/[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 409 by ph10, Sat Mar 28 17:10:56 2009 UTC revision 600 by ph10, Mon May 9 08:54:11 2011 UTC
# Line 6  Line 6 
6  and semantics are as close as possible to those of the Perl 5 language.  and semantics are as close as possible to those of the Perl 5 language.
7    
8                         Written by Philip Hazel                         Written by Philip Hazel
9             Copyright (c) 1997-2009 University of Cambridge             Copyright (c) 1997-2011 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 71  defined PCRE_ERROR_xxx codes, which are Line 71  defined PCRE_ERROR_xxx codes, which are
71  /* Special internal returns from the match() function. Make them sufficiently  /* Special internal returns from the match() function. Make them sufficiently
72  negative to avoid the external error codes. */  negative to avoid the external error codes. */
73    
74  #define MATCH_COMMIT       (-999)  #define MATCH_ACCEPT       (-999)
75  #define MATCH_PRUNE        (-998)  #define MATCH_COMMIT       (-998)
76  #define MATCH_SKIP         (-997)  #define MATCH_PRUNE        (-997)
77  #define MATCH_THEN         (-996)  #define MATCH_SKIP         (-996)
78    #define MATCH_SKIP_ARG     (-995)
79    #define MATCH_THEN         (-994)
80    
81    /* This is a convenience macro for code that occurs many times. */
82    
83    #define MRRETURN(ra) \
84      { \
85      md->mark = markptr; \
86      RRETURN(ra); \
87      }
88    
89  /* 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.
90  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,
# Line 89  static const char rep_max[] = { 0, 0, 0, Line 99  static const char rep_max[] = { 0, 0, 0,
99    
100    
101    
102  #ifdef DEBUG  #ifdef PCRE_DEBUG
103  /*************************************************  /*************************************************
104  *        Debugging function to print chars       *  *        Debugging function to print chars       *
105  *************************************************/  *************************************************/
# Line 122  while (length-- > 0) Line 132  while (length-- > 0)
132  *          Match a back-reference                *  *          Match a back-reference                *
133  *************************************************/  *************************************************/
134    
135  /* If a back reference hasn't been set, the length that is passed is greater  /* Normally, if a back reference hasn't been set, the length that is passed is
136  than the number of characters left in the string, so the match fails.  negative, so the match always fails. However, in JavaScript compatibility mode,
137    the length passed is zero. Note that in caseless UTF-8 mode, the number of
138    subject bytes matched may be different to the number of reference bytes.
139    
140  Arguments:  Arguments:
141    offset      index into the offset vector    offset      index into the offset vector
142    eptr        points into the subject    eptr        pointer into the subject
143    length      length to be matched    length      length of reference to be matched (number of bytes)
144    md          points to match data block    md          points to match data block
145    ims         the ims flags    ims         the ims flags
146    
147  Returns:      TRUE if matched  Returns:      < 0 if not matched, otherwise the number of subject bytes matched
148  */  */
149    
150  static BOOL  static int
151  match_ref(int offset, register USPTR eptr, int length, match_data *md,  match_ref(int offset, register USPTR eptr, int length, match_data *md,
152    unsigned long int ims)    unsigned long int ims)
153  {  {
154  USPTR p = md->start_subject + md->offset_vector[offset];  USPTR eptr_start = eptr;
155    register USPTR p = md->start_subject + md->offset_vector[offset];
156    
157  #ifdef DEBUG  #ifdef PCRE_DEBUG
158  if (eptr >= md->end_subject)  if (eptr >= md->end_subject)
159    printf("matching subject <null>");    printf("matching subject <null>");
160  else  else
# Line 154  pchars(p, length, FALSE, md); Line 167  pchars(p, length, FALSE, md);
167  printf("\n");  printf("\n");
168  #endif  #endif
169    
170  /* Always fail if not enough characters left */  /* Always fail if reference not set (and not JavaScript compatible). */
171    
172  if (length > md->end_subject - eptr) return FALSE;  if (length < 0) return -1;
173    
174  /* Separate the caseless case for speed. In UTF-8 mode we can only do this  /* Separate the caseless case for speed. In UTF-8 mode we can only do this
175  properly if Unicode properties are supported. Otherwise, we can check only  properly if Unicode properties are supported. Otherwise, we can check only
# Line 168  if ((ims & PCRE_CASELESS) != 0) Line 181  if ((ims & PCRE_CASELESS) != 0)
181  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
182    if (md->utf8)    if (md->utf8)
183      {      {
184      USPTR endptr = eptr + length;      /* Match characters up to the end of the reference. NOTE: the number of
185      while (eptr < endptr)      bytes matched may differ, because there are some characters whose upper and
186        lower case versions code as different numbers of bytes. For example, U+023A
187        (2 bytes in UTF-8) is the upper case version of U+2C65 (3 bytes in UTF-8);
188        a sequence of 3 of the former uses 6 bytes, as does a sequence of two of
189        the latter. It is important, therefore, to check the length along the
190        reference, not along the subject (earlier code did this wrong). */
191    
192        USPTR endptr = p + length;
193        while (p < endptr)
194        {        {
195        int c, d;        int c, d;
196          if (eptr >= md->end_subject) return -1;
197        GETCHARINC(c, eptr);        GETCHARINC(c, eptr);
198        GETCHARINC(d, p);        GETCHARINC(d, p);
199        if (c != d && c != UCD_OTHERCASE(d)) return FALSE;        if (c != d && c != UCD_OTHERCASE(d)) return -1;
200        }        }
201      }      }
202    else    else
# Line 183  if ((ims & PCRE_CASELESS) != 0) Line 205  if ((ims & PCRE_CASELESS) != 0)
205    
206    /* The same code works when not in UTF-8 mode and in UTF-8 mode when there    /* The same code works when not in UTF-8 mode and in UTF-8 mode when there
207    is no UCP support. */    is no UCP support. */
208        {
209    while (length-- > 0)      if (eptr + length > md->end_subject) return -1;
210      { if (md->lcc[*p++] != md->lcc[*eptr++]) return FALSE; }      while (length-- > 0)
211          { if (md->lcc[*p++] != md->lcc[*eptr++]) return -1; }
212        }
213    }    }
214    
215  /* In the caseful case, we can just compare the bytes, whether or not we  /* In the caseful case, we can just compare the bytes, whether or not we
216  are in UTF-8 mode. */  are in UTF-8 mode. */
217    
218  else  else
219    { while (length-- > 0) if (*p++ != *eptr++) return FALSE; }    {
220      if (eptr + length > md->end_subject) return -1;
221      while (length-- > 0) if (*p++ != *eptr++) return -1;
222      }
223    
224  return TRUE;  return eptr - eptr_start;
225  }  }
226    
227    
# Line 245  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM Line 272  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM
272         RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,         RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,
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 };         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,
276           RM61,  RM62 };
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
280  actuall used in this definition. */  actually used in this definition. */
281    
282  #ifndef NO_RECURSE  #ifndef NO_RECURSE
283  #define REGISTER register  #define REGISTER register
284    
285  #ifdef DEBUG  #ifdef PCRE_DEBUG
286  #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw) \  #define RMATCH(ra,rb,rc,rd,re,rf,rg,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,rc,rd,re,rf,rg,rdepth+1); \    rrc = match(ra,rb,mstart,markptr,rc,rd,re,rf,rg,rdepth+1); \
290    printf("to line %d\n", __LINE__); \    printf("to line %d\n", __LINE__); \
291    }    }
292  #define RRETURN(ra) \  #define RRETURN(ra) \
# Line 268  actuall used in this definition. */ Line 296  actuall used in this definition. */
296    }    }
297  #else  #else
298  #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw) \  #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw) \
299    rrc = match(ra,rb,mstart,rc,rd,re,rf,rg,rdepth+1)    rrc = match(ra,rb,mstart,markptr,rc,rd,re,rf,rg,rdepth+1)
300  #define RRETURN(ra) return ra  #define RRETURN(ra) return ra
301  #endif  #endif
302    
# Line 283  argument of match(), which never changes Line 311  argument of match(), which never changes
311    
312  #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw)\  #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw)\
313    {\    {\
314    heapframe *newframe = (pcre_stack_malloc)(sizeof(heapframe));\    heapframe *newframe