REPEATCHAR;
3180    
3181      case OP_POSSTAR:      case OP_POSSTAR:
# Line 2994  for (;;) Line 3223  for (;;)
3223      /* Common code for all repeated single-character matches. */      /* Common code for all repeated single-character matches. */
3224    
3225      REPEATCHAR:      REPEATCHAR:
3226  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3227      if (utf8)      if (utf)
3228        {        {
3229        length = 1;        length = 1;
3230        charptr = ecode;        charptr = ecode;
# Line 3011  for (;;) Line 3240  for (;;)
3240          unsigned int othercase;          unsigned int othercase;
3241          if (op >= OP_STARI &&     /* Caseless */          if (op >= OP_STARI &&     /* Caseless */
3242              (othercase = UCD_OTHERCASE(fc)) != fc)              (othercase = UCD_OTHERCASE(fc)) != fc)
3243            oclength = _pcre_ord2utf8(othercase, occhars);            oclength = PRIV(ord2utf)(othercase, occhars);
3244          else oclength = 0;          else oclength = 0;
3245  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3246    
3247          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3248            {            {
3249            if (eptr <= md->end_subject - length &&            if (eptr <= md->end_subject - length &&
3250              memcmp(eptr, charptr, length) == 0) eptr += length;              memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3251  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3252            else if (oclength > 0 &&            else if (oclength > 0 &&
3253                     eptr <= md->end_subject - oclength &&                     eptr <= md->end_subject - oclength &&
3254                     memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                     memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3255  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3256            else            else
3257              {              {
3258              CHECK_PARTIAL();              CHECK_PARTIAL();
3259              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3260              }              }
3261            }            }
3262    
# Line 3039  for (;;) Line 3268  for (;;)
3268              {              {
3269              RMATCH(eptr, ecode, offset_top, md, eptrb, RM22);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM22);
3270              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3271              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3272              if (eptr <= md->end_subject - length &&              if (eptr <= md->end_subject - length &&
3273                memcmp(eptr, charptr, length) == 0) eptr += length;                memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3274  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3275              else if (oclength > 0 &&              else if (oclength > 0 &&
3276                       eptr <= md->end_subject - oclength &&                       eptr <= md->end_subject - oclength &&
3277                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                       memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3278  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3279              else              else
3280                {                {
3281                CHECK_PARTIAL();                CHECK_PARTIAL();
3282                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3283                }                }
3284              }              }
3285            /* Control never gets here */            /* Control never gets here */
# Line 3062  for (;;) Line 3291  for (;;)
3291            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3292              {              {
3293              if (eptr <= md->end_subject - length &&              if (eptr <= md->end_subject - length &&
3294                  memcmp(eptr, charptr, length) == 0) eptr += length;                  memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3295  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3296              else if (oclength > 0 &&              else if (oclength > 0 &&
3297                       eptr <= md->end_subject - oclength &&                       eptr <= md->end_subject - oclength &&
3298                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                       memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3299  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3300              else              else
3301                {                {
# Line 3081  for (;;) Line 3310  for (;;)
3310              {              {
3311              RMATCH(eptr, ecode, offset_top, md, eptrb, RM23);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM23);
3312              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3313              if (eptr == pp) { MRRETURN(MATCH_NOMATCH); }              if (eptr == pp) { RRETURN(MATCH_NOMATCH); }
3314  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3315              eptr--;              eptr--;
3316              BACKCHAR(eptr);              BACKCHAR(eptr);
# Line 3098  for (;;) Line 3327  for (;;)
3327        value of fc will always be < 128. */        value of fc will always be < 128. */
3328        }        }
3329      else      else
3330  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
3331          /* When not in UTF-8 mode, load a single-byte character. */
3332          fc = *ecode++;
3333    
3334      /* When not in UTF-8 mode, load a single-byte character. */      /* The value of fc at this point is always one character, though we may
3335        or may not be in UTF mode. The code is duplicated for the caseless and
     fc = *ecode++;  
   
     /* 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  
3336      caseful cases, for speed, since matching characters is likely to be quite      caseful cases, for speed, since matching characters is likely to be quite
3337      common. First, ensure the minimum number of matches are present. If min =      common. First, ensure the minimum number of matches are present. If min =
3338      max, continue at the same level without recursing. Otherwise, if      max, continue at the same level without recursing. Otherwise, if
# Line 3118  for (;;) Line 3345  for (;;)
3345    
3346      if (op >= OP_STARI)  /* Caseless */      if (op >= OP_STARI)  /* Caseless */
3347        {        {
3348        fc = md->lcc[fc];  #ifdef COMPILE_PCRE8
3349          /* fc must be < 128 if UTF is enabled. */
3350          foc = md->fcc[fc];
3351    #else
3352    #ifdef SUPPORT_UTF
3353    #ifdef SUPPORT_UCP
3354          if (utf && fc > 127)
3355            foc = UCD_OTHERCASE(fc);
3356    #else
3357          if (utf && fc > 127)
3358            foc = fc;
3359    #endif /* SUPPORT_UCP */
3360          else
3361    #endif /* SUPPORT_UTF */
3362            foc = TABLE_GET(fc, md->fcc, fc);
3363    #endif /* COMPILE_PCRE8 */
3364    
3365        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
3366          {          {
3367          if (eptr >= md->end_subject)          if (eptr >= md->end_subject)
3368            {            {
3369            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3370            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3371            }            }
3372          if (fc != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);          if (fc != *eptr && foc != *eptr) RRETURN(MATCH_NOMATCH);
3373            eptr++;
3374          }          }
3375        if (min == max) continue;        if (min == max) continue;
