/[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 667 by ph10, Mon Aug 22 14:57:32 2011 UTC revision 893 by ph10, Thu Jan 19 17:15:11 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);\    if (oldframe != &frame_zero) (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;
485  int condcode;  int condcode;
486    
487  /* When recursion is not being used, all "local" variables that have to be  /* When recursion is not being used, all "local" variables that have to be
488  preserved over calls to RMATCH() are part of a "frame" which is obtained from  preserved over calls to RMATCH() are part of a "frame". We set up the top-level
489  heap storage. Set up the top-level frame here; others are obtained from the  frame on the stack here; subsequent instantiations are obtained from the heap
490  heap whenever RMATCH() does a "recursion". See the macro definitions above. */  whenever RMATCH() does a "recursion". See the macro definitions above. Putting
491    the top-level on the stack rather than malloc-ing them all gives a performance
492    boost in many cases where there is not much "recursion". */
493    
494  #ifdef NO_RECURSE  #ifdef NO_RECURSE
495  heapframe *frame = (heapframe *)(pcre_stack_malloc)(sizeof(heapframe));  heapframe frame_zero;
496  if (frame == NULL) RRETURN(PCRE_ERROR_NOMEMORY);  heapframe *frame = &frame_zero;
497  frame->Xprevframe = NULL;            /* Marks the top level */  frame->Xprevframe = NULL;            /* Marks the top level */
498    
499  /* Copy in the original argument variables */  /* Copy in the original argument variables */
# Line 506  frame->Xprevframe = NULL;            /* Line 501  frame->Xprevframe = NULL;            /*
501  frame->Xeptr = eptr;  frame->Xeptr = eptr;
502  frame->Xecode = ecode;  frame->Xecode = ecode;
503  frame->Xmstart = mstart;  frame->Xmstart = mstart;
 frame->Xmarkptr = markptr;  
504  frame->Xoffset_top = offset_top;  frame->Xoffset_top = offset_top;
505  frame->Xeptrb = eptrb;  frame->Xeptrb = eptrb;
506  frame->Xrdepth = rdepth;  frame->Xrdepth = rdepth;
# Line 520  HEAP_RECURSE: Line 514  HEAP_RECURSE:
514  #define eptr               frame->Xeptr  #define eptr               frame->Xeptr
515  #define ecode              frame->Xecode  #define ecode              frame->Xecode
516  #define mstart             frame->Xmstart  #define mstart             frame->Xmstart
 #define markptr            frame->Xmarkptr  
517  #define offset_top         frame->Xoffset_top  #define offset_top         frame->Xoffset_top
518  #define eptrb              frame->Xeptrb  #define eptrb              frame->Xeptrb
519  #define rdepth             frame->Xrdepth  #define rdepth             frame->Xrdepth
520    
521  /* Ditto for the local variables */  /* Ditto for the local variables */
522    
523  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
524  #define charptr            frame->Xcharptr  #define charptr            frame->Xcharptr
525  #endif  #endif
526  #define callpat            frame->Xcallpat  #define callpat            frame->Xcallpat
# Line 585  declarations can be cut out in a block. Line 578  declarations can be cut out in a block.
578  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
579  to RMATCH(). */  to RMATCH(). */
580    
581  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
582  const uschar *charptr;  const pcre_uchar *charptr;
583  #endif  #endif
584  const uschar *callpat;  const pcre_uchar *callpat;
585  const uschar *data;  const pcre_uchar *data;
586  const uschar *next;  const pcre_uchar *next;
587  USPTR         pp;  PCRE_PUCHAR       pp;
588  const uschar *prev;  const pcre_uchar *prev;
589  USPTR         saved_eptr;  PCRE_PUCHAR       saved_eptr;
590    
591  recursion_info new_recursive;  recursion_info new_recursive;
592    
# Line 606  int prop_type; Line 599  int prop_type;
599  int prop_value;  int prop_value;
600  int prop_fail_result;  int prop_fail_result;
601  int oclength;  int oclength;
602  uschar occhars[8];  pcre_uchar occhars[6];
603  #endif  #endif
604    
605  int codelink;  int codelink;
# Line 622  int save_offset1, save_offset2, save_off Line 615  int save_offset1, save_offset2, save_off
615  int stacksave[REC_STACK_SAVE_MAX];  int stacksave[REC_STACK_SAVE_MAX];
616    
617  eptrblock newptrb;  eptrblock newptrb;
618    
619    /* There is a special fudge for calling match() in a way that causes it to
620    measure the size of its basic stack frame when the stack is being used for
621    recursion. The first argument (eptr) points to a pointer that is used
622    "statically" for doing the calculation. The second argument (ecode) being NULL
623    triggers this behaviour. It cannot normally every be NULL. The return is the
624    negated value of the frame size. */
625    
626    if (ecode == NULL)
627      {
628      char **aptr = (char **)eptr;
629      if (rdepth == 0)
630        {
631        *aptr = (char *)&rdepth;
632        return match(eptr, NULL, NULL, 0, NULL, NULL, 1);
633        }
634      else
635        {
636        int len = (char *)&rdepth - *aptr;
637        return (len > 0)? -len : len;
638        }
639      }
640  #endif     /* NO_RECURSE */  #endif     /* NO_RECURSE */
641    
642  /* To save space on the stack and in the heap frame, I have doubled up on some  /* To save space on the stack and in the heap frame, I have doubled up on some
# Line 634  the alternative names that are used. */ Line 649  the alternative names that are used. */
649  #define code_offset   codelink  #define code_offset   codelink
650  #define condassert    condition  #define condassert    condition
651  #define matched_once  prev_is_word  #define matched_once  prev_is_word
652    #define foc           number
653    #define save_mark     data
654    
655  /* These statements are here to stop the compiler complaining about unitialized  /* These statements are here to stop the compiler complaining about unitialized
656  variables. */  variables. */
# Line 659  defined). However, RMATCH isn't like a f Line 676  defined). However, RMATCH isn't like a f
676  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,
677  however, impact performance when true recursion is being used. */  however, impact performance when true recursion is being used. */
678    
679  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
680  utf8 = md->utf8;       /* Local copy of the flag */  utf = md->utf;       /* Local copy of the flag */
681  #else  #else
682  utf8 = FALSE;  utf = FALSE;
683  #endif  #endif
684    
685  /* 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 718  for (;;)
718    switch(op)    switch(op)
719      {      {
720      case OP_MARK:      case OP_MARK:
721      markptr = ecode + 2;      md->nomatch_mark = ecode + 2;
722      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      md->mark = NULL;    /* In case previously set by assertion */
723        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
724        eptrb, RM55);        eptrb, RM55);
725        if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
726             md->mark == NULL) md->mark = ecode + 2;
727    
728      /* 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
729      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 732  for (;;)
732      position and return MATCH_SKIP. Otherwise, pass back the return code      position and return MATCH_SKIP. Otherwise, pass back the return code
733      unaltered. */      unaltered. */
734    
735      if (rrc == MATCH_SKIP_ARG &&      else if (rrc == MATCH_SKIP_ARG &&
736          strcmp((char *)markptr, (char *)(md->start_match_ptr)) == 0)          STRCMP_UC_UC(ecode + 2, md->start_match_ptr) == 0)
737        {        {
738        md->start_match_ptr = eptr;        md->start_match_ptr = eptr;
739        RRETURN(MATCH_SKIP);        RRETURN(MATCH_SKIP);
740        }        }
   
     if (md->mark == NULL) md->mark = markptr;  
741      RRETURN(rrc);      RRETURN(rrc);
742    
743      case OP_FAIL:      case OP_FAIL:
744      MRRETURN(MATCH_NOMATCH);      RRETURN(MATCH_NOMATCH);
745    
746      /* COMMIT overrides PRUNE, SKIP, and THEN */      /* COMMIT overrides PRUNE, SKIP, and THEN */
747    
748      case OP_COMMIT:      case OP_COMMIT:
749      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
750        eptrb, RM52);        eptrb, RM52);
751      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE &&      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE &&
752          rrc != MATCH_SKIP && rrc != MATCH_SKIP_ARG &&          rrc != MATCH_SKIP && rrc != MATCH_SKIP_ARG &&
753          rrc != MATCH_THEN)          rrc != MATCH_THEN)
754        RRETURN(rrc);        RRETURN(rrc);
755      MRRETURN(MATCH_COMMIT);      RRETURN(MATCH_COMMIT);
756    
757      /* PRUNE overrides THEN */      /* PRUNE overrides THEN */
758    
759      case OP_PRUNE:      case OP_PRUNE:
760      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
761        eptrb, RM51);        eptrb, RM51);
762      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
763      MRRETURN(MATCH_PRUNE);      RRETURN(MATCH_PRUNE);
764    
765      case OP_PRUNE_ARG:      case OP_PRUNE_ARG:
766      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      md->nomatch_mark = ecode + 2;
767        md->mark = NULL;    /* In case previously set by assertion */
768        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
769        eptrb, RM56);        eptrb, RM56);
770        if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
771             md->mark == NULL) md->mark = ecode + 2;
772      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
     md->mark = ecode + 2;  
773      RRETURN(MATCH_PRUNE);      RRETURN(MATCH_PRUNE);
774    
775      /* SKIP overrides PRUNE and THEN */      /* SKIP overrides PRUNE and THEN */
776    
777      case OP_SKIP:      case OP_SKIP:
778      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
779        eptrb, RM53);        eptrb, RM53);
780      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)
781        RRETURN(rrc);        RRETURN(rrc);
782      md->start_match_ptr = eptr;   /* Pass back current position */      md->start_match_ptr = eptr;   /* Pass back current position */
783      MRRETURN(MATCH_SKIP);      RRETURN(MATCH_SKIP);
784    
785        /* Note that, for Perl compatibility, SKIP with an argument does NOT set
786        nomatch_mark. There is a flag that disables this opcode when re-matching a
787        pattern that ended with a SKIP for which there was not a matching MARK. */
788    
789      case OP_SKIP_ARG:      case OP_SKIP_ARG:
790      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      if (md->ignore_skip_arg)
791          {
792          ecode += PRIV(OP_lengths)[*ecode] + ecode[1];
793          break;
794          }
795        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
796        eptrb, RM57);        eptrb, RM57);
797      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)
798        RRETURN(rrc);        RRETURN(rrc);
799    
800      /* 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
801      returning the special MATCH_SKIP_ARG return code. This will either be      returning the special MATCH_SKIP_ARG return code. This will either be
802      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
803      as PRUNE. */      with the md->ignore_skip_arg flag set. */
804    
805      md->start_match_ptr = ecode + 2;      md->start_match_ptr = ecode + 2;
806      RRETURN(MATCH_SKIP_ARG);      RRETURN(MATCH_SKIP_ARG);
807    
808      /* 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
809      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
810      to skip back past alternatives that precede the THEN within the current      match pointer to do this. */
     branch. */  
811    
812      case OP_THEN:      case OP_THEN:
813      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
814        eptrb, RM54);        eptrb, RM54);
815      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
816      md->start_match_ptr = ecode - GET(ecode, 1);      md->start_match_ptr = ecode;
817      MRRETURN(MATCH_THEN);      RRETURN(MATCH_THEN);
818    
819      case OP_THEN_ARG:      case OP_THEN_ARG:
820      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1+LINK_SIZE],      md->nomatch_mark = ecode + 2;
821        offset_top, md, eptrb, RM58);      md->mark = NULL;    /* In case previously set by assertion */
822        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top,
823          md, eptrb, RM58);
824        if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
825             md->mark == NULL) md->mark = ecode + 2;
826      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
827      md->start_match_ptr = ecode - GET(ecode, 1);      md->start_match_ptr = ecode;
     md->mark = ecode + LINK_SIZE + 2;  
828      RRETURN(MATCH_THEN);      RRETURN(MATCH_THEN);
829    
830        /* Handle an atomic group that does not contain any capturing parentheses.
831        This can be handled like an assertion. Prior to 8.13, all atomic groups
832        were handled this way. In 8.13, the code was changed as below for ONCE, so
833        that backups pass through the group and thereby reset captured values.
834        However, this uses a lot more stack, so in 8.20, atomic groups that do not
835        contain any captures generate OP_ONCE_NC, which can be handled in the old,
836        less stack intensive way.
837    
838        Check the alternative branches in turn - the matching won't pass the KET
839        for this kind of subpattern. If any one branch matches, we carry on as at
840        the end of a normal bracket, leaving the subject pointer, but resetting
841        the start-of-match value in case it was changed by \K. */
842    
843        case OP_ONCE_NC:
844        prev = ecode;
845        saved_eptr = eptr;
846        save_mark = md->mark;
847        do
848          {
849          RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM64);
850          if (rrc == MATCH_MATCH)  /* Note: _not_ MATCH_ACCEPT */
851            {
852            mstart = md->start_match_ptr;
853            break;
854            }
855          if (rrc == MATCH_THEN)
856            {
857            next = ecode + GET(ecode,1);
858            if (md->start_match_ptr < next &&
859                (*ecode == OP_ALT || *next == OP_ALT))
860              rrc = MATCH_NOMATCH;
861            }
862    
863          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
864          ecode += GET(ecode,1);
865          md->mark = save_mark;
866          }
867        while (*ecode == OP_ALT);
868    
869        /* If hit the end of the group (which could be repeated), fail */
870    
871        if (*ecode != OP_ONCE_NC && *ecode != OP_ALT) RRETURN(MATCH_NOMATCH);
872    
873        /* Continue as from after the group, updating the offsets high water
874        mark, since extracts may have been taken. */
875    
876        do ecode += GET(ecode, 1); while (*ecode == OP_ALT);
877    
878        offset_top = md->end_offset_top;
879        eptr = md->end_match_ptr;
880    
881        /* For a non-repeating ket, just continue at this level. This also
882        happens for a repeating ket if no characters were matched in the group.
883        This is the forcible breaking of infinite loops as implemented in Perl
884        5.005. */
885    
886        if (*ecode == OP_KET || eptr == saved_eptr)
887          {
888          ecode += 1+LINK_SIZE;
889          break;
890          }
891    
892        /* The repeating kets try the rest of the pattern or restart from the
893        preceding bracket, in the appropriate order. The second "call" of match()
894        uses tail recursion, to avoid using another stack frame. */
895    
896        if (*ecode == OP_KETRMIN)
897          {
898          RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM65);
899          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
900          ecode = prev;
901          goto TAIL_RECURSE;
902          }
903        else  /* OP_KETRMAX */
904          {
905          md->match_function_type = MATCH_CBEGROUP;
906          RMATCH(eptr, prev, offset_top, md, eptrb, RM66);
907          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
908          ecode += 1 + LINK_SIZE;
909          goto TAIL_RECURSE;
910          }
911        /* Control never gets here */
912    
913      /* Handle a capturing bracket, other than those that are possessive with an      /* Handle a capturing bracket, other than those that are possessive with an
914      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
915      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 942  for (;;)
942        save_offset2 = md->offset_vector[offset+1];        save_offset2 = md->offset_vector[offset+1];
943        save_offset3 = md->offset_vector[md->offset_end - number];        save_offset3 = md->offset_vector[md->offset_end - number];
944        save_capture_last = md->capture_last;        save_capture_last = md->capture_last;
945          save_mark = md->mark;
946    
947        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));
948        md->offset_vector[md->offset_end - number] =        md->offset_vector[md->offset_end - number] =
# Line 835  for (;;) Line 951  for (;;)
951        for (;;)        for (;;)
952          {          {
953          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
954          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
955            eptrb, RM1);            eptrb, RM1);
956          if (rrc == MATCH_ONCE) break;  /* Backing up through an atomic group */          if (rrc == MATCH_ONCE) break;  /* Backing up through an atomic group */
957          if (rrc != MATCH_NOMATCH &&  
958              (rrc != MATCH_THEN || md->start_match_ptr != ecode))          /* If we backed up to a THEN, check whether it is within the current
959            RRETURN(rrc);          branch by comparing the address of the THEN that is passed back with
960            the end of the branch. If it is within the current branch, and the
961            branch is one of two or more alternatives (it either starts or ends
962            with OP_ALT), we have reached the limit of THEN's action, so convert
963            the return code to NOMATCH, which will cause normal backtracking to
964            happen from now on. Otherwise, THEN is passed back to an outer
965            alternative. This implements Perl's treatment of parenthesized groups,
966            where a group not containing | does not affect the current alternative,
967            that is, (X) is NOT the same as (X|(*F)). */
968    
969            if (rrc == MATCH_THEN)
970              {
971              next = ecode + GET(ecode,1);
972              if (md->start_match_ptr < next &&
973                  (*ecode == OP_ALT || *next == OP_ALT))
974                rrc = MATCH_NOMATCH;
975              }
976    
977            /* Anything other than NOMATCH is passed back. */
978    
979            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
980          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
981          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
982            md->mark = save_mark;
983          if (*ecode != OP_ALT) break;          if (*ecode != OP_ALT) break;
984          }          }
985    
# Line 851  for (;;) Line 988  for (;;)
988        md->offset_vector[offset+1] = save_offset2;        md->offset_vector[offset+1] = save_offset2;
989        md->offset_vector[md->offset_end - number] = save_offset3;        md->offset_vector[md->offset_end - number] = save_offset3;
990    
991        /* 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. */  
992    
993        if (rrc != MATCH_THEN && md->mark == NULL) md->mark = markptr;        RRETURN(rrc);
       RRETURN(((rrc == MATCH_ONCE)? MATCH_ONCE:MATCH_NOMATCH));  
994        }        }
995    
996      /* FALL THROUGH ... Insufficient room for saving captured contents. Treat      /* FALL THROUGH ... Insufficient room for saving captured contents. Treat
# Line 870  for (;;) Line 1005  for (;;)
1005      /* VVVVVVVVVVVVVVVVVVVVVVVVV */      /* VVVVVVVVVVVVVVVVVVVVVVVVV */
1006    
1007      /* Non-capturing or atomic group, except for possessive with unlimited      /* Non-capturing or atomic group, except for possessive with unlimited
1008      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.
1009      within the brackets, we used to return the result of a recursive call to  
1010      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
1011      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
1012      group. However, now that there is the possiblity of (*THEN) occurring in      possible to reduce stack usage by turning this into a tail recursion,
1013      the final alternative, this optimization is no longer possible.      except in the case of a possibly empty group. However, now that there is
1014        the possiblity of (*THEN) occurring in the final alternative, this
1015        optimization is no longer always possible.
1016    
1017        We can optimize if we know there are no (*THEN)s in the pattern; at present
1018        this is the best that can be done.
1019    
1020      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
1021      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 1032  for (;;)
1032      for (;;)      for (;;)
1033        {        {
1034        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;
1035        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, eptrb,  
1036          /* If this is not a possibly empty group, and there are no (*THEN)s in
1037          the pattern, and this is the final alternative, optimize as described
1038          above. */
1039    
1040          else if (!md->hasthen && ecode[GET(ecode, 1)] != OP_ALT)
1041            {
1042            ecode += PRIV(OP_lengths)[*ecode];
1043            goto TAIL_RECURSE;
1044            }
1045    
1046          /* In all other cases, we have to make another call to match(). */
1047    
1048          save_mark = md->mark;
1049          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb,
1050          RM2);          RM2);
1051        if (rrc != MATCH_NOMATCH &&  
1052            (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* See comment in the code for capturing groups above about handling
1053          THEN. */
1054    
1055          if (rrc == MATCH_THEN)
1056            {
1057            next = ecode + GET(ecode,1);
1058            if (md->start_match_ptr < next &&
1059                (*ecode == OP_ALT || *next == OP_ALT))
1060              rrc = MATCH_NOMATCH;
1061            }
1062    
1063          if (rrc != MATCH_NOMATCH)
1064          {          {
1065          if (rrc == MATCH_ONCE)          if (rrc == MATCH_ONCE)
1066            {            {
1067            const uschar *scode = ecode;            const pcre_uchar *scode = ecode;
1068            if (*scode != OP_ONCE)           /* If not at start, find it */            if (*scode != OP_ONCE)           /* If not at start, find it */
1069              {              {
1070              while (*scode == OP_ALT) scode += GET(scode, 1);              while (*scode == OP_ALT) scode += GET(scode, 1);
# Line 910  for (;;) Line 1075  for (;;)
1075          RRETURN(rrc);          RRETURN(rrc);
1076          }          }
1077        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1078          md->mark = save_mark;
1079        if (*ecode != OP_ALT) break;        if (*ecode != OP_ALT) break;
1080        }        }
1081      if (rrc != MATCH_THEN && md->mark == NULL) md->mark = markptr;  
1082      RRETURN(MATCH_NOMATCH);      RRETURN(MATCH_NOMATCH);
1083    
1084      /* Handle possessive capturing brackets with an unlimited repeat. We come      /* Handle possessive capturing brackets with an unlimited repeat. We come
# Line 941  for (;;) Line 1107  for (;;)
1107      if (offset < md->offset_max)      if (offset < md->offset_max)
1108        {        {
1109        matched_once = FALSE;        matched_once = FALSE;
1110        code_offset = ecode - md->start_code;        code_offset = (int)(ecode - md->start_code);
1111    
1112        save_offset1 = md->offset_vector[offset];        save_offset1 = md->offset_vector[offset];
1113        save_offset2 = md->offset_vector[offset+1];        save_offset2 = md->offset_vector[offset+1];
# Line 964  for (;;) Line 1130  for (;;)
1130          md->offset_vector[md->offset_end - number] =          md->offset_vector[md->offset_end - number] =
1131            (int)(eptr - md->start_subject);            (int)(eptr - md->start_subject);
1132          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
1133          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
1134            eptrb, RM63);            eptrb, RM63);
1135          if (rrc == MATCH_KETRPOS)          if (rrc == MATCH_KETRPOS)
1136            {            {
# Line 975  for (;;) Line 1141  for (;;)
1141            matched_once = TRUE;            matched_once = TRUE;
1142            continue;            continue;
1143            }            }
1144          if (rrc != MATCH_NOMATCH &&  
1145              (rrc != MATCH_THEN || md->start_match_ptr != ecode))          /* See comment in the code for capturing groups above about handling
1146            RRETURN(rrc);          THEN. */
1147    
1148            if (rrc == MATCH_THEN)
1149              {
1150              next = ecode + GET(ecode,1);
1151              if (md->start_match_ptr < next &&
1152                  (*ecode == OP_ALT || *next == OP_ALT))
1153                rrc = MATCH_NOMATCH;
1154              }
1155    
1156            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1157          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
1158          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
1159          if (*ecode != OP_ALT) break;          if (*ecode != OP_ALT) break;
# Line 990  for (;;) Line 1166  for (;;)
1166          md->offset_vector[md->offset_end - number] = save_offset3;          md->offset_vector[md->offset_end - number] = save_offset3;
1167          }          }
1168    
       if (rrc != MATCH_THEN && md->mark == NULL) md->mark = markptr;  
1169        if (allow_zero || matched_once)        if (allow_zero || matched_once)
1170          {          {
1171          ecode += 1 + LINK_SIZE;          ecode += 1 + LINK_SIZE;
# Line 1022  for (;;) Line 1197  for (;;)
1197    
1198      POSSESSIVE_NON_CAPTURE:      POSSESSIVE_NON_CAPTURE:
1199      matched_once = FALSE;      matched_once = FALSE;
1200      code_offset = ecode - md->start_code;      code_offset = (int)(ecode - md->start_code);
1201    
1202      for (;;)      for (;;)
1203        {        {
1204        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
1205        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
1206          eptrb, RM48);          eptrb, RM48);
1207        if (rrc == MATCH_KETRPOS)        if (rrc == MATCH_KETRPOS)
1208          {          {
# Line 1037  for (;;) Line 1212  for (;;)
1212          matched_once = TRUE;          matched_once = TRUE;
1213          continue;          continue;
1214          }          }
1215        if (rrc != MATCH_NOMATCH &&  
1216            (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* See comment in the code for capturing groups above about handling
1217          RRETURN(rrc);        THEN. */
1218    
1219          if (rrc == MATCH_THEN)
1220            {
1221            next = ecode + GET(ecode,1);
1222            if (md->start_match_ptr < next &&
1223                (*ecode == OP_ALT || *next == OP_ALT))
1224              rrc = MATCH_NOMATCH;
1225            }
1226    
1227          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1228        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1229        if (*ecode != OP_ALT) break;        if (*ecode != OP_ALT) break;
1230        }        }
# Line 1067  for (;;) Line 1252  for (;;)
1252    
1253      if (ecode[LINK_SIZE+1] == OP_CALLOUT)      if (ecode[LINK_SIZE+1] == OP_CALLOUT)
1254        {        {
1255        if (pcre_callout != NULL)        if (PUBL(callout) != NULL)
1256          {          {
1257          pcre_callout_block cb;          PUBL(callout_block) cb;
1258          cb.version          = 2;   /* Version 1 of the callout block */          cb.version          = 2;   /* Version 1 of the callout block */
1259          cb.callout_number   = ecode[LINK_SIZE+2];          cb.callout_number   = ecode[LINK_SIZE+2];
1260          cb.offset_vector    = md->offset_vector;          cb.offset_vector    = md->offset_vector;
1261    #ifdef COMPILE_PCRE8
1262          cb.subject          = (PCRE_SPTR)md->start_subject;          cb.subject          = (PCRE_SPTR)md->start_subject;
1263    #else
1264            cb.subject          = (PCRE_SPTR16)md->start_subject;
1265    #endif
1266          cb.subject_length   = (int)(md->end_subject - md->start_subject);          cb.subject_length   = (int)(md->end_subject - md->start_subject);
1267          cb.start_match      = (int)(mstart - md->start_subject);          cb.start_match      = (int)(mstart - md->start_subject);
1268          cb.current_position = (int)(eptr - md->start_subject);          cb.current_position = (int)(eptr - md->start_subject);
# Line 1082  for (;;) Line 1271  for (;;)
1271          cb.capture_top      = offset_top/2;          cb.capture_top      = offset_top/2;
1272          cb.capture_last     = md->capture_last;          cb.capture_last     = md->capture_last;
1273          cb.callout_data     = md->callout_data;          cb.callout_data     = md->callout_data;
1274          cb.mark             = markptr;          cb.mark             = md->nomatch_mark;
1275          if ((rrc = (*pcre_callout)(&cb)) > 0) MRRETURN(MATCH_NOMATCH);          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1276          if (rrc < 0) RRETURN(rrc);          if (rrc < 0) RRETURN(rrc);
1277          }          }
1278        ecode += _pcre_OP_lengths[OP_CALLOUT];        ecode += PRIV(OP_lengths)[OP_CALLOUT];
1279        }        }
1280    
1281      condcode = ecode[LINK_SIZE+1];      condcode = ecode[LINK_SIZE+1];
# Line 1103  for (;;) Line 1292  for (;;)
1292        else        else
1293          {          {
1294          int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/          int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/
1295          condition =  (recno == RREF_ANY || recno == md->recursive->group_num);          condition = (recno == RREF_ANY || recno == md->recursive->group_num);
1296    
1297          /* 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
1298          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
1299          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
1300          if any one is set. */          if any one is set. */
1301    
1302          if (!condition && condcode == OP_NRREF && recno != RREF_ANY)          if (!condition && condcode == OP_NRREF)
1303            {            {
1304            uschar *slotA = md->name_table;            pcre_uchar *slotA = md->name_table;
1305            for (i = 0; i < md->name_count; i++)            for (i = 0; i < md->name_count; i++)
1306              {              {
1307              if (GET2(slotA, 0) == recno) break;              if (GET2(slotA, 0) == recno) break;
# Line 1125  for (;;) Line 1314  for (;;)
1314    
1315            if (i < md->name_count)            if (i < md->name_count)
1316              {              {
1317              uschar *slotB = slotA;              pcre_uchar *slotB = slotA;
1318              while (slotB > md->name_table)              while (slotB > md->name_table)
1319                {                {
1320                slotB -= md->name_entry_size;                slotB -= md->name_entry_size;
1321                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1322                  {                  {
1323                  condition = GET2(slotB, 0) == md->recursive->group_num;                  condition = GET2(slotB, 0) == md->recursive->group_num;
1324                  if (condition) break;                  if (condition) break;
# Line 1145  for (;;) Line 1334  for (;;)
1334                for (i++; i < md->name_count; i++)                for (i++; i < md->name_count; i++)
1335                  {                  {
1336                  slotB += md->name_entry_size;                  slotB += md->name_entry_size;
1337                  if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                  if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1338                    {                    {
1339                    condition = GET2(slotB, 0) == md->recursive->group_num;                    condition = GET2(slotB, 0) == md->recursive->group_num;
1340                    if (condition) break;                    if (condition) break;
# Line 1158  for (;;) Line 1347  for (;;)
1347    
1348          /* Chose branch according to the condition */          /* Chose branch according to the condition */
1349    
1350          ecode += condition? 3 : GET(ecode, 1);          ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
1351          }          }
1352        }        }
1353    
# Line 1175  for (;;) Line 1364  for (;;)
1364        if (!condition && condcode == OP_NCREF)        if (!condition && condcode == OP_NCREF)
1365          {          {
1366          int refno = offset >> 1;          int refno = offset >> 1;
1367          uschar *slotA = md->name_table;          pcre_uchar *slotA = md->name_table;
1368    
1369          for (i = 0; i < md->name_count; i++)          for (i = 0; i < md->name_count; i++)
1370            {            {
# Line 1189  for (;;) Line 1378  for (;;)
1378    
1379          if (i < md->name_count)          if (i < md->name_count)
1380            {            {
1381            uschar *slotB = slotA;            pcre_uchar *slotB = slotA;
1382            while (slotB > md->name_table)            while (slotB > md->name_table)
1383              {              {
1384              slotB -= md->name_entry_size;              slotB -= md->name_entry_size;
1385              if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)              if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1386                {                {
1387                offset = GET2(slotB, 0) << 1;                offset = GET2(slotB, 0) << 1;
1388                condition = offset < offset_top &&                condition = offset < offset_top &&
# Line 1211  for (;;) Line 1400  for (;;)
1400              for (i++; i < md->name_count; i++)              for (i++; i < md->name_count; i++)
1401                {                {
1402                slotB += md->name_entry_size;                slotB += md->name_entry_size;
1403                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1404                  {                  {
1405                  offset = GET2(slotB, 0) << 1;                  offset = GET2(slotB, 0) << 1;
1406                  condition = offset < offset_top &&                  condition = offset < offset_top &&
# Line 1226  for (;;) Line 1415  for (;;)
1415    
1416        /* Chose branch according to the condition */        /* Chose branch according to the condition */
1417    
1418        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
1419        }        }
1420    
1421      else if (condcode == OP_DEF)     /* DEFINE - always false */      else if (condcode == OP_DEF)     /* DEFINE - always false */
# Line 1251  for (;;) Line 1440  for (;;)
1440          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
1441          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
1442          }          }
1443        else if (rrc != MATCH_NOMATCH &&  
1444                (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* PCRE doesn't allow the effect of (*THEN) to escape beyond an
1445          assertion; it is therefore treated as NOMATCH. */
1446    
1447          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1448          {          {
1449          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
1450          }          }
# Line 1263  for (;;) Line 1455  for (;;)
1455          }          }
1456        }        }
1457    
1458      /* 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
1459      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
1460      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
1461      strategy no longer works because of the possibilty of (*THEN) being      call to match() is always required, unless the second alternative doesn't
1462      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
1463      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
1464      plough on. */      alternatives. If a THEN is encountered in the branch, it propagates out to
1465        the enclosing alternative (unless nested in a deeper set of alternatives,
1466        of course). */
1467    
1468      if (condition || *ecode == OP_ALT)      if (condition || *ecode == OP_ALT)
1469        {        {
1470        if (op == OP_SCOND) md->match_function_type = MATCH_CBEGROUP;        if (op != OP_SCOND)
1471            {
1472            ecode += 1 + LINK_SIZE;
1473            goto TAIL_RECURSE;
1474            }
1475    
1476          md->match_function_type = MATCH_CBEGROUP;
1477        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;  
1478        RRETURN(rrc);        RRETURN(rrc);
1479        }        }
1480      else                         /* Condition false & no alternative */  
1481         /* Condition false & no alternative; continue after the group. */
1482    
1483        else
1484        {        {
1485        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1486        }        }
# Line 1306  for (;;) Line 1507  for (;;)
1507        md->offset_vector[offset+1] = (int)(eptr - md->start_subject);        md->offset_vector[offset+1] = (int)(eptr - md->start_subject);
1508        if (offset_top <= offset) offset_top = offset + 2;        if (offset_top <= offset) offset_top = offset + 2;
1509        }        }
1510      ecode += 3;      ecode += 1 + IMM2_SIZE;
1511      break;      break;
1512    
1513    
# Line 1326  for (;;) Line 1527  for (;;)
1527           (md->notempty ||           (md->notempty ||
1528             (md->notempty_atstart &&             (md->notempty_atstart &&
1529               mstart == md->start_subject + md->start_offset)))               mstart == md->start_subject + md->start_offset)))
1530        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
1531    
1532      /* Otherwise, we have a match. */      /* Otherwise, we have a match. */
1533    
# Line 1335  for (;;) Line 1536  for (;;)
1536      md->start_match_ptr = mstart;       /* and the start (\K can modify) */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
1537    
1538      /* 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
1539      given as the argument to MRRETURN when the heap is in use. */      given as the argument to RRETURN when the heap is in use. */
1540    
1541      rrc = (op == OP_END)? MATCH_MATCH : MATCH_ACCEPT;      rrc = (op == OP_END)? MATCH_MATCH : MATCH_ACCEPT;
1542      MRRETURN(rrc);      RRETURN(rrc);
1543    
1544      /* Assertion brackets. Check the alternative branches in turn - the      /* Assertion brackets. Check the alternative branches in turn - the
1545      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 1353  for (;;) Line 1554  for (;;)
1554    
1555      case OP_ASSERT:      case OP_ASSERT:
1556      case OP_ASSERTBACK:      case OP_ASSERTBACK:
1557        save_mark = md->mark;
1558      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1559        {        {
1560        condassert = TRUE;        condassert = TRUE;
# Line 1366  for (;;) Line 1568  for (;;)
1568        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
1569          {          {
1570          mstart = md->start_match_ptr;   /* In case \K reset it */          mstart = md->start_match_ptr;   /* In case \K reset it */
         markptr = md->mark;  
1571          break;          break;
1572          }          }
1573        if (rrc != MATCH_NOMATCH &&  
1574            (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* PCRE does not allow THEN to escape beyond an assertion; it is treated
1575          RRETURN(rrc);        as NOMATCH. */
1576    
1577          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1578        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1579          md->mark = save_mark;
1580        }        }
1581      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
1582    
1583      if (*ecode == OP_KET) MRRETURN(MATCH_NOMATCH);      if (*ecode == OP_KET) RRETURN(MATCH_NOMATCH);
1584    
1585      /* If checking an assertion for a condition, return MATCH_MATCH. */      /* If checking an assertion for a condition, return MATCH_MATCH. */
1586    
# Line 1396  for (;;) Line 1600  for (;;)
1600    
1601      case OP_ASSERT_NOT:      case OP_ASSERT_NOT:
1602      case OP_ASSERTBACK_NOT:      case OP_ASSERTBACK_NOT:
1603        save_mark = md->mark;
1604      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1605        {        {
1606        condassert = TRUE;        condassert = TRUE;
# Line 1406  for (;;) Line 1611  for (;;)
1611      do      do
1612        {        {
1613        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);
1614        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) MRRETURN(MATCH_NOMATCH);        md->mark = save_mark;
1615          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) RRETURN(MATCH_NOMATCH);
1616        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)
1617          {          {
1618          do ecode += GET(ecode,1); while (*ecode == OP_ALT);          do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1619          break;          break;
1620          }          }
1621        if (rrc != MATCH_NOMATCH &&  
1622            (rrc != MATCH_THEN || md->start_match_ptr != ecode))        /* PCRE does not allow THEN to escape beyond an assertion; it is treated
1623          RRETURN(rrc);        as NOMATCH. */
1624    
1625          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1626        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1627        }        }
1628      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 1430  for (;;) Line 1638  for (;;)
1638      back a number of characters, not bytes. */      back a number of characters, not bytes. */
1639    
1640      case OP_REVERSE:      case OP_REVERSE:
1641  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
1642      if (utf8)      if (utf)
1643        {        {
1644        i = GET(ecode, 1);        i = GET(ecode, 1);
1645        while (i-- > 0)        while (i-- > 0)
1646          {          {
1647          eptr--;          eptr--;
1648          if (eptr < md->start_subject) MRRETURN(MATCH_NOMATCH);          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1649          BACKCHAR(eptr);          BACKCHAR(eptr);
1650          }          }
1651        }        }
# Line 1448  for (;;) Line 1656  for (;;)
1656    
1657        {        {
1658        eptr -= GET(ecode, 1);        eptr -= GET(ecode, 1);
1659        if (eptr < md->start_subject) MRRETURN(MATCH_NOMATCH);        if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1660        }        }
1661    
1662      /* Save the earliest consulted character, then skip to next op code */      /* Save the earliest consulted character, then skip to next op code */
# Line 1462  for (;;) Line 1670  for (;;)
1670      function is able to force a failure. */      function is able to force a failure. */
1671    
1672      case OP_CALLOUT:      case OP_CALLOUT:
1673      if (pcre_callout != NULL)      if (PUBL(callout) != NULL)
1674        {        {
1675        pcre_callout_block cb;        PUBL(callout_block) cb;
1676        cb.version          = 2;   /* Version 1 of the callout block */        cb.version          = 2;   /* Version 1 of the callout block */
1677        cb.callout_number   = ecode[1];        cb.callout_number   = ecode[1];
1678        cb.offset_vector    = md->offset_vector;        cb.offset_vector    = md->offset_vector;
1679    #ifdef COMPILE_PCRE8
1680        cb.subject          = (PCRE_SPTR)md->start_subject;        cb.subject          = (PCRE_SPTR)md->start_subject;
1681    #else
1682          cb.subject          = (PCRE_SPTR16)md->start_subject;
1683    #endif
1684        cb.subject_length   = (int)(md->end_subject - md->start_subject);        cb.subject_length   = (int)(md->end_subject - md->start_subject);
1685        cb.start_match      = (int)(mstart - md->start_subject);        cb.start_match      = (int)(mstart - md->start_subject);
1686        cb.current_position = (int)(eptr - md->start_subject);        cb.current_position = (int)(eptr - md->start_subject);
# Line 1477  for (;;) Line 1689  for (;;)
1689        cb.capture_top      = offset_top/2;        cb.capture_top      = offset_top/2;
1690        cb.capture_last     = md->capture_last;        cb.capture_last     = md->capture_last;
1691        cb.callout_data     = md->callout_data;        cb.callout_data     = md->callout_data;
1692        cb.mark             = markptr;        cb.mark             = md->nomatch_mark;
1693        if ((rrc = (*pcre_callout)(&cb)) > 0) MRRETURN(MATCH_NOMATCH);        if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1694        if (rrc < 0) RRETURN(rrc);        if (rrc < 0) RRETURN(rrc);
1695        }        }
1696      ecode += 2 + 2*LINK_SIZE;      ecode += 2 + 2*LINK_SIZE;
# Line 1537  for (;;) Line 1749  for (;;)
1749        else        else
1750          {          {
1751          new_recursive.offset_save =          new_recursive.offset_save =
1752            (int *)(pcre_malloc)(new_recursive.saved_max * sizeof(int));            (int *)(PUBL(malloc))(new_recursive.saved_max * sizeof(int));
1753          if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY);          if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY);
1754          }          }
1755        memcpy(new_recursive.offset_save, md->offset_vector,        memcpy(new_recursive.offset_save, md->offset_vector,
# Line 1552  for (;;) Line 1764  for (;;)
1764        do        do
1765          {          {
1766          if (cbegroup) md->match_function_type = MATCH_CBEGROUP;          if (cbegroup) md->match_function_type = MATCH_CBEGROUP;
1767          RMATCH(eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,          RMATCH(eptr, callpat + PRIV(OP_lengths)[*callpat], offset_top,
1768            md, eptrb, RM6);            md, eptrb, RM6);
1769          memcpy(md->offset_vector, new_recursive.offset_save,          memcpy(md->offset_vector, new_recursive.offset_save,
1770              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
1771            md->recursive = new_recursive.prevrec;
1772          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
1773            {            {
1774            DPRINTF(("Recursion matched\n"));            DPRINTF(("Recursion matched\n"));
           md->recursive = new_recursive.prevrec;  
1775            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
1776              (pcre_free)(new_recursive.offset_save);              (PUBL(free))(new_recursive.offset_save);
1777    
1778            /* 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
1779            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 1571  for (;;) Line 1783  for (;;)
1783            mstart = md->start_match_ptr;            mstart = md->start_match_ptr;
1784            goto RECURSION_MATCHED;        /* Exit loop; end processing */            goto RECURSION_MATCHED;        /* Exit loop; end processing */
1785            }            }
1786          else if (rrc != MATCH_NOMATCH &&  
1787                  (rrc != MATCH_THEN || md->start_match_ptr != ecode))          /* PCRE does not allow THEN to escape beyond a recursion; it is treated
1788            as NOMATCH. */
1789    
1790            else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1791            {            {
1792            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1793            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
1794              (pcre_free)(new_recursive.offset_save);              (PUBL(free))(new_recursive.offset_save);
1795            RRETURN(rrc);            RRETURN(rrc);
1796            }            }
1797    
# Line 1588  for (;;) Line 1803  for (;;)
1803        DPRINTF(("Recursion didn't match\n"));        DPRINTF(("Recursion didn't match\n"));
1804        md->recursive = new_recursive.prevrec;        md->recursive = new_recursive.prevrec;
1805        if (new_recursive.offset_save != stacksave)        if (new_recursive.offset_save != stacksave)
1806          (pcre_free)(new_recursive.offset_save);          (PUBL(free))(new_recursive.offset_save);
1807        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
1808        }        }
1809    
1810      RECURSION_MATCHED:      RECURSION_MATCHED:
# Line 1658  for (;;) Line 1873  for (;;)
1873        }        }
1874      else saved_eptr = NULL;      else saved_eptr = NULL;
1875    
1876      /* 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
1877      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
1878      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
1879      by \K. */      start in case it was changed by \K. */
1880    
1881      if (*prev == OP_ASSERT || *prev == OP_ASSERT_NOT ||      if ((*prev >= OP_ASSERT && *prev <= OP_ASSERTBACK_NOT) ||
1882          *prev == OP_ASSERTBACK || *prev == OP_ASSERTBACK_NOT)           *prev == OP_ONCE_NC)
1883        {        {
1884        md->end_match_ptr = eptr;      /* For ONCE */        md->end_match_ptr = eptr;      /* For ONCE_NC */
1885        md->end_offset_top = offset_top;        md->end_offset_top = offset_top;
1886        md->start_match_ptr = mstart;        md->start_match_ptr = mstart;
1887        MRRETURN(MATCH_MATCH);         /* Sets md->mark */        RRETURN(MATCH_MATCH);         /* Sets md->mark */
1888        }        }
1889    
1890      /* 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 1734  for (;;) Line 1949  for (;;)
1949      /* For an ordinary non-repeating ket, just continue at this level. This      /* For an ordinary non-repeating ket, just continue at this level. This
1950      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
1951      group. This is the forcible breaking of infinite loops as implemented in      group. This is the forcible breaking of infinite loops as implemented in
1952      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,
1953      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
1954      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
1955      bypassing intermediate backup points, but resetting any captures that      original OP_ONCE level, thereby bypassing intermediate backup points, but
1956      happened along the way. */      resetting any captures that happened along the way. */
1957    
1958      if (*ecode == OP_KET || eptr == saved_eptr)      if (*ecode == OP_KET || eptr == saved_eptr)
1959        {        {
# Line 1811  for (;;) Line 2026  for (;;)
2026      /* Not multiline mode: start of subject assertion, unless notbol. */      /* Not multiline mode: start of subject assertion, unless notbol. */
2027    
2028      case OP_CIRC:      case OP_CIRC:
2029      if (md->notbol && eptr == md->start_subject) MRRETURN(MATCH_NOMATCH);      if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH);
2030    
2031      /* Start of subject assertion */      /* Start of subject assertion */
2032    
2033      case OP_SOD:      case OP_SOD:
2034      if (eptr != md->start_subject) MRRETURN(MATCH_NOMATCH);      if (eptr != md->start_subject) RRETURN(MATCH_NOMATCH);
2035      ecode++;      ecode++;
2036      break;      break;
2037    
2038      /* Multiline mode: start of subject unless notbol, or after any newline. */      /* Multiline mode: start of subject unless notbol, or after any newline. */
2039    
2040      case OP_CIRCM:      case OP_CIRCM:
2041      if (md->notbol && eptr == md->start_subject) MRRETURN(MATCH_NOMATCH);      if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH);
2042      if (eptr != md->start_subject &&      if (eptr != md->start_subject &&
2043          (eptr == md->end_subject || !WAS_NEWLINE(eptr)))          (eptr == md->end_subject || !WAS_NEWLINE(eptr)))
2044        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2045      ecode++;      ecode++;
2046      break;      break;
2047    
2048      /* Start of match assertion */      /* Start of match assertion */
2049    
2050      case OP_SOM:      case OP_SOM:
2051      if (eptr != md->start_subject + md->start_offset) MRRETURN(MATCH_NOMATCH);      if (eptr != md->start_subject + md->start_offset) RRETURN(MATCH_NOMATCH);
2052      ecode++;      ecode++;
2053      break;      break;
2054    
# Line 1849  for (;;) Line 2064  for (;;)
2064    
2065      case OP_DOLLM:      case OP_DOLLM:
2066      if (eptr < md->end_subject)      if (eptr < md->end_subject)
2067        { if (!IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH); }        { if (!IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); }
2068      else      else
2069        {        {
2070        if (md->noteol) MRRETURN(MATCH_NOMATCH);        if (md->noteol) RRETURN(MATCH_NOMATCH);
2071        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2072        }        }
2073      ecode++;      ecode++;
# Line 1862  for (;;) Line 2077  for (;;)
2077      subject unless noteol is set. */      subject unless noteol is set. */
2078    
2079      case OP_DOLL:      case OP_DOLL:
2080      if (md->noteol) MRRETURN(MATCH_NOMATCH);      if (md->noteol) RRETURN(MATCH_NOMATCH);
2081      if (!md->endonly) goto ASSERT_NL_OR_EOS;      if (!md->endonly) goto ASSERT_NL_OR_EOS;
2082    
2083      /* ... else fall through for endonly */      /* ... else fall through for endonly */
# Line 1870  for (;;) Line 2085  for (;;)
2085      /* End of subject assertion (\z) */      /* End of subject assertion (\z) */
2086    
2087      case OP_EOD:      case OP_EOD:
2088      if (eptr < md->end_subject) MRRETURN(MATCH_NOMATCH);      if (eptr < md->end_subject) RRETURN(MATCH_NOMATCH);
2089      SCHECK_PARTIAL();      SCHECK_PARTIAL();
2090      ecode++;      ecode++;
2091      break;      break;
# Line 1881  for (;;) Line 2096  for (;;)
2096      ASSERT_NL_OR_EOS:      ASSERT_NL_OR_EOS:
2097      if (eptr < md->end_subject &&      if (eptr < md->end_subject &&
2098          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))
2099        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2100    
2101      /* Either at end of string or \n before end. */      /* Either at end of string or \n before end. */
2102    
# Line 1900  for (;;) Line 2115  for (;;)
2115        be "non-word" characters. Remember the earliest consulted character for        be "non-word" characters. Remember the earliest consulted character for
2116        partial matching. */        partial matching. */
2117    
2118  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2119        if (utf8)        if (utf)
2120          {          {
2121          /* Get status of previous character */          /* Get status of previous character */
2122    
2123          if (eptr == md->start_subject) prev_is_word = FALSE; else          if (eptr == md->start_subject) prev_is_word = FALSE; else
2124            {            {
2125            USPTR lastptr = eptr - 1;            PCRE_PUCHAR lastptr = eptr - 1;
2126            while((*lastptr & 0xc0) == 0x80) lastptr--;            BACKCHAR(lastptr);
2127            if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;            if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;
2128            GETCHAR(c, lastptr);            GETCHAR(c, lastptr);
2129  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 1973  for (;;) Line 2188  for (;;)
2188              }              }
2189            else            else
2190  #endif  #endif
2191            prev_is_word = ((md->ctypes[eptr[-1]] & ctype_word) != 0);            prev_is_word = MAX_255(eptr[-1])
2192                && ((md->ctypes[eptr[-1]] & ctype_word) != 0);
2193            }            }
2194    
2195          /* Get status of next character */          /* Get status of next character */
# Line 1996  for (;;) Line 2212  for (;;)
2212            }            }
2213          else          else
2214  #endif  #endif
2215          cur_is_word = ((md->ctypes[*eptr] & ctype_word) != 0);          cur_is_word = MAX_255(*eptr)
2216              && ((md->ctypes[*eptr] & ctype_word) != 0);
2217          }          }
2218    
2219        /* Now see if the situation is what we want */        /* Now see if the situation is what we want */
2220    
2221        if ((*ecode++ == OP_WORD_BOUNDARY)?        if ((*ecode++ == OP_WORD_BOUNDARY)?
2222             cur_is_word == prev_is_word : cur_is_word != prev_is_word)             cur_is_word == prev_is_word : cur_is_word != prev_is_word)
2223          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2224        }        }
2225      break;      break;
2226    
2227      /* Match a single character type; inline for speed */      /* Match a single character type; inline for speed */
2228    
2229      case OP_ANY:      case OP_ANY:
2230      if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
2231      /* Fall through */      /* Fall through */
2232    
2233      case OP_ALLANY:      case OP_ALLANY:
2234      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */
2235        {                            /* not be updated before SCHECK_PARTIAL. */        {                            /* not be updated before SCHECK_PARTIAL. */
2236        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2237        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2238        }        }
2239      eptr++;      eptr++;
2240      if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  #ifdef SUPPORT_UTF
2241        if (utf) ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
2242    #endif
2243      ecode++;      ecode++;
2244      break;      break;
2245    
# Line 2031  for (;;) Line 2250  for (;;)
2250      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */
2251        {                            /* not be updated before SCHECK_PARTIAL. */        {                            /* not be updated before SCHECK_PARTIAL. */
2252        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2253        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2254        }        }
2255      eptr++;      eptr++;
2256      ecode++;      ecode++;
# Line 2041  for (;;) Line 2260  for (;;)
2260      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2261        {        {
2262        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2263        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2264        }        }
2265      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2266      if (      if (
2267  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2268         c < 256 &&         c < 256 &&
2269  #endif  #endif
2270         (md->ctypes[c] & ctype_digit) != 0         (md->ctypes[c] & ctype_digit) != 0
2271         )         )
2272        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2273      ecode++;      ecode++;
2274      break;      break;
2275    
# Line 2058  for (;;) Line 2277  for (;;)
2277      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2278        {        {
2279        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2280        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2281        }        }
2282      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2283      if (      if (
2284  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2285         c >= 256 ||         c > 255 ||
2286  #endif  #endif
2287         (md->ctypes[c] & ctype_digit) == 0         (md->ctypes[c] & ctype_digit) == 0
2288         )         )
2289        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2290      ecode++;      ecode++;
2291      break;      break;
2292    
# Line 2075  for (;;) Line 2294  for (;;)
2294      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2295        {        {
2296        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2297        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2298        }        }
2299      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2300      if (      if (
2301  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2302         c < 256 &&         c < 256 &&
2303  #endif  #endif
2304         (md->ctypes[c] & ctype_space) != 0         (md->ctypes[c] & ctype_space) != 0
2305         )         )
2306        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2307      ecode++;      ecode++;
2308      break;      break;
2309    
# Line 2092  for (;;) Line 2311  for (;;)
2311      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2312        {        {
2313        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2314        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2315        }        }
2316      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2317      if (      if (
2318  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2319         c >= 256 ||         c > 255 ||
2320  #endif  #endif
2321         (md->ctypes[c] & ctype_space) == 0         (md->ctypes[c] & ctype_space) == 0
2322         )         )
2323        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2324      ecode++;      ecode++;
2325      break;      break;
2326    
# Line 2109  for (;;) Line 2328  for (;;)
2328      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2329        {        {
2330        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2331        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2332        }        }
2333      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2334      if (      if (
2335  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2336         c < 256 &&         c < 256 &&
2337  #endif  #endif
2338         (md->ctypes[c] & ctype_word) != 0         (md->ctypes[c] & ctype_word) != 0
2339         )         )
2340        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2341      ecode++;      ecode++;
2342      break;      break;
2343    
# Line 2126  for (;;) Line 2345  for (;;)
2345      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2346        {        {
2347        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2348        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2349        }        }
2350      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2351      if (      if (
2352  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2353         c >= 256 ||         c > 255 ||
2354  #endif  #endif
2355         (md->ctypes[c] & ctype_word) == 0         (md->ctypes[c] & ctype_word) == 0
2356         )         )
2357        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2358      ecode++;      ecode++;
2359      break;      break;
2360    
# Line 2143  for (;;) Line 2362  for (;;)
2362      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2363        {        {
2364        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2365        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2366        }        }
2367      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2368      switch(c)      switch(c)
2369        {        {
2370        default: MRRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
2371    
2372        case 0x000d:        case 0x000d:
2373        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
# Line 2162  for (;;) Line 2381  for (;;)
2381        case 0x0085:        case 0x0085:
2382        case 0x2028:        case 0x2028:
2383        case 0x2029:        case 0x2029:
2384        if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);        if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
2385        break;        break;
2386        }        }
2387      ecode++;      ecode++;
# Line 2172  for (;;) Line 2391  for (;;)
2391      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2392        {        {
2393        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2394        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2395        }        }
2396      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2397      switch(c)      switch(c)
# Line 2197  for (;;) Line 2416  for (;;)
2416        case 0x202f:    /* NARROW NO-BREAK SPACE */        case 0x202f:    /* NARROW NO-BREAK SPACE */
2417        case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */        case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
2418        case 0x3000:    /* IDEOGRAPHIC SPACE */        case 0x3000:    /* IDEOGRAPHIC SPACE */
2419        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2420        }        }
2421      ecode++;      ecode++;
2422      break;      break;
# Line 2206  for (;;) Line 2425  for (;;)
2425      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2426        {        {
2427        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2428        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2429        }        }
2430      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2431      switch(c)      switch(c)
2432        {        {
2433        default: MRRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
2434        case 0x09:      /* HT */        case 0x09:      /* HT */
2435        case 0x20:      /* SPACE */        case 0x20:      /* SPACE */
2436        case 0xa0:      /* NBSP */        case 0xa0:      /* NBSP */
# Line 2240  for (;;) Line 2459  for (;;)
2459      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2460        {        {
2461        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2462        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2463        }        }
2464      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2465      switch(c)      switch(c)
# Line 2253  for (;;) Line 2472  for (;;)
2472        case 0x85:      /* NEL */        case 0x85:      /* NEL */
2473        case 0x2028:    /* LINE SEPARATOR */        case 0x2028:    /* LINE SEPARATOR */
2474        case 0x2029:    /* PARAGRAPH SEPARATOR */        case 0x2029:    /* PARAGRAPH SEPARATOR */
2475        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2476        }        }
2477      ecode++;      ecode++;
2478      break;      break;
# Line 2262  for (;;) Line 2481  for (;;)
2481      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2482        {        {
2483        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2484        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2485        }        }
2486      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2487      switch(c)      switch(c)
2488        {        {
2489        default: MRRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
2490        case 0x0a:      /* LF */        case 0x0a:      /* LF */
2491        case 0x0b:      /* VT */        case 0x0b:      /* VT */
2492        case 0x0c:      /* FF */        case 0x0c:      /* FF */
# Line 2289  for (;;) Line 2508  for (;;)
2508      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2509        {        {
2510        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2511        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2512        }        }
2513      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2514        {        {
# Line 2298  for (;;) Line 2517  for (;;)
2517        switch(ecode[1])        switch(ecode[1])
2518          {          {
2519          case PT_ANY:          case PT_ANY:
2520          if (op == OP_NOTPROP) MRRETURN(MATCH_NOMATCH);          if (op == OP_NOTPROP) RRETURN(MATCH_NOMATCH);
2521          break;          break;
2522    
2523          case PT_LAMP:          case PT_LAMP:
2524          if ((prop->chartype == ucp_Lu ||          if ((prop->chartype == ucp_Lu ||
2525               prop->chartype == ucp_Ll ||               prop->chartype == ucp_Ll ||
2526               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))
2527            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2528          break;          break;
2529    
2530          case PT_GC:          case PT_GC:
2531          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))          if ((ecode[2] != PRIV(ucp_gentype)[prop->chartype]) == (op == OP_PROP))
2532            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2533          break;          break;
2534    
2535          case PT_PC:          case PT_PC:
2536          if ((ecode[2] != prop->chartype) == (op == OP_PROP))          if ((ecode[2] != prop->chartype) == (op == OP_PROP))
2537            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2538          break;          break;
2539    
2540          case PT_SC:          case PT_SC:
2541          if ((ecode[2] != prop->script) == (op == OP_PROP))          if ((ecode[2] != prop->script) == (op == OP_PROP))
2542            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2543          break;          break;
2544    
2545          /* These are specials */          /* These are specials */
2546    
2547          case PT_ALNUM:          case PT_ALNUM:
2548          if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
2549               _pcre_ucp_gentype[prop->chartype] == ucp_N) == (op == OP_NOTPROP))               PRIV(ucp_gentype)[prop->chartype] == ucp_N) == (op == OP_NOTPROP))
2550            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2551          break;          break;
2552    
2553          case PT_SPACE:    /* Perl space */          case PT_SPACE:    /* Perl space */
2554          if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
2555               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
2556                 == (op == OP_NOTPROP))                 == (op == OP_NOTPROP))
2557            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2558          break;          break;
2559    
2560          case PT_PXSPACE:  /* POSIX space */          case PT_PXSPACE:  /* POSIX space */
2561          if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
2562               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
2563               c == CHAR_FF || c == CHAR_CR)               c == CHAR_FF || c == CHAR_CR)
2564                 == (op == OP_NOTPROP))                 == (op == OP_NOTPROP))
2565            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2566          break;          break;
2567    
2568          case PT_WORD:          case PT_WORD:
2569          if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
2570               _pcre_ucp_gentype[prop->chartype] == ucp_N ||               PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
2571               c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))               c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))
2572            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2573          break;          break;
2574    
2575          /* This should never occur */          /* This should never occur */
# Line 2370  for (;;) Line 2589  for (;;)
2589      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2590        {        {
2591        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2592        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2593        }        }
2594      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2595      if (UCD_CATEGORY(c) == ucp_M) MRRETURN(MATCH_NOMATCH);      if (UCD_CATEGORY(c) == ucp_M) RRETURN(MATCH_NOMATCH);
2596      while (eptr < md->end_subject)      while (eptr < md->end_subject)
2597        {        {
2598        int len = 1;        int len = 1;
2599        if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }        if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
2600        if (UCD_CATEGORY(c) != ucp_M) break;        if (UCD_CATEGORY(c) != ucp_M) break;
2601        eptr += len;        eptr += len;
2602        }        }
# Line 2398  for (;;) Line 2617  for (;;)
2617      case OP_REFI:      case OP_REFI:
2618      caseless = op == OP_REFI;      caseless = op == OP_REFI;
2619      offset = GET2(ecode, 1) << 1;               /* Doubled ref number */      offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
2620      ecode += 3;      ecode += 1 + IMM2_SIZE;
2621    
2622      /* If the reference is unset, there are two possibilities:      /* If the reference is unset, there are two possibilities:
2623    
# Line 2438  for (;;) Line 2657  for (;;)
2657        case OP_CRMINRANGE:        case OP_CRMINRANGE:
2658        minimize = (*ecode == OP_CRMINRANGE);        minimize = (*ecode == OP_CRMINRANGE);
2659        min = GET2(ecode, 1);        min = GET2(ecode, 1);
2660        max = GET2(ecode, 3);        max = GET2(ecode, 1 + IMM2_SIZE);
2661        if (max == 0) max = INT_MAX;        if (max == 0) max = INT_MAX;
2662        ecode += 5;        ecode += 1 + 2 * IMM2_SIZE;
2663        break;        break;
2664    
2665        default:               /* No repeat follows */        default:               /* No repeat follows */
2666        if ((length = match_ref(offset, eptr, length, md, caseless)) < 0)        if ((length = match_ref(offset, eptr, length, md, caseless)) < 0)
2667          {          {
2668          CHECK_PARTIAL();          CHECK_PARTIAL();
2669          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2670          }          }
2671        eptr += length;        eptr += length;
2672        continue;              /* With the main loop */        continue;              /* With the main loop */
2673        }        }
2674    
2675      /* Handle repeated back references. If the length of the reference is      /* Handle repeated back references. If the length of the reference is
2676      zero, just continue with the main loop. */      zero, just continue with the main loop. If the length is negative, it
2677        means the reference is unset in non-Java-compatible mode. If the minimum is
2678        zero, we can continue at the same level without recursion. For any other
2679        minimum, carrying on will result in NOMATCH. */
2680    
2681      if (length == 0) continue;      if (length == 0) continue;
2682        if (length < 0 && min == 0) continue;
2683    
2684      /* First, ensure the minimum number of matches are present. We get back      /* First, ensure the minimum number of matches are present. We get back
2685      the length of the reference string explicitly rather than passing the      the length of the reference string explicitly rather than passing the
# Line 2468  for (;;) Line 2691  for (;;)
2691        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2692          {          {
2693          CHECK_PARTIAL();          CHECK_PARTIAL();
2694          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2695          }          }
2696        eptr += slength;        eptr += slength;
2697        }        }
# Line 2487  for (;;) Line 2710  for (;;)
2710          int slength;          int slength;
2711          RMATCH(eptr, ecode, offset_top, md, eptrb, RM14);          RMATCH(eptr, ecode, offset_top, md, eptrb, RM14);
2712          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2713          if (fi >= max) MRRETURN(MATCH_NOMATCH);          if (fi >= max) RRETURN(MATCH_NOMATCH);
2714          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2715            {            {
2716            CHECK_PARTIAL();            CHECK_PARTIAL();
2717            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2718            }            }
2719          eptr += slength;          eptr += slength;
2720          }          }
# Line 2519  for (;;) Line 2742  for (;;)
2742          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2743          eptr -= length;          eptr -= length;
2744          }          }
2745        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2746        }        }
2747      /* Control never gets here */      /* Control never gets here */
2748    
# Line 2537  for (;;) Line 2760  for (;;)
2760      case OP_NCLASS:      case OP_NCLASS:
2761      case OP_CLASS:      case OP_CLASS:
2762        {        {
2763          /* The data variable is saved across frames, so the byte map needs to
2764          be stored there. */
2765    #define BYTE_MAP ((pcre_uint8 *)data)
2766        data = ecode + 1;                /* Save for matching */        data = ecode + 1;                /* Save for matching */
2767        ecode += 33;                     /* Advance past the item */        ecode += 1 + (32 / sizeof(pcre_uchar)); /* Advance past the item */
2768    
2769        switch (*ecode)        switch (*ecode)
2770          {          {
# Line 2559  for (;;) Line 2785  for (;;)
2785          case OP_CRMINRANGE:          case OP_CRMINRANGE:
2786          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
2787          min = GET2(ecode, 1);          min = GET2(ecode, 1);
2788          max = GET2(ecode, 3);          max = GET2(ecode, 1 + IMM2_SIZE);
2789          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
2790          ecode += 5;          ecode += 1 + 2 * IMM2_SIZE;
2791          break;          break;
2792    
2793          default:               /* No repeat follows */          default:               /* No repeat follows */
# Line 2571  for (;;) Line 2797  for (;;)
2797    
2798        /* First, ensure the minimum number of matches are present. */        /* First, ensure the minimum number of matches are present. */
2799    
2800  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2801        /* UTF-8 mode */        if (utf)
       if (utf8)  
2802          {          {
2803          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2804            {            {
2805            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
2806              {              {
2807              SCHECK_PARTIAL();              SCHECK_PARTIAL();
2808              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2809              }              }
2810            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
2811            if (c > 255)            if (c > 255)
2812              {              {
2813              if (op == OP_CLASS) MRRETURN(MATCH_NOMATCH);              if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2814              }              }
2815            else            else
2816              {              if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
             if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  
             }  
2817            }            }
2818          }          }
2819        else        else
2820  #endif  #endif
2821        /* Not UTF-8 mode */        /* Not UTF mode */
2822          {          {
2823          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2824            {            {
2825            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
2826              {              {
2827              SCHECK_PARTIAL();              SCHECK_PARTIAL();
2828              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2829              }              }
2830            c = *eptr++;            c = *eptr++;
2831            if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  #ifndef COMPILE_PCRE8
2832              if (c > 255)
2833                {
2834                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2835                }
2836              else
2837    #endif
2838                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2839            }            }
2840          }          }
2841    
# Line 2619  for (;;) Line 2849  for (;;)
2849    
2850        if (minimize)        if (minimize)
2851          {          {
2852  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2853          /* UTF-8 mode */          if (utf)
         if (utf8)  
2854            {            {
2855            for (fi = min;; fi++)            for (fi = min;; fi++)
2856              {              {
2857              RMATCH(eptr, ecode, offset_top, md, eptrb, RM16);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM16);
2858              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2859              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2860              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
2861                {                {
2862                SCHECK_PARTIAL();                SCHECK_PARTIAL();
2863                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2864                }                }
2865              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
2866              if (c > 255)              if (c > 255)
2867                {                {
2868                if (op == OP_CLASS) MRRETURN(MATCH_NOMATCH);                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2869                }                }
2870              else              else
2871                {                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
               if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  
               }  
2872              }              }
2873            }            }
2874          else          else
2875  #endif  #endif
2876          /* Not UTF-8 mode */          /* Not UTF mode */
2877            {            {
2878            for (fi = min;; fi++)            for (fi = min;; fi++)
2879              {              {
2880              RMATCH(eptr, ecode, offset_top, md, eptrb, RM17);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM17);
2881              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2882              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2883              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
2884                {                {
2885                SCHECK_PARTIAL();                SCHECK_PARTIAL();
2886                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2887                }                }
2888              c = *eptr++;              c = *eptr++;
2889              if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  #ifndef COMPILE_PCRE8
2890                if (c > 255)
2891                  {
2892                  if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2893                  }
2894                else
2895    #endif
2896                  if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2897              }              }
2898            }            }
2899          /* Control never gets here */          /* Control never gets here */
# Line 2671  for (;;) Line 2905  for (;;)
2905          {          {
2906          pp = eptr;          pp = eptr;
2907    
2908  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2909          /* UTF-8 mode */          if (utf)
         if (utf8)  
2910            {            {
2911            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2912              {              {
# Line 2689  for (;;) Line 2922  for (;;)
2922                if (op == OP_CLASS) break;                if (op == OP_CLASS) break;
2923                }                }
2924              else              else
2925                {                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
               if ((data[c/8] & (1 << (c&7))) == 0) break;  
               }  
2926              eptr += len;              eptr += len;
2927              }              }
2928            for (;;)            for (;;)
# Line 2704  for (;;) Line 2935  for (;;)
2935            }            }
2936          else          else
2937  #endif  #endif
2938            /* Not UTF-8 mode */            /* Not UTF mode */
2939            {            {
2940            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2941              {              {
# Line 2714  for (;;) Line 2945  for (;;)
2945                break;                break;
2946                }                }
2947              c = *eptr;              c = *eptr;
2948              if ((data[c/8] & (1 << (c&7))) == 0) break;  #ifndef COMPILE_PCRE8
2949                if (c > 255)
2950                  {
2951                  if (op == OP_CLASS) break;
2952                  }
2953                else
2954    #endif
2955                  if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
2956              eptr++;              eptr++;
2957              }              }
2958            while (eptr >= pp)            while (eptr >= pp)
# Line 2725  for (;;) Line 2963  for (;;)
2963              }              }
2964            }            }
2965    
2966          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2967          }          }
2968    #undef BYTE_MAP
2969        }        }
2970      /* Control never gets here */      /* Control never gets here */
2971    
# Line 2735  for (;;) Line 2974  for (;;)
2974      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
2975      mode, because Unicode properties are supported in non-UTF-8 mode. */      mode, because Unicode properties are supported in non-UTF-8 mode. */
2976    
2977  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
2978      case OP_XCLASS:      case OP_XCLASS:
2979        {        {
2980        data = ecode + 1 + LINK_SIZE;                /* Save for matching */        data = ecode + 1 + LINK_SIZE;                /* Save for matching */
# Line 2760  for (;;) Line 2999  for (;;)
2999          case OP_CRMINRANGE:          case OP_CRMINRANGE:
3000          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
3001          min = GET2(ecode, 1);          min = GET2(ecode, 1);
3002          max = GET2(ecode, 3);          max = GET2(ecode, 1 + IMM2_SIZE);
3003          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
3004          ecode += 5;          ecode += 1 + 2 * IMM2_SIZE;
3005          break;          break;
3006    
3007          default:               /* No repeat follows */          default:               /* No repeat follows */
# Line 2777  for (;;) Line 3016  for (;;)
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    
3025        /* 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 2797  for (;;) Line 3036  for (;;)
3036            {            {
3037            RMATCH(eptr, ecode, offset_top, md, eptrb, RM20);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM20);
3038            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3039            if (fi >= max) MRRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3040            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3041              {              {
3042              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3043              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3044              }              }
3045            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3046            if (!_pcre_xclass(c, data)) MRRETURN(MATCH_NOMATCH);            if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH);
3047            }            }
3048          /* Control never gets here */          /* Control never gets here */
3049          }          }
# Line 2822  for (;;) Line 3061  for (;;)
3061              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3062              break;              break;
3063              }              }
3064    #ifdef SUPPORT_UTF
3065            GETCHARLENTEST(c, eptr, len);            GETCHARLENTEST(c, eptr, len);
3066            if (!_pcre_xclass(c, data)) break;  #else
3067              c = *eptr;
3068    #endif
3069              if (!PRIV(xclass)(c, data, utf)) break;
3070            eptr += len;            eptr += len;
3071            }            }
3072          for(;;)          for(;;)
# Line 2831  for (;;) Line 3074  for (;;)
3074            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);
3075            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3076            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
3077            if (utf8) BACKCHAR(eptr);  #ifdef SUPPORT_UTF
3078              if (utf) BACKCHAR(eptr);
3079    #endif
3080            }            }
3081          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3082          }          }
3083    
3084        /* Control never gets here */        /* Control never gets here */
# Line 2843  for (;;) Line 3088  for (;;)
3088      /* Match a single character, casefully */      /* Match a single character, casefully */
3089    
3090      case OP_CHAR:      case OP_CHAR:
3091  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3092      if (utf8)      if (utf)
3093        {        {
3094        length = 1;        length = 1;
3095        ecode++;        ecode++;
# Line 2852  for (;;) Line 3097  for (;;)
3097        if (length > md->end_subject - eptr)        if (length > md->end_subject - eptr)
3098          {          {
3099          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
3100          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3101          }          }
3102        while (length-- > 0) if (*ecode++ != *eptr++) MRRETURN(MATCH_NOMATCH);        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);
3103        }        }
3104      else      else
3105  #endif  #endif
3106        /* Not UTF mode */
     /* Non-UTF-8 mode */  
3107        {        {
3108        if (md->end_subject - eptr < 1)        if (md->end_subject - eptr < 1)
3109          {          {
3110          SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */          SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
3111          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3112          }          }
3113        if (ecode[1] != *eptr++) MRRETURN(MATCH_NOMATCH);        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);
3114        ecode += 2;        ecode += 2;
3115        }        }
3116      break;      break;
3117    
3118      /* Match a single character, caselessly */      /* Match a single character, caselessly. If we are at the end of the
3119        subject, give up immediately. */
3120    
3121      case OP_CHARI:      case OP_CHARI:
3122  #ifdef SUPPORT_UTF8      if (eptr >= md->end_subject)
3123      if (utf8)        {
3124          SCHECK_PARTIAL();
3125          RRETURN(MATCH_NOMATCH);
3126          }
3127    
3128    #ifdef SUPPORT_UTF
3129        if (utf)
3130        {        {
3131        length = 1;        length = 1;
3132        ecode++;        ecode++;
3133        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
3134    
       if (length > md->end_subject - eptr)  
         {  
         CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */  
         MRRETURN(MATCH_NOMATCH);  
         }  
   
3135        /* 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
3136        can use the fast lookup table. */        we know that its other case must also be one byte long, so we can use the
3137          fast lookup table. We know that there is at least one byte left in the
3138          subject. */
3139    
3140        if (fc < 128)        if (fc < 128)
3141          {          {
3142          if (md->lcc[*ecode++] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);          if (md->lcc[fc]
3143                != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH);
3144            ecode++;
3145            eptr++;
3146          }          }
3147    
3148        /* Otherwise we must pick up the subject character */        /* Otherwise we must pick up the subject character. Note that we cannot
3149          use the value of "length" to check for sufficient bytes left, because the
3150          other case of the character may have more or fewer bytes.  */
3151    
3152        else        else
3153          {          {
# Line 2911  for (;;) Line 3163  for (;;)
3163  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3164            if (dc != UCD_OTHERCASE(fc))            if (dc != UCD_OTHERCASE(fc))
3165  #endif  #endif
3166              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3167            }            }
3168          }          }
3169        }        }
3170      else      else
3171  #endif   /* SUPPORT_UTF8 */  #endif   /* SUPPORT_UTF */
3172    
3173      /* Non-UTF-8 mode */      /* Not UTF mode */
3174        {        {
3175        if (md->end_subject - eptr < 1)        if (TABLE_GET(ecode[1], md->lcc, ecode[1])
3176          {            != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH);
3177          SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */        eptr++;
         MRRETURN(MATCH_NOMATCH);  
         }  
       if (md->lcc[ecode[1]] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);  
3178        ecode += 2;        ecode += 2;
3179        }        }
3180      break;      break;
# Line 2935  for (;;) Line 3184  for (;;)
3184      case OP_EXACT:      case OP_EXACT:
3185      case OP_EXACTI:      case OP_EXACTI:
3186      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3187      ecode += 3;      ecode += 1 + IMM2_SIZE;
3188      goto REPEATCHAR;      goto REPEATCHAR;
3189    
3190      case OP_POSUPTO:      case OP_POSUPTO:
# Line 2950  for (;;) Line 3199  for (;;)
3199      min = 0;      min = 0;
3200      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3201      minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI;      minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI;
3202      ecode += 3;      ecode += 1 + IMM2_SIZE;
3203      goto REPEATCHAR;      goto REPEATCHAR;
3204    
3205      case OP_POSSTAR:      case OP_POSSTAR:
# Line 2998  for (;;) Line 3247  for (;;)
3247      /* Common code for all repeated single-character matches. */      /* Common code for all repeated single-character matches. */
3248    
3249      REPEATCHAR:      REPEATCHAR:
3250  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3251      if (utf8)      if (utf)
3252        {        {
3253        length = 1;        length = 1;
3254        charptr = ecode;        charptr = ecode;
# Line 3015  for (;;) Line 3264  for (;;)
3264          unsigned int othercase;          unsigned int othercase;
3265          if (op >= OP_STARI &&     /* Caseless */          if (op >= OP_STARI &&     /* Caseless */
3266              (othercase = UCD_OTHERCASE(fc)) != fc)              (othercase = UCD_OTHERCASE(fc)) != fc)
3267            oclength = _pcre_ord2utf8(othercase, occhars);            oclength = PRIV(ord2utf)(othercase, occhars);
3268          else oclength = 0;          else oclength = 0;
3269  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3270    
3271          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3272            {            {
3273            if (eptr <= md->end_subject - length &&            if (eptr <= md->end_subject - length &&
3274              memcmp(eptr, charptr, length) == 0) eptr += length;              memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3275  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3276            else if (oclength > 0 &&            else if (oclength > 0 &&
3277                     eptr <= md->end_subject - oclength &&                     eptr <= md->end_subject - oclength &&
3278                     memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                     memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3279  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3280            else            else
3281              {              {
3282              CHECK_PARTIAL();              CHECK_PARTIAL();
3283              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3284              }              }
3285            }            }
3286    
# Line 3043  for (;;) Line 3292  for (;;)
3292              {              {
3293              RMATCH(eptr, ecode, offset_top, md, eptrb, RM22);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM22);
3294              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3295              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3296              if (eptr <= md->end_subject - length &&              if (eptr <= md->end_subject - length &&
3297                memcmp(eptr, charptr, length) == 0) eptr += length;                memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3298  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3299              else if (oclength > 0 &&              else if (oclength > 0 &&
3300                       eptr <= md->end_subject - oclength &&                       eptr <= md->end_subject - oclength &&
3301                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                       memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3302  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3303              else              else
3304                {                {
3305                CHECK_PARTIAL();                CHECK_PARTIAL();
3306                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3307                }                }
3308              }              }
3309            /* Control never gets here */            /* Control never gets here */
# Line 3066  for (;;) Line 3315  for (;;)
3315            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3316              {              {
3317              if (eptr <= md->end_subject - length &&              if (eptr <= md->end_subject - length &&
3318                  memcmp(eptr, charptr, length) == 0) eptr += length;                  memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3319  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3320              else if (oclength > 0 &&              else if (oclength > 0 &&
3321                       eptr <= md->end_subject - oclength &&                       eptr <= md->end_subject - oclength &&
3322                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                       memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3323  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3324              else              else
3325                {                {
# Line 3085  for (;;) Line 3334  for (;;)
3334              {              {
3335              RMATCH(eptr, ecode, offset_top, md, eptrb, RM23);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM23);
3336              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3337              if (eptr == pp) { MRRETURN(MATCH_NOMATCH); }              if (eptr == pp) { RRETURN(MATCH_NOMATCH); }
3338  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3339              eptr--;              eptr--;
3340              BACKCHAR(eptr);              BACKCHAR(eptr);
# Line 3102  for (;;) Line 3351  for (;;)
3351        value of fc will always be < 128. */        value of fc will always be < 128. */
3352        }        }
3353      else      else
3354  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
3355          /* When not in UTF-8 mode, load a single-byte character. */
3356      /* When not in UTF-8 mode, load a single-byte character. */        fc = *ecode++;
3357    
3358      fc = *ecode++;      /* The value of fc at this point is always one character, though we may
3359        or may not be in UTF mode. The code is duplicated for the caseless and
     /* The value of fc at this point is always less than 256, though we may or  
     may not be in UTF-8 mode. The code is duplicated for the caseless and  
3360      caseful cases, for speed, since matching characters is likely to be quite      caseful cases, for speed, since matching characters is likely to be quite
3361      common. First, ensure the minimum number of matches are present. If min =      common. First, ensure the minimum number of matches are present. If min =
3362      max, continue at the same level without recursing. Otherwise, if      max, continue at the same level without recursing. Otherwise, if
# Line 3122  for (;;) Line 3369  for (;;)
3369    
3370      if (op >= OP_STARI)  /* Caseless */      if (op >= OP_STARI)  /* Caseless */
3371        {        {
3372        fc = md->lcc[fc];  #ifdef COMPILE_PCRE8
3373          /* fc must be < 128 if UTF is enabled. */
3374          foc = md->fcc[fc];
3375    #else
3376    #ifdef SUPPORT_UTF
3377    #ifdef SUPPORT_UCP
3378          if (utf && fc > 127)
3379            foc = UCD_OTHERCASE(fc);
3380    #else
3381          if (utf && fc > 127)
3382            foc = fc;
3383    #endif /* SUPPORT_UCP */
3384          else
3385    #endif /* SUPPORT_UTF */
3386            foc = TABLE_GET(fc, md->fcc, fc);
3387    #endif /* COMPILE_PCRE8 */
3388    
3389        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
3390          {          {
3391          if (eptr >= md->end_subject)          if (eptr >= md->end_subject)
3392            {            {
3393            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3394            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3395            }            }
3396          if (fc != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);          if (fc != *eptr && foc != *eptr) RRETURN(MATCH_NOMATCH);
3397            eptr++;
3398          }          }
3399        if (min == max) continue;        if (min == max) continue;
3400        if (minimize)        if (minimize)
# Line 3139  for (;;) Line 3403  for (;;)
3403            {            {
3404            RMATCH(eptr, ecode, offset_top, md, eptrb, RM24);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM24);
3405            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3406            if (fi >= max) MRRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3407            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3408              {              {
3409              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3410              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3411              }              }
3412            if (fc != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);            if (fc != *eptr && foc != *eptr) RRETURN(MATCH_NOMATCH);
3413              eptr++;
3414            }            }
3415          /* Control never gets here */          /* Control never gets here */
3416          }          }
# Line 3159  for (;;) Line 3424  for (;;)
3424              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3425              break;              break;
3426              }              }
3427            if (fc != md->lcc[*eptr]) break;            if (fc != *eptr && foc != *eptr) break;
3428            eptr++;            eptr++;
3429            }            }
3430    
# Line 3171  for (;;) Line 3436  for (;;)
3436            eptr--;            eptr--;
3437            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3438            }            }
3439          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3440          }          }
3441        /* Control never gets here */        /* Control never gets here */
3442        }        }
# Line 3185  for (;;) Line 3450  for (;;)
3450          if (eptr >= md->end_subject)          if (eptr >= md->end_subject)
3451            {            {
3452            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3453            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3454            }            }
3455          if (fc != *eptr++) MRRETURN(MATCH_NOMATCH);          if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
3456          }          }
3457    
3458        if (min == max) continue;        if (min == max) continue;
# Line 3198  for (;;) Line 3463  for (;;)
3463            {            {
3464            RMATCH(eptr, ecode, offset_top, md, eptrb, RM26);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM26);
3465            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3466            if (fi >= max) MRRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3467            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3468              {              {
3469              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3470              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3471              }              }
3472            if (fc != *eptr++) MRRETURN(MATCH_NOMATCH);            if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
3473            }            }
3474          /* Control never gets here */          /* Control never gets here */
3475          }          }
# Line 3229  for (;;) Line 3494  for (;;)
3494            eptr--;            eptr--;
3495            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3496            }            }
3497          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3498          }          }
3499        }        }
3500      /* Control never gets here */      /* Control never gets here */
# Line 3242  for (;;) Line 3507  for (;;)
3507      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
3508        {        {
3509        SCHECK_PARTIAL();        SCHECK_PARTIAL();
3510        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
3511        }        }
3512      ecode++;      ecode++;
3513      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
3514      if (op == OP_NOTI)         /* The caseless case */      if (op == OP_NOTI)         /* The caseless case */
3515        {        {
3516  #ifdef SUPPORT_UTF8        register int ch, och;
3517        if (c < 256)        ch = *ecode++;
3518  #endif  #ifdef COMPILE_PCRE8
3519        c = md->lcc[c];        /* ch must be < 128 if UTF is enabled. */
3520        if (md->lcc[*ecode++] == c) MRRETURN(MATCH_NOMATCH);        och = md->fcc[ch];
3521    #else
3522    #ifdef SUPPORT_UTF
3523    #ifdef SUPPORT_UCP
3524          if (utf && ch > 127)
3525            och = UCD_OTHERCASE(ch);
3526    #else
3527          if (utf && ch > 127)
3528            och = ch;
3529    #endif /* SUPPORT_UCP */
3530          else
3531    #endif /* SUPPORT_UTF */
3532            och = TABLE_GET(ch, md->fcc, ch);
3533    #endif /* COMPILE_PCRE8 */
3534          if (ch == c || och == c) RRETURN(MATCH_NOMATCH);
3535        }        }
3536      else    /* Caseful */      else    /* Caseful */
3537        {        {
3538        if (*ecode++ == c) MRRETURN(MATCH_NOMATCH);        if (*ecode++ == c) RRETURN(MATCH_NOMATCH);
3539        }        }
3540      break;      break;
3541    
# Line 3270  for (;;) Line 3549  for (;;)
3549      case OP_NOTEXACT:      case OP_NOTEXACT:
3550      case OP_NOTEXACTI:      case OP_NOTEXACTI:
3551      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3552      ecode += 3;      ecode += 1 + IMM2_SIZE;
3553      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3554    
3555      case OP_NOTUPTO:      case OP_NOTUPTO:
# Line 3280  for (;;) Line 3559  for (;;)
3559      min = 0;      min = 0;
3560      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3561      minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI;      minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI;
3562      ecode += 3;      ecode += 1 + IMM2_SIZE;
3563      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3564    
3565      case OP_NOTPOSSTAR:      case OP_NOTPOSSTAR:
# Line 3312  for (;;) Line 3591  for (;;)
3591      possessive = TRUE;      possessive = TRUE;
3592      min = 0;      min = 0;
3593      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3594      ecode += 3;      ecode += 1 + IMM2_SIZE;
3595      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3596    
3597      case OP_NOTSTAR:      case OP_NOTSTAR:
# Line 3351  for (;;) Line 3630  for (;;)
3630    
3631      if (op >= OP_NOTSTARI)     /* Caseless */      if (op >= OP_NOTSTARI)     /* Caseless */
3632        {        {
3633        fc = md->lcc[fc];  #ifdef COMPILE_PCRE8
3634          /* fc must be < 128 if UTF is enabled. */
3635          foc = md->fcc[fc];
3636    #else
3637    #ifdef SUPPORT_UTF
3638    #ifdef SUPPORT_UCP
3639          if (utf && fc > 127)
3640            foc = UCD_OTHERCASE(fc);
3641    #else
3642          if (utf && fc > 127)
3643            foc = fc;
3644    #endif /* SUPPORT_UCP */
3645          else
3646    #endif /* SUPPORT_UTF */
3647            foc = TABLE_GET(fc, md->fcc, fc);
3648    #endif /* COMPILE_PCRE8 */
3649    
3650  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3651        /* UTF-8 mode */        if (utf)
       if (utf8)  
3652          {          {
3653          register unsigned int d;          register unsigned int d;
3654          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 3363  for (;;) Line 3656  for (;;)
3656            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3657              {              {
3658              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3659              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3660              }              }
3661            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3662            if (d < 256) d = md->lcc[d];            if (fc == d || foc == d) RRETURN(MATCH_NOMATCH);
           if (fc == d) MRRETURN(MATCH_NOMATCH);  
3663            }            }
3664          }          }
3665        else        else
3666  #endif  #endif
3667          /* Not UTF mode */
       /* Not UTF-8 mode */  
3668          {          {
3669          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3670            {            {
3671            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3672              {              {
3673              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3674              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3675              }              }
3676            if (fc == md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);            if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
3677              eptr++;
3678            }            }
3679          }          }
3680    
# Line 3390  for (;;) Line 3682  for (;;)
3682    
3683        if (minimize)        if (minimize)
3684          {          {
3685  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3686          /* UTF-8 mode */          if (utf)
         if (utf8)  
3687            {            {
3688            register unsigned int d;            register unsigned int d;
3689            for (fi = min;; fi++)            for (fi = min;; fi++)
3690              {              {
3691              RMATCH(eptr, ecode, offset_top, md, eptrb, RM28);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM28);
3692              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3693              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3694              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3695                {                {
3696                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3697                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3698                }                }
3699              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3700              if (d < 256) d = md->lcc[d];              if (fc == d || foc == d) RRETURN(MATCH_NOMATCH);
             if (fc == d) MRRETURN(MATCH_NOMATCH);  
3701              }              }
3702            }            }
3703          else          else
3704  #endif  #endif
3705          /* Not UTF-8 mode */          /* Not UTF mode */
3706            {            {
3707            for (fi = min;; fi++)            for (fi = min;; fi++)
3708              {              {
3709              RMATCH(eptr, ecode, offset_top, md, eptrb, RM29);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM29);
3710              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3711              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3712              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3713                {                {
3714                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3715                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3716                }                }
3717              if (fc == md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);              if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
3718                eptr++;
3719              }              }
3720            }            }
3721          /* Control never gets here */          /* Control never gets here */
# Line 3436  for (;;) Line 3727  for (;;)
3727          {          {
3728          pp = eptr;          pp = eptr;
3729    
3730  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3731          /* UTF-8 mode */          if (utf)
         if (utf8)  
3732            {            {
3733            register unsigned int d;            register unsigned int d;
3734            for (i = min; i < max; i++)            for (i = min; i < max; i++)
# Line 3450  for (;;) Line 3740  for (;;)
3740                break;                break;
3741                }                }
3742              GETCHARLEN(d, eptr, len);              GETCHARLEN(d, eptr, len);
3743              if (d < 256) d = md->lcc[d];              if (fc == d || foc == d) break;
             if (fc == d) break;  
3744              eptr += len;              eptr += len;
3745              }              }
3746          if (possessive) continue;            if (possessive) continue;
3747          for(;;)            for(;;)
3748              {              {
3749              RMATCH(eptr, ecode, offset_top, md, eptrb, RM30);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM30);
3750              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
# Line 3465  for (;;) Line 3754  for (;;)
3754            }            }
3755          else          else
3756  #endif  #endif
3757          /* Not UTF-8 mode */          /* Not UTF mode */
3758            {            {
3759            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3760              {              {
# Line 3474  for (;;) Line 3763  for (;;)
3763                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3764                break;                break;
3765                }                }
3766              if (fc == md->lcc[*eptr]) break;              if (fc == *eptr || foc == *eptr) break;
3767              eptr++;              eptr++;
3768              }              }
3769            if (possessive) continue;            if (possessive) continue;
# Line 3486  for (;;) Line 3775  for (;;)
3775              }              }
3776            }            }
3777    
3778          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3779          }          }
3780        /* Control never gets here */        /* Control never gets here */
3781        }        }
# Line 3495  for (;;) Line 3784  for (;;)
3784    
3785      else      else
3786        {        {
3787  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3788        /* UTF-8 mode */        if (utf)
       if (utf8)  
3789          {          {
3790          register unsigned int d;          register unsigned int d;
3791          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 3505  for (;;) Line 3793  for (;;)
3793            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3794              {              {
3795              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3796              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3797              }              }
3798            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3799            if (fc == d) MRRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
3800            }            }
3801          }          }
3802        else        else
3803  #endif  #endif
3804        /* Not UTF-8 mode */        /* Not UTF mode */
3805          {          {
3806          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3807            {            {
3808            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3809              {              {
3810              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3811              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3812              }              }
3813            if (fc == *eptr++) MRRETURN(MATCH_NOMATCH);            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
3814            }            }
3815          }          }
3816    
# Line 3530  for (;;) Line 3818  for (;;)
3818    
3819        if (minimize)        if (minimize)
3820          {          {
3821  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3822          /* UTF-8 mode */          if (utf)
         if (utf8)  
3823            {            {
3824            register unsigned int d;            register unsigned int d;
3825            for (fi = min;; fi++)            for (fi = min;; fi++)
3826              {              {
3827              RMATCH(eptr, ecode, offset_top, md, eptrb, RM32);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM32);
3828              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3829              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3830              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3831                {                {
3832                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3833                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3834                }                }
3835              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3836              if (fc == d) MRRETURN(MATCH_NOMATCH);              if (fc == d) RRETURN(MATCH_NOMATCH);
3837              }              }
3838            }            }
3839          else          else
3840  #endif  #endif
3841          /* Not UTF-8 mode */          /* Not UTF mode */
3842            {            {
3843            for (fi = min;; fi++)            for (fi = min;; fi++)
3844              {              {
3845              RMATCH(eptr, ecode, offset_top, md, eptrb, RM33);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM33);
3846              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3847              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3848              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3849                {                {
3850                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3851                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3852                }                }
3853              if (fc == *eptr++) MRRETURN(MATCH_NOMATCH);              if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
3854              }              }
3855            }            }
3856          /* Control never gets here */          /* Control never gets here */
# Line 3575  for (;;) Line 3862  for (;;)
3862          {          {
3863          pp = eptr;          pp = eptr;
3864    
3865  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3866          /* UTF-8 mode */          if (utf)
         if (utf8)  
3867            {            {
3868            register unsigned int d;            register unsigned int d;
3869            for (i = min; i < max; i++)            for (i = min; i < max; i++)
# Line 3603  for (;;) Line 3889  for (;;)
3889            }            }
3890          else          else
3891  #endif  #endif
3892          /* Not UTF-8 mode */          /* Not UTF mode */
3893            {            {
3894            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3895              {              {
# Line 3624  for (;;) Line 3910  for (;;)
3910              }              }
3911            }            }
3912    
3913          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3914          }          }
3915        }        }
3916      /* Control never gets here */      /* Control never gets here */
# Line 3636  for (;;) Line 3922  for (;;)
3922      case OP_TYPEEXACT:      case OP_TYPEEXACT:
3923      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3924      minimize = TRUE;      minimize = TRUE;
3925      ecode += 3;      ecode += 1 + IMM2_SIZE;
3926      goto REPEATTYPE;      goto REPEATTYPE;
3927    
3928      case OP_TYPEUPTO:      case OP_TYPEUPTO:
# Line 3644  for (;;) Line 3930  for (;;)
3930      min = 0;      min = 0;
3931      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3932      minimize = *ecode == OP_TYPEMINUPTO;      minimize = *ecode == OP_TYPEMINUPTO;
3933      ecode += 3;      ecode += 1 + IMM2_SIZE;
3934      goto REPEATTYPE;      goto REPEATTYPE;
3935    
3936      case OP_TYPEPOSSTAR:      case OP_TYPEPOSSTAR:
# Line 3672  for (;;) Line 3958  for (;;)
3958      possessive = TRUE;      possessive = TRUE;
3959      min = 0;      min = 0;
3960      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3961      ecode += 3;      ecode += 1 + IMM2_SIZE;
3962      goto REPEATTYPE;      goto REPEATTYPE;
3963    
3964      case OP_TYPESTAR:      case OP_TYPESTAR:
# Line 3718  for (;;) Line 4004  for (;;)
4004          switch(prop_type)          switch(prop_type)
4005            {            {
4006            case PT_ANY:            case PT_ANY:
4007            if (prop_fail_result) MRRETURN(MATCH_NOMATCH);            if (prop_fail_result) RRETURN(MATCH_NOMATCH);
4008            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
4009              {              {
4010              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4011                {                {
4012                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4013                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4014                }                }
4015              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4016              }              }
# Line 3737  for (;;) Line 4023  for (;;)
4023              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4024                {                {
4025                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4026                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4027                }                }
4028              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4029              chartype = UCD_CHARTYPE(c);              chartype = UCD_CHARTYPE(c);
4030              if ((chartype == ucp_Lu ||              if ((chartype == ucp_Lu ||
4031                   chartype == ucp_Ll ||                   chartype == ucp_Ll ||
4032                   chartype == ucp_Lt) == prop_fail_result)                   chartype == ucp_Lt) == prop_fail_result)
4033                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4034              }              }
4035            break;            break;
4036    
# Line 3754  for (;;) Line 4040  for (;;)
4040              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4041                {                {
4042                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4043                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4044                }                }
4045              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4046              if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)              if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)
4047                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4048              }              }
4049            break;            break;
4050    
# Line 3768  for (;;) Line 4054  for (;;)
4054              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4055                {                {
4056                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4057                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4058                }                }
4059              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4060              if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)              if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)
4061                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4062              }              }
4063            break;            break;
4064    
# Line 3782  for (;;) Line 4068  for (;;)
4068              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4069                {                {
4070                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4071                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4072                }                }
4073              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4074              if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result)              if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result)
4075                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4076              }              }
4077            break;            break;
4078    
# Line 3797  for (;;) Line 4083  for (;;)
4083              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4084                {                {
4085                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4086                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4087                }                }
4088              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4089              category = UCD_CATEGORY(c);              category = UCD_CATEGORY(c);
4090              if ((category == ucp_L || category == ucp_N) == prop_fail_result)              if ((category == ucp_L || category == ucp_N) == prop_fail_result)
4091                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4092              }              }
4093            break;            break;
4094    
# Line 3812  for (;;) Line 4098  for (;;)
4098              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4099                {                {
4100                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4101                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4102                }                }
4103              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4104              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
4105                   c == CHAR_FF || c == CHAR_CR)                   c == CHAR_FF || c == CHAR_CR)
4106                     == prop_fail_result)                     == prop_fail_result)
4107                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4108              }              }
4109            break;            break;
4110    
# Line 3828  for (;;) Line 4114  for (;;)
4114              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4115                {                {
4116                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4117                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4118                }                }
4119              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4120              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
4121                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)
4122                     == prop_fail_result)                     == prop_fail_result)
4123                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4124              }              }
4125            break;            break;
4126    
# Line 3845  for (;;) Line 4131  for (;;)
4131              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4132                {                {
4133                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4134                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4135                }                }
4136              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4137              category = UCD_CATEGORY(c);              category = UCD_CATEGORY(c);
4138              if ((category == ucp_L || category == ucp_N || c == CHAR_UNDERSCORE)              if ((category == ucp_L || category == ucp_N || c == CHAR_UNDERSCORE)
4139                     == prop_fail_result)                     == prop_fail_result)
4140                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4141              }              }
4142            break;            break;
4143    
# Line 3872  for (;;) Line 4158  for (;;)
4158            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4159              {              {
4160              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4161              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4162              }              }
4163            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
4164            if (UCD_CATEGORY(c) == ucp_M) MRRETURN(MATCH_NOMATCH);            if (UCD_CATEGORY(c) == ucp_M) RRETURN(MATCH_NOMATCH);
4165            while (eptr < md->end_subject)            while (eptr < md->end_subject)
4166              {              {
4167              int len = 1;              int len = 1;
4168              if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
4169              if (UCD_CATEGORY(c) != ucp_M) break;              if (UCD_CATEGORY(c) != ucp_M) break;
4170              eptr += len;              eptr += len;
4171              }              }
# Line 3891  for (;;) Line 4177  for (;;)
4177    
4178  /* Handle all other cases when the coding is UTF-8 */  /* Handle all other cases when the coding is UTF-8 */
4179    
4180  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
4181        if (utf8) switch(ctype)        if (utf) switch(ctype)
4182          {          {
4183          case OP_ANY:          case OP_ANY:
4184          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 3900  for (;;) Line 4186  for (;;)
4186            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4187              {              {
4188              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4189              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4190              }              }
4191            if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
4192            eptr++;            eptr++;
4193            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4194            }            }
4195          break;          break;
4196    
# Line 3914  for (;;) Line 4200  for (;;)
4200            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4201              {              {
4202              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4203              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4204              }              }
4205            eptr++;            eptr++;
4206            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4207            }            }
4208          break;          break;
4209    
4210          case OP_ANYBYTE:          case OP_ANYBYTE:
4211          if (eptr > md->end_subject - min) MRRETURN(MATCH_NOMATCH);          if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
4212          eptr += min;          eptr += min;
4213          break;          break;
4214    
# Line 3932  for (;;) Line 4218  for (;;)
4218            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4219              {              {
4220              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4221              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4222              }              }
4223            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4224            switch(c)            switch(c)
4225              {              {
4226              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4227    
4228              case 0x000d:              case 0x000d:
4229              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
# Line 3951  for (;;) Line 4237  for (;;)
4237              case 0x0085:              case 0x0085:
4238              case 0x2028:              case 0x2028:
4239              case 0x2029:              case 0x2029:
4240              if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4241              break;              break;
4242              }              }
4243            }            }
# Line 3963  for (;;) Line 4249  for (;;)
4249            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4250              {              {
4251              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4252              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4253              }              }
4254            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4255            switch(c)            switch(c)
# Line 3988  for (;;) Line 4274  for (;;)
4274              case 0x202f:    /* NARROW NO-BREAK SPACE */              case 0x202f:    /* NARROW NO-BREAK SPACE */
4275              case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */              case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4276              case 0x3000:    /* IDEOGRAPHIC SPACE */              case 0x3000:    /* IDEOGRAPHIC SPACE */
4277              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4278              }              }
4279            }            }
4280          break;          break;
# Line 3999  for (;;) Line 4285  for (;;)
4285            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4286              {              {
4287              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4288              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4289              }              }
4290            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4291            switch(c)            switch(c)
4292              {              {
4293              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4294              case 0x09:      /* HT */              case 0x09:      /* HT */
4295              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4296              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
# Line 4035  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)
# Line 4048  for (;;) Line 4334  for (;;)
4334              case 0x85:      /* NEL */              case 0x85:      /* NEL */
4335              case 0x2028:    /* LINE SEPARATOR */              case 0x2028:    /* LINE SEPARATOR */
4336              case 0x2029:    /* PARAGRAPH SEPARATOR */              case 0x2029:    /* PARAGRAPH SEPARATOR */
4337              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4338              }              }
4339            }            }
4340          break;          break;
# Line 4059  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            switch(c)            switch(c)
4352              {              {
4353              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4354              case 0x0a:      /* LF */              case 0x0a:      /* LF */
4355              case 0x0b:      /* VT */              case 0x0b:      /* VT */
4356              case 0x0c:      /* FF */              case 0x0c:      /* FF */
# Line 4083  for (;;) Line 4369  for (;;)
4369            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4370              {              {
4371              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4372              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4373              }              }
4374            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4375            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)
4376              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4377            }            }
4378          break;          break;
4379    
# Line 4097  for (;;) Line 4383  for (;;)
4383            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4384              {              {
4385              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4386              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4387              }              }
4388            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_digit) == 0)
4389              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4390              eptr++;
4391            /* 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 */
4392            }            }
4393          break;          break;
# Line 4111  for (;;) Line 4398  for (;;)
4398            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4399              {              {
4400              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4401              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4402              }              }
4403            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)
4404              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4405            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);            eptr++;
4406              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4407            }            }
4408          break;          break;
4409    
# Line 4125  for (;;) Line 4413  for (;;)
4413            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4414              {              {
4415              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4416              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4417              }              }
4418            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_space) == 0)
4419              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4420              eptr++;
4421            /* 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 */
4422            }            }
4423          break;          break;
# Line 4139  for (;;) Line 4428  for (;;)
4428            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4429              {              {
4430              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4431              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4432              }              }
4433            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0)            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0)
4434              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4435            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);            eptr++;
4436              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4437            }            }
4438          break;          break;
4439    
# Line 4153  for (;;) Line 4443  for (;;)
4443            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4444              {              {
4445              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4446              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4447              }              }
4448            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_word) == 0)
4449              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4450              eptr++;
4451            /* 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 */
4452            }            }
4453          break;          break;
# Line 4166  for (;;) Line 4457  for (;;)
4457          }  /* End switch(ctype) */          }  /* End switch(ctype) */
4458    
4459        else        else
4460  #endif     /* SUPPORT_UTF8 */  #endif     /* SUPPORT_UTF */
4461    
4462        /* 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
4463        than OP_PROP and OP_NOTPROP. */        than OP_PROP and OP_NOTPROP. */
# Line 4179  for (;;) Line 4470  for (;;)
4470            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4471              {              {
4472              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4473              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4474              }              }
4475            if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
4476            eptr++;            eptr++;
4477            }            }
4478          break;          break;
# Line 4190  for (;;) Line 4481  for (;;)
4481          if (eptr > md->end_subject - min)          if (eptr > md->end_subject - min)
4482            {            {
4483            SCHECK_PARTIAL();            SCHECK_PARTIAL();
4484            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
4485            }            }
4486          eptr += min;          eptr += min;
4487          break;          break;
# Line 4199  for (;;) Line 4490  for (;;)
4490          if (eptr > md->end_subject - min)          if (eptr > md->end_subject - min)
4491            {            {
4492            SCHECK_PARTIAL();            SCHECK_PARTIAL();
4493            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
4494            }            }
4495          eptr += min;          eptr += min;
4496          break;          break;
# Line 4210  for (;;) Line 4501  for (;;)
4501            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4502              {              {
4503              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4504              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4505              }              }
4506            switch(*eptr++)            switch(*eptr++)
4507              {              {
4508              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4509    
4510              case 0x000d:              case 0x000d:
4511              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
# Line 4226  for (;;) Line 4517  for (;;)
4517              case 0x000b:              case 0x000b:
4518              case 0x000c:              case 0x000c:
4519              case 0x0085:              case 0x0085:
4520              if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);  #ifdef COMPILE_PCRE16
4521                case 0x2028:
4522                case 0x2029:
4523    #endif
4524                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4525              break;              break;
4526              }              }
4527            }            }
# Line 4238  for (;;) Line 4533  for (;;)
4533            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4534              {              {
4535              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4536              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4537              }              }
4538            switch(*eptr++)            switch(*eptr++)
4539              {              {
# Line 4246  for (;;) Line 4541  for (;;)
4541              case 0x09:      /* HT */              case 0x09:      /* HT */
4542              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4543              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
4544              MRRETURN(MATCH_NOMATCH);  #ifdef COMPILE_PCRE16
4545                case 0x1680:    /* OGHAM SPACE MARK */
4546                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
4547                case 0x2000:    /* EN QUAD */
4548                case 0x2001:    /* EM QUAD */
4549                case 0x2002:    /* EN SPACE */
4550                case 0x2003:    /* EM SPACE */
4551                case 0x2004:    /* THREE-PER-EM SPACE */
4552                case 0x2005:    /* FOUR-PER-EM SPACE */
4553                case 0x2006:    /* SIX-PER-EM SPACE */
4554                case 0x2007:    /* FIGURE SPACE */
4555                case 0x2008:    /* PUNCTUATION SPACE */
4556                case 0x2009:    /* THIN SPACE */
4557                case 0x200A:    /* HAIR SPACE */
4558                case 0x202f:    /* NARROW NO-BREAK SPACE */
4559                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4560                case 0x3000:    /* IDEOGRAPHIC SPACE */
4561    #endif
4562                RRETURN(MATCH_NOMATCH);
4563              }              }
4564            }            }
4565          break;          break;
# Line 4257  for (;;) Line 4570  for (;;)
4570            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4571              {              {
4572              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4573              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4574              }              }
4575            switch(*eptr++)            switch(*eptr++)
4576              {              {
4577              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4578              case 0x09:      /* HT */              case 0x09:      /* HT */
4579              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4580              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
4581    #ifdef COMPILE_PCRE16
4582                case 0x1680:    /* OGHAM SPACE MARK */
4583                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
4584                case 0x2000:    /* EN QUAD */
4585                case 0x2001:    /* EM QUAD */
4586                case 0x2002:    /* EN SPACE */
4587                case 0x2003:    /* EM SPACE */
4588                case 0x2004:    /* THREE-PER-EM SPACE */
4589                case 0x2005:    /* FOUR-PER-EM SPACE */
4590                case 0x2006:    /* SIX-PER-EM SPACE */
4591                case 0x2007:    /* FIGURE SPACE */
4592                case 0x2008:    /* PUNCTUATION SPACE */
4593                case 0x2009:    /* THIN SPACE */
4594                case 0x200A:    /* HAIR SPACE */
4595                case 0x202f:    /* NARROW NO-BREAK SPACE */
4596                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4597                case 0x3000:    /* IDEOGRAPHIC SPACE */
4598    #endif
4599              break;              break;
4600              }              }
4601            }            }
# Line 4276  for (;;) Line 4607  for (;;)
4607            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4608              {              {
4609              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4610              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4611              }              }
4612            switch(*eptr++)            switch(*eptr++)
4613              {              {
# Line 4286  for (;;) Line 4617  for (;;)
4617              case 0x0c:      /* FF */              case 0x0c:      /* FF */
4618              case 0x0d:      /* CR */              case 0x0d:      /* CR */
4619              case 0x85:      /* NEL */              case 0x85:      /* NEL */
4620              MRRETURN(MATCH_NOMATCH);  #ifdef COMPILE_PCRE16
4621                case 0x2028:    /* LINE SEPARATOR */
4622                case 0x2029:    /* PARAGRAPH SEPARATOR */
4623    #endif
4624                RRETURN(MATCH_NOMATCH);
4625              }              }
4626            }            }
4627          break;          break;
# Line 4297  for (;;) Line 4632  for (;;)
4632            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4633              {              {
4634              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4635              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4636              }              }
4637            switch(*eptr++)            switch(*eptr++)
4638              {              {
4639              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4640              case 0x0a:      /* LF */              case 0x0a:      /* LF */
4641              case 0x0b:      /* VT */              case 0x0b:      /* VT */
4642              case 0x0c:      /* FF */              case 0x0c:      /* FF */
4643              case 0x0d:      /* CR */              case 0x0d:      /* CR */
4644              case 0x85:      /* NEL */              case 0x85:      /* NEL */
4645    #ifdef COMPILE_PCRE16
4646                case 0x2028:    /* LINE SEPARATOR */
4647                case 0x2029:    /* PARAGRAPH SEPARATOR */
4648    #endif
4649              break;              break;
4650              }              }
4651            }            }
# Line 4318  for (;;) Line 4657  for (;;)
4657            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4658              {              {
4659              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4660              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4661              }              }
4662            if ((md->ctypes[*eptr++] & ctype_digit) != 0) MRRETURN(MATCH_NOMATCH);            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_digit) != 0)
4663                RRETURN(MATCH_NOMATCH);
4664              eptr++;
4665            }            }
4666          break;          break;
4667    
# Line 4330  for (;;) Line 4671  for (;;)
4671            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4672              {              {
4673              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4674              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4675              }              }
4676            if ((md->ctypes[*eptr++] & ctype_digit) == 0) MRRETURN(MATCH_NOMATCH);            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_digit) == 0)
4677                RRETURN(MATCH_NOMATCH);
4678              eptr++;
4679            }            }
4680          break;          break;
4681    
# Line 4342  for (;;) Line 4685  for (;;)
4685            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4686              {              {
4687              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4688              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4689              }              }
4690            if ((md->ctypes[*eptr++] & ctype_space) != 0) MRRETURN(MATCH_NOMATCH);            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_space) != 0)
4691                RRETURN(MATCH_NOMATCH);
4692              eptr++;
4693            }            }
4694          break;          break;
4695    
# Line 4354  for (;;) Line 4699  for (;;)
4699            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4700              {              {
4701              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4702              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4703              }              }
4704            if ((md->ctypes[*eptr++] & ctype_space) == 0) MRRETURN(MATCH_NOMATCH);            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_space) == 0)
4705                RRETURN(MATCH_NOMATCH);
4706              eptr++;
4707            }            }
4708          break;          break;
4709    
# Line 4366  for (;;) Line 4713  for (;;)
4713            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4714              {              {
4715              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4716              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4717              }              }
4718            if ((md->ctypes[*eptr++] & ctype_word) != 0)            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_word) != 0)
4719              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4720              eptr++;
4721            }            }
4722          break;          break;
4723    
# Line 4379  for (;;) Line 4727  for (;;)
4727            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4728              {              {
4729              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4730              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4731              }              }
4732            if ((md->ctypes[*eptr++] & ctype_word) == 0)            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_word) == 0)
4733              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4734              eptr++;
4735            }            }
4736          break;          break;
4737    
# Line 4411  for (;;) Line 4760  for (;;)
4760              {              {
4761              RMATCH(eptr, ecode, offset_top, md, eptrb, RM36);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM36);
4762              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4763              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4764              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4765                {                {
4766                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4767                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4768                }                }
4769              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4770              if (prop_fail_result) MRRETURN(MATCH_NOMATCH);              if (prop_fail_result) RRETURN(MATCH_NOMATCH);
4771              }              }
4772            /* Control never gets here */            /* Control never gets here */
4773    
# Line 4428  for (;;) Line 4777  for (;;)
4777              int chartype;              int chartype;
4778              RMATCH(eptr, ecode, offset_top, md, eptrb, RM37);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM37);
4779              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4780              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4781              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4782                {                {
4783                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4784                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4785                }                }
4786              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4787              chartype = UCD_CHARTYPE(c);              chartype = UCD_CHARTYPE(c);
4788              if ((chartype == ucp_Lu ||              if ((chartype == ucp_Lu ||
4789                   chartype == ucp_Ll ||                   chartype == ucp_Ll ||
4790                   chartype == ucp_Lt) == prop_fail_result)                   chartype == ucp_Lt) == prop_fail_result)
4791                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4792              }              }
4793            /* Control never gets here */            /* Control never gets here */
4794    
# Line 4448  for (;;) Line 4797  for (;;)
4797              {              {
4798              RMATCH(eptr, ecode, offset_top, md, eptrb, RM38);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM38);
4799              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4800              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4801              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4802                {                {
4803                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4804                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4805                }                }
4806              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4807              if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)              if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)
4808                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4809              }              }
4810            /* Control never gets here */            /* Control never gets here */
4811    
# Line 4465  for (;;) Line 4814  for (;;)
4814              {              {
4815              RMATCH(eptr, ecode, offset_top, md, eptrb, RM39);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM39);
4816              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4817              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4818              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4819                {                {
4820                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4821                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4822                }                }
4823              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4824              if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)              if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)
4825                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4826              }              }
4827            /* Control never gets here */            /* Control never gets here */
4828    
# Line 4482  for (;;) Line 4831  for (;;)
4831              {              {
4832              RMATCH(eptr, ecode, offset_top, md, eptrb, RM40);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM40);
4833              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4834              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4835              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4836                {                {
4837                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4838           &nbs