3376        if (minimize)        if (minimize)
# Line 3135  for (;;) Line 3379  for (;;)
3379            {            {
3380            RMATCH(eptr, ecode, offset_top, md, eptrb, RM24);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM24);
3381            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3382            if (fi >= max) MRRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3383            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3384              {              {
3385              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3386              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3387              }              }
3388            if (fc != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);            if (fc != *eptr && foc != *eptr) RRETURN(MATCH_NOMATCH);
3389              eptr++;
3390            }            }
3391          /* Control never gets here */          /* Control never gets here */
3392          }          }
# Line 3155  for (;;) Line 3400  for (;;)
3400              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3401              break;              break;
3402              }              }
3403            if (fc != md->lcc[*eptr]) break;            if (fc != *eptr && foc != *eptr) break;
3404            eptr++;            eptr++;
3405            }            }
3406    
# Line 3167  for (;;) Line 3412  for (;;)
3412            eptr--;            eptr--;
3413            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3414            }            }
3415          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3416          }          }
3417        /* Control never gets here */        /* Control never gets here */
3418        }        }
# Line 3181  for (;;) Line 3426  for (;;)
3426          if (eptr >= md->end_subject)          if (eptr >= md->end_subject)
3427            {            {
3428            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3429            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3430            }            }
3431          if (fc != *eptr++) MRRETURN(MATCH_NOMATCH);          if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
3432          }          }
3433    
3434        if (min == max) continue;        if (min == max) continue;
# Line 3194  for (;;) Line 3439  for (;;)
3439            {            {
3440            RMATCH(eptr, ecode, offset_top, md, eptrb, RM26);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM26);
3441            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3442            if (fi >= max) MRRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3443            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3444              {              {
3445              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3446              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3447              }              }
3448            if (fc != *eptr++) MRRETURN(MATCH_NOMATCH);            if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
3449            }            }
3450          /* Control never gets here */          /* Control never gets here */
3451          }          }
# Line 3225  for (;;) Line 3470  for (;;)
3470            eptr--;            eptr--;
3471            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3472            }            }
3473          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3474          }          }
3475        }        }
3476      /* Control never gets here */      /* Control never gets here */
# Line 3238  for (;;) Line 3483  for (;;)
3483      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
3484        {        {
3485        SCHECK_PARTIAL();        SCHECK_PARTIAL();
3486        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
3487        }        }
3488      ecode++;      ecode++;
3489      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
3490      if (op == OP_NOTI)         /* The caseless case */      if (op == OP_NOTI)         /* The caseless case */
3491        {        {
3492  #ifdef SUPPORT_UTF8        register int ch, och;
3493        if (c < 256)        ch = *ecode++;
3494  #endif  #ifdef COMPILE_PCRE8
3495        c = md->lcc[c];        /* ch must be < 128 if UTF is enabled. */
3496        if (md->lcc[*ecode++] == c) MRRETURN(MATCH_NOMATCH);        och = md->fcc[ch];
3497    #else
3498    #ifdef SUPPORT_UTF
3499    #ifdef SUPPORT_UCP
3500          if (utf && ch > 127)
3501            och = UCD_OTHERCASE(ch);
3502    #else
3503          if (utf && ch > 127)
3504            och = ch;
3505    #endif /* SUPPORT_UCP */
3506          else
3507    #endif /* SUPPORT_UTF */
3508            och = TABLE_GET(ch, md->fcc, ch);
3509    #endif /* COMPILE_PCRE8 */
3510          if (ch == c || och == c) RRETURN(MATCH_NOMATCH);
3511        }        }
3512      else    /* Caseful */      else    /* Caseful */
3513        {        {
3514        if (*ecode++ == c) MRRETURN(MATCH_NOMATCH);        if (*ecode++ == c) RRETURN(MATCH_NOMATCH);
3515        }        }
3516      break;      break;
3517    
# Line 3266  for (;;) Line 3525  for (;;)
3525      case OP_NOTEXACT:      case OP_NOTEXACT:
3526      case OP_NOTEXACTI:      case OP_NOTEXACTI:
3527      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3528      ecode += 3;      ecode += 1 + IMM2_SIZE;
3529      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3530    
3531      case OP_NOTUPTO:      case OP_NOTUPTO:
# Line 3276  for (;;) Line 3535  for (;;)
3535      min = 0;      min = 0;
3536      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3537      minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI;      minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI;
3538      ecode += 3;      ecode += 1 + IMM2_SIZE;
3539      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3540    
3541      case OP_NOTPOSSTAR:      case OP_NOTPOSSTAR:
# Line 3308  for (;;) Line 3567  for (;;)
3567      possessive = TRUE;      possessive = TRUE;
3568      min = 0;      min = 0;
3569      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3570      ecode += 3;      ecode += 1 + IMM2_SIZE;
3571      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3572    
3573      case OP_NOTSTAR:      case OP_NOTSTAR:
# Line 3347  for (;;) Line 3606  for (;;)
3606    
3607      if (op >= OP_NOTSTARI)     /* Caseless */      if (op >= OP_NOTSTARI)     /* Caseless */
3608        {        {
3609        fc = md->lcc[fc];  #ifdef COMPILE_PCRE8
3610          /* fc must be < 128 if UTF is enabled. */
3611          foc = md->fcc[fc];
3612    #else
3613    #ifdef SUPPORT_UTF
3614    #ifdef SUPPORT_UCP
3615          if (utf && fc > 127)
3616            foc = UCD_OTHERCASE(fc);
3617    #else
3618          if (utf && fc > 127)
3619            foc = fc;
3620    #endif /* SUPPORT_UCP */
3621          else
3622    #endif /* SUPPORT_UTF */
3623            foc = TABLE_GET(fc, md->fcc, fc);
3624    #endif /* COMPILE_PCRE8 */
3625    
3626  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3627        /* UTF-8 mode */        if (utf)
       if (utf8)  
3628          {          {
3629          register unsigned int d;          register unsigned int d;
3630          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 3359  for (;;) Line 3632  for (;;)
3632            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3633              {              {
3634              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3635              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3636              }              }
3637            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3638            if (d < 256) d = md->lcc[d];            if (fc == d || foc == d) RRETURN(MATCH_NOMATCH);
           if (fc == d) MRRETURN(MATCH_NOMATCH);  
3639            }            }
3640          }          }
3641        else        else
3642  #endif  #endif
3643          /* Not UTF mode */
       /* Not UTF-8 mode */  
3644          {          {
3645          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3646            {            {
3647            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3648              {              {
3649              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3650              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3651              }              }
3652            if (fc == md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);            if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
3653              eptr++;
3654            }            }
3655          }          }
3656    
# Line 3386  for (;;) Line 3658  for (;;)
3658    
3659        if (minimize)        if (minimize)
3660          {          {
3661  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3662          /* UTF-8 mode */          if (utf)
         if (utf8)  
3663            {            {
3664            register unsigned int d;            register unsigned int d;
3665            for (fi = min;; fi++)            for (fi = min;; fi++)
3666              {              {
3667              RMATCH(eptr, ecode, offset_top, md, eptrb, RM28);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM28);
3668              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3669              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3670              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3671                {                {
3672                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3673                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3674                }                }
3675              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3676              if (d < 256) d = md->lcc[d];              if (fc == d || foc == d) RRETURN(MATCH_NOMATCH);
             if (fc == d) MRRETURN(MATCH_NOMATCH);  
3677              }              }
3678            }            }
3679          else          else
3680  #endif  #endif
3681          /* Not UTF-8 mode */          /* Not UTF mode */
3682            {            {
3683            for (fi = min;; fi++)            for (fi = min;; fi++)
3684              {              {
3685              RMATCH(eptr, ecode, offset_top, md, eptrb, RM29);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM29);
3686              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3687              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3688              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3689                {                {
3690                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3691                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3692                }                }
3693              if (fc == md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);              if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
3694                eptr++;
3695              }              }
3696            }            }
3697          /* Control never gets here */          /* Control never gets here */
# Line 3432  for (;;) Line 3703  for (;;)
3703          {          {
3704          pp = eptr;          pp = eptr;
3705    
3706  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3707          /* UTF-8 mode */          if (utf)
         if (utf8)  
3708            {            {
3709            register unsigned int d;            register unsigned int d;
3710            for (i = min; i < max; i++)            for (i = min; i < max; i++)
# Line 3446  for (;;) Line 3716  for (;;)
3716                break;                break;
3717                }                }
3718              GETCHARLEN(d, eptr, len);              GETCHARLEN(d, eptr, len);
3719              if (d < 256) d = md->lcc[d];              if (fc == d || foc == d) break;
             if (fc == d) break;  
3720              eptr += len;              eptr += len;
3721              }              }
3722          if (possessive) continue;          if (possessive) continue;
# Line 3461  for (;;) Line 3730  for (;;)
3730            }            }
3731          else          else
3732  #endif  #endif
3733          /* Not UTF-8 mode */          /* Not UTF mode */
3734            {            {
3735            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3736              {              {
# Line 3470  for (;;) Line 3739  for (;;)
3739                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3740                break;                break;
3741                }                }
3742              if (fc == md->lcc[*eptr]) break;              if (fc == *eptr || foc == *eptr) break;
3743              eptr++;              eptr++;
3744              }              }
3745            if (possessive) continue;            if (possessive) continue;
# Line 3482  for (;;) Line 3751  for (;;)
3751              }              }
3752            }            }
3753    
3754          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3755          }          }
3756        /* Control never gets here */        /* Control never gets here */
3757        }        }
# Line 3491  for (;;) Line 3760  for (;;)
3760    
3761      else      else
3762        {        {
3763  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3764        /* UTF-8 mode */        if (utf)
       if (utf8)  
3765          {          {
3766          register unsigned int d;          register unsigned int d;
3767          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 3501  for (;;) Line 3769  for (;;)
3769            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3770              {              {
3771              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3772              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3773              }              }
3774            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3775            if (fc == d) MRRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
3776            }            }
3777          }          }
3778        else        else
3779  #endif  #endif
3780        /* Not UTF-8 mode */        /* Not UTF mode */
3781          {          {
3782          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3783            {            {
3784            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3785              {              {
3786              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3787              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3788              }              }
3789            if (fc == *eptr++) MRRETURN(MATCH_NOMATCH);            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
3790            }            }
3791          }          }
3792    
# Line 3526  for (;;) Line 3794  for (;;)
3794    
3795        if (minimize)        if (minimize)
3796          {          {
3797  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3798          /* UTF-8 mode */          if (utf)
         if (utf8)  
3799            {            {
3800            register unsigned int d;            register unsigned int d;
3801            for (fi = min;; fi++)            for (fi = min;; fi++)
3802              {              {
3803              RMATCH(eptr, ecode, offset_top, md, eptrb, RM32);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM32);
3804              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3805              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3806              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3807                {                {
3808                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3809                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3810                }                }
3811              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3812              if (fc == d) MRRETURN(MATCH_NOMATCH);              if (fc == d) RRETURN(MATCH_NOMATCH);
3813              }              }
3814            }            }
3815          else          else
3816  #endif  #endif
3817          /* Not UTF-8 mode */          /* Not UTF mode */
3818            {            {
3819            for (fi = min;; fi++)            for (fi = min;; fi++)
3820              {              {
3821              RMATCH(eptr, ecode, offset_top, md, eptrb, RM33);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM33);
3822              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3823              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3824              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3825                {                {
3826                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3827                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3828                }                }
3829              if (fc == *eptr++) MRRETURN(MATCH_NOMATCH);              if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
3830              }              }
3831            }            }
3832          /* Control never gets here */          /* Control never gets here */
# Line 3571  for (;;) Line 3838  for (;;)
3838          {          {
3839          pp = eptr;          pp = eptr;
3840    
3841  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3842          /* UTF-8 mode */          if (utf)
         if (utf8)  
3843            {            {
3844            register unsigned int d;            register unsigned int d;
3845            for (i = min; i < max; i++)            for (i = min; i < max; i++)
# Line 3599  for (;;) Line 3865  for (;;)
3865            }            }
3866          else          else
3867  #endif  #endif
3868          /* Not UTF-8 mode */          /* Not UTF mode */
3869            {            {
3870            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3871              {              {
# Line 3620  for (;;) Line 3886  for (;;)
3886              }              }
3887            }            }
3888    
3889          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3890          }          }
3891        }        }
3892      /* Control never gets here */      /* Control never gets here */
# Line 3632  for (;;) Line 3898  for (;;)
3898      case OP_TYPEEXACT:      case OP_TYPEEXACT:
3899      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3900      minimize = TRUE;      minimize = TRUE;
3901      ecode += 3;      ecode += 1 + IMM2_SIZE;
3902      goto REPEATTYPE;      goto REPEATTYPE;
3903    
3904      case OP_TYPEUPTO:      case OP_TYPEUPTO:
# Line 3640  for (;;) Line 3906  for (;;)
3906      min = 0;      min = 0;
3907      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3908      minimize = *ecode == OP_TYPEMINUPTO;      minimize = *ecode == OP_TYPEMINUPTO;
3909      ecode += 3;      ecode += 1 + IMM2_SIZE;
3910      goto REPEATTYPE;      goto REPEATTYPE;
3911    
3912      case OP_TYPEPOSSTAR:      case OP_TYPEPOSSTAR:
# Line 3668  for (;;) Line 3934  for (;;)
3934      possessive = TRUE;      possessive = TRUE;
3935      min = 0;      min = 0;
3936      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3937      ecode += 3;      ecode += 1 + IMM2_SIZE;
3938      goto REPEATTYPE;      goto REPEATTYPE;
3939    
3940      case OP_TYPESTAR:      case OP_TYPESTAR:
# Line 3714  for (;;) Line 3980  for (;;)
3980          switch(prop_type)          switch(prop_type)
3981            {            {
3982            case PT_ANY:            case PT_ANY:
3983            if (prop_fail_result) MRRETURN(MATCH_NOMATCH);            if (prop_fail_result) RRETURN(MATCH_NOMATCH);
3984            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3985              {              {
3986              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3987                {                {
3988                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3989                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3990                }                }
3991              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3992              }              }
# Line 3733  for (;;) Line 3999  for (;;)
3999              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4000                {                {
4001                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4002                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4003                }                }
4004              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4005              chartype = UCD_CHARTYPE(c);              chartype = UCD_CHARTYPE(c);
4006              if ((chartype == ucp_Lu ||              if ((chartype == ucp_Lu ||
4007                   chartype == ucp_Ll ||                   chartype == ucp_Ll ||
4008                   chartype == ucp_Lt) == prop_fail_result)                   chartype == ucp_Lt) == prop_fail_result)
4009                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4010              }              }
4011            break;            break;
4012    
# Line 3750  for (;;) Line 4016  for (;;)
4016              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4017                {                {
4018                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4019                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4020                }                }
4021              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4022              if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)              if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)
4023                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4024              }              }
4025            break;            break;
4026    
# Line 3764  for (;;) Line 4030  for (;;)
4030              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4031                {                {
4032                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4033                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4034                }                }
4035              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4036              if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)              if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)
4037                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4038              }              }
4039            break;            break;
4040    
# Line 3778  for (;;) Line 4044  for (;;)
4044              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4045                {                {
4046                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4047                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4048                }                }
4049              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4050              if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result)              if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result)
4051                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4052              }              }
4053            break;            break;
4054    
# Line 3793  for (;;) Line 4059  for (;;)
4059              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4060                {                {
4061                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4062                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4063                }                }
4064              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4065              category = UCD_CATEGORY(c);              category = UCD_CATEGORY(c);
4066              if ((category == ucp_L || category == ucp_N) == prop_fail_result)              if ((category == ucp_L || category == ucp_N) == prop_fail_result)
4067                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4068              }              }
4069            break;            break;
4070    
# Line 3808  for (;;) Line 4074  for (;;)
4074              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4075                {                {
4076                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4077                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4078                }                }
4079              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4080              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
4081                   c == CHAR_FF || c == CHAR_CR)                   c == CHAR_FF || c == CHAR_CR)
4082                     == prop_fail_result)                     == prop_fail_result)
4083                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4084              }              }
4085            break;            break;
4086    
# Line 3824  for (;;) Line 4090  for (;;)
4090              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4091                {                {
4092                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4093                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4094                }                }
4095              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4096              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
4097                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)
4098                     == prop_fail_result)                     == prop_fail_result)
4099                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4100              }              }
4101            break;            break;
4102    
# Line 3841  for (;;) Line 4107  for (;;)
4107              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4108                {                {
4109                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4110                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4111                }                }
4112              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4113              category = UCD_CATEGORY(c);              category = UCD_CATEGORY(c);
4114              if ((category == ucp_L || category == ucp_N || c == CHAR_UNDERSCORE)              if ((category == ucp_L || category == ucp_N || c == CHAR_UNDERSCORE)
4115                     == prop_fail_result)                     == prop_fail_result)
4116                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4117              }              }
4118            break;            break;
4119    
# Line 3868  for (;;) Line 4134  for (;;)
4134            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4135              {              {
4136              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4137              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4138              }              }
4139            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
4140            if (UCD_CATEGORY(c) == ucp_M) MRRETURN(MATCH_NOMATCH);            if (UCD_CATEGORY(c) == ucp_M) RRETURN(MATCH_NOMATCH);
4141            while (eptr < md->end_subject)            while (eptr < md->end_subject)
4142              {              {
4143              int len = 1;              int len = 1;
4144              if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
4145              if (UCD_CATEGORY(c) != ucp_M) break;              if (UCD_CATEGORY(c) != ucp_M) break;
4146              eptr += len;              eptr += len;
4147              }              }
# Line 3887  for (;;) Line 4153  for (;;)
4153    
4154  /* Handle all other cases when the coding is UTF-8 */  /* Handle all other cases when the coding is UTF-8 */
4155    
4156  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
4157        if (utf8) switch(ctype)        if (utf) switch(ctype)
4158          {          {
4159          case OP_ANY:          case OP_ANY:
4160          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 3896  for (;;) Line 4162  for (;;)
4162            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4163              {              {
4164              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4165              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4166              }              }
4167            if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
4168            eptr++;            eptr++;
4169            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4170            }            }
4171          break;          break;
4172    
# Line 3910  for (;;) Line 4176  for (;;)
4176            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4177              {              {
4178              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4179              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4180              }              }
4181            eptr++;            eptr++;
4182            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4183            }            }
4184          break;          break;
4185    
4186          case OP_ANYBYTE:          case OP_ANYBYTE:
4187          if (eptr > md->end_subject - min) MRRETURN(MATCH_NOMATCH);          if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
4188          eptr += min;          eptr += min;
4189          break;          break;
4190    
# Line 3928  for (;;) Line 4194  for (;;)
4194            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4195              {              {
4196              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4197              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4198              }              }
4199            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4200            switch(c)            switch(c)
4201              {              {
4202              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4203    
4204              case 0x000d:              case 0x000d:
4205              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
# Line 3947  for (;;) Line 4213  for (;;)
4213              case 0x0085:              case 0x0085:
4214              case 0x2028:              case 0x2028:
4215              case 0x2029:              case 0x2029:
4216              if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4217              break;              break;
4218              }              }
4219            }            }
# Line 3959  for (;;) Line 4225  for (;;)
4225            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4226              {              {
4227              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4228              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4229              }              }
4230            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4231            switch(c)            switch(c)
# Line 3984  for (;;) Line 4250  for (;;)
4250              case 0x202f:    /* NARROW NO-BREAK SPACE */              case 0x202f:    /* NARROW NO-BREAK SPACE */
4251              case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */              case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4252              case 0x3000:    /* IDEOGRAPHIC SPACE */              case 0x3000:    /* IDEOGRAPHIC SPACE */
4253              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4254              }              }
4255            }            }
4256          break;          break;
# Line 3995  for (;;) Line 4261  for (;;)
4261            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4262              {              {
4263              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4264              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4265              }              }
4266            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4267            switch(c)            switch(c)
4268              {              {
4269              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4270              case 0x09:      /* HT */              case 0x09:      /* HT */
4271              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4272              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
# Line 4031  for (;;) Line 4297  for (;;)
4297            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4298              {              {
4299              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4300              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4301              }              }
4302            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4303            switch(c)            switch(c)
# Line 4044  for (;;) Line 4310  for (;;)
4310              case 0x85:      /* NEL */              case 0x85:      /* NEL */
4311              case 0x2028:    /* LINE SEPARATOR */              case 0x2028:    /* LINE SEPARATOR */
4312              case 0x2029:    /* PARAGRAPH SEPARATOR */              case 0x2029:    /* PARAGRAPH SEPARATOR */
4313              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4314              }              }
4315            }            }
4316          break;          break;
# Line 4055  for (;;) Line 4321  for (;;)
4321            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4322              {              {
4323              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4324              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4325              }              }
4326            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4327            switch(c)            switch(c)
4328              {              {
4329              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4330              case 0x0a:      /* LF */              case 0x0a:      /* LF */
4331              case 0x0b:      /* VT */              case 0x0b:      /* VT */
4332              case 0x0c:      /* FF */              case 0x0c:      /* FF */
# Line 4079  for (;;) Line 4345  for (;;)
4345            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4346              {              {
4347              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4348              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4349              }              }
4350            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4351            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)
4352              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4353            }            }
4354          break;          break;
4355    
# Line 4093  for (;;) Line 4359  for (;;)
4359            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4360              {              {
4361              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4362              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4363              }              }
4364            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_digit) == 0)
4365              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4366              eptr++;
4367            /* 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 */
4368            }            }
4369          break;          break;
# Line 4107  for (;;) Line 4374  for (;;)
4374            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4375              {              {
4376              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4377              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4378              }              }
4379            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)
4380              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4381            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);            eptr++;
4382              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4383            }            }
4384          break;          break;
4385    
# Line 4121  for (;;) Line 4389  for (;;)
4389            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4390              {              {
4391              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4392              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4393              }              }
4394            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_space) == 0)
4395              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4396              eptr++;
4397            /* 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 */
4398            }            }
4399          break;          break;
# Line 4135  for (;;) Line 4404  for (;;)
4404            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4405              {              {
4406              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4407              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4408              }              }
4409            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0)            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0)
4410              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4411            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);            eptr++;
4412              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4413            }            }
4414          break;          break;
4415    
# Line 4149  for (;;) Line 4419  for (;;)
4419            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4420              {              {
4421              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4422              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4423              }              }
4424            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_word) == 0)
4425              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4426              eptr++;
4427            /* 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 */
4428            }            }
4429          break;          break;
# Line 4162  for (;;) Line 4433  for (;;)
4433          }  /* End switch(ctype) */          }  /* End switch(ctype) */
4434    
4435        else        else
4436  #endif     /* SUPPORT_UTF8 */  #endif     /* SUPPORT_UTF */
4437    
4438        /* 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
4439        than OP_PROP and OP_NOTPROP. */        than OP_PROP and OP_NOTPROP. */
# Line 4175  for (;;) Line 4446  for (;;)
4446            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4447              {              {
4448              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4449              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4450              }              }
4451            if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
4452            eptr++;            eptr++;
4453            }            }
4454          break;          break;
# Line 4186  for (;;) Line 4457  for (;;)
4457          if (eptr > md->end_subject - min)          if (eptr > md->end_subject - min)
4458            {            {
4459            SCHECK_PARTIAL();            SCHECK_PARTIAL();
4460            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
4461            }            }
4462          eptr += min;          eptr += min;
4463          break;          break;
# Line 4195  for (;;) Line 4466  for (;;)
4466          if (eptr > md->end_subject - min)          if (eptr > md->end_subject - min)
4467            {            {
4468            SCHECK_PARTIAL();            SCHECK_PARTIAL();
4469            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
4470            }            }
4471          eptr += min;          eptr += min;
4472          break;          break;
# Line 4206  for (;;) Line 4477  for (;;)
4477            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4478              {              {
4479              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4480              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4481              }              }
4482            switch(*eptr++)            switch(*eptr++)
4483              {              {
4484              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4485    
4486              case 0x000d:              case 0x000d:
4487              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
# Line 4222  for (;;) Line 4493  for (;;)
4493              case 0x000b:              case 0x000b:
4494              case 0x000c:              case 0x000c:
4495              case 0x0085:              case 0x0085:
4496              if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);  #ifdef COMPILE_PCRE16
4497                case 0x2028:
4498                case 0x2029:
4499    #endif
4500                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4501              break;              break;
4502              }              }
4503            }            }
# Line 4234  for (;;) Line 4509  for (;;)
4509            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4510              {              {
4511              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4512              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4513              }              }
4514            switch(*eptr++)            switch(*eptr++)
4515              {              {
# Line 4242  for (;;) Line 4517  for (;;)
4517              case 0x09:      /* HT */              case 0x09:      /* HT */
4518              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4519              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
4520              MRRETURN(MATCH_NOMATCH);  #ifdef COMPILE_PCRE16
4521                case 0x1680:    /* OGHAM SPACE MARK */
4522                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
4523                case 0x2000:    /* EN QUAD */
4524                case 0x2001:    /* EM QUAD */
4525                case 0x2002:    /* EN SPACE */
4526                case 0x2003:    /* EM SPACE */
4527                case 0x2004:    /* THREE-PER-EM SPACE */
4528                case 0x2005:    /* FOUR-PER-EM SPACE */
4529                case 0x2006:    /* SIX-PER-EM SPACE */
4530                case 0x2007:    /* FIGURE SPACE */
4531                case 0x2008:    /* PUNCTUATION SPACE */
4532                case 0x2009:    /* THIN SPACE */
4533                case 0x200A:    /* HAIR SPACE */
4534                case 0x202f:    /* NARROW NO-BREAK SPACE */
4535                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4536                case 0x3000:    /* IDEOGRAPHIC SPACE */
4537    #endif
4538                RRETURN(MATCH_NOMATCH);
4539              }              }
4540            }            }
4541          break;          break;
# Line 4253  for (;;) Line 4546  for (;;)
4546            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4547              {              {
4548              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4549              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4550              }              }
4551            switch(*eptr++)            switch(*eptr++)
4552              {              {
4553              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4554              case 0x09:      /* HT */              case 0x09:      /* HT */
4555              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4556              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
4557    #ifdef COMPILE_PCRE16
4558                case 0x1680:    /* OGHAM SPACE MARK */
4559                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
4560                case 0x2000:    /* EN QUAD */
4561                case 0x2001:    /* EM QUAD */
4562                case 0x2002:    /* EN SPACE */
4563                case 0x2003:    /* EM SPACE */
4564                case 0x2004:    /* THREE-PER-EM SPACE */
4565                case 0x2005:    /* FOUR-PER-EM SPACE */
4566                case 0x2006:    /* SIX-PER-EM SPACE */
4567                case 0x2007:    /* FIGURE SPACE */
4568                case 0x2008:    /* PUNCTUATION SPACE */
4569                case 0x2009:    /* THIN SPACE */
4570                case 0x200A:    /* HAIR SPACE */
4571                case 0x202f:    /* NARROW NO-BREAK SPACE */
4572                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4573                case 0x3000:    /* IDEOGRAPHIC SPACE */
4574    #endif
4575              break;              break;
4576              }              }
4577            }            }
# Line 4272  for (;;) Line 4583  for (;;)
4583            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4584              {              {
4585              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4586              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4587              }              }
4588            switch(*eptr++)            switch(*eptr++)
4589              {              {
# Line 4282  for (;;) Line 4593  for (;;)
4593              case 0x0c:      /* FF */              case 0x0c:      /* FF */
4594              case 0x0d:      /* CR */              case 0x0d:      /* CR */
4595              case 0x85:      /* NEL */              case 0x85:      /* NEL */
4596              MRRETURN(MATCH_NOMATCH);  #ifdef COMPILE_PCRE16
4597                case 0x2028:    /* LINE SEPARATOR */
4598                case 0x2029:    /* PARAGRAPH SEPARATOR */
4599    #endif
4600                RRETURN(MATCH_NOMATCH);
4601              }              }
4602            }            }
4603          break;          break;
# Line 4293  for (;;) Line 4608  for (;;)
4608            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4609              {              {
4610              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4611              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4612              }              }
4613            switch(*eptr++)            switch(*eptr++)
4614              {              {
4615              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4616              case 0x0a:      /* LF */              case 0x0a:      /* LF */
4617              case 0x0b:      /* VT */              case 0x0b:      /* VT */
4618              case 0x0c:      /* FF */              case 0x0c:      /* FF */
4619              case 0x0d:      /* CR */              case 0x0d:      /* CR */
4620              case 0x85:      /* NEL */              case 0x85:      /* NEL */
4621    #ifdef COMPILE_PCRE16
4622                case 0x2028:    /* LINE SEPARATOR */
4623                case 0x2029:    /* PARAGRAPH SEPARATOR */
4624    #endif
4625              break;              break;
4626              }              }
4627            }            }
# Line 4314  for (;;) Line 4633  for (;;)
4633            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4634              {              {
4635              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4636              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4637              }              }
4638            if ((md->ctypes[*eptr++] & ctype_digit) != 0) MRRETURN(MATCH_NOMATCH);            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_digit) != 0)
4639                RRETURN(MATCH_NOMATCH);
4640              eptr++;
4641            }            }
4642          break;          break;
4643    
# Line 4326  for (;;) Line 4647  for (;;)
4647            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4648              {              {
4649              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4650              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4651              }              }
4652            if ((md->ctypes[*eptr++] & ctype_digit) == 0) MRRETURN(MATCH_NOMATCH);            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_digit) == 0)
4653                RRETURN(MATCH_NOMATCH);
4654              eptr++;
4655            }            }
4656          break;          break;
4657    
# Line 4338  for (;;) Line 4661  for (;;)
4661            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4662              {              {
4663              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4664              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4665              }              }
4666            if ((md->ctypes[*eptr++] & ctype_space) != 0) MRRETURN(MATCH_NOMATCH);            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_space) != 0)
4667                RRETURN(MATCH_NOMATCH);
4668              eptr++;
4669            }            }
4670          break;          break;
4671    
# Line 4350  for (;;) Line 4675  for (;;)
4675            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4676              {              {
4677              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4678              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4679              }              }
4680            if ((md->ctypes[*eptr++] & ctype_space) == 0) MRRETURN(MATCH_NOMATCH);            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_space) == 0)
4681                RRETURN(MATCH_NOMATCH);
4682              eptr++;
4683            }            }
4684          break;          break;
4685    
# Line 4362  for (;;) Line 4689  for (;;)
4689            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4690              {              {
4691              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4692              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4693              }              }
4694            if ((md->ctypes[*eptr++] & ctype_word) != 0)            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_word) != 0)
4695              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4696              eptr++;
4697            }            }
4698          break;          break;
4699    
# Line 4375  for (;;) Line 4703  for (;;)
4703            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4704              {              {
4705              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4706              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4707              }              }
4708            if ((md->ctypes[*eptr++] & ctype_word) == 0)            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_word) == 0)
4709              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4710              eptr++;
4711            }            }
4712          break;          break;
4713    
# Line 4407  for (;;) Line 4736  for (;;)
4736              {              {
4737              RMATCH(eptr, ecode, offset_top, md, eptrb, RM36);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM36);
4738              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4739              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4740              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4741                {                {
4742                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4743                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4744                }                }
4745              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4746              if (prop_fail_result) MRRETURN(MATCH_NOMATCH);              if (prop_fail_result) RRETURN(MATCH_NOMATCH);
4747              }              }
4748            /* Control never gets here */            /* Control never gets here */
4749    
# Line 4424  for (;;) Line 4753  for (;;)
4753              int chartype;              int chartype;
4754              RMATCH(eptr, ecode, offset_top, md, eptrb, RM37);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM37);
4755              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4756              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4757              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4758                {                {
4759                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4760                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4761                }                }
4762              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4763              chartype = UCD_CHARTYPE(c);              chartype = UCD_CHARTYPE(c);
4764              if ((chartype == ucp_Lu ||              if ((chartype == ucp_Lu ||
4765                   chartype == ucp_Ll ||                   chartype == ucp_Ll ||
4766                   chartype == ucp_Lt) == prop_fail_result)                   chartype == ucp_Lt) == prop_fail_result)
4767                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4768              }              }
4769            /* Control never gets here */            /* Control never gets here */
4770    
# Line 4444  for (;;) Line 4773  for (;;)
4773              {              {
4774              RMATCH(eptr, ecode, offset_top, md, eptrb, RM38);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM38);
4775              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4776              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4777              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4778                {                {
4779                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4780                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4781                }                }
4782              GETCHARINCTEST(c, eptr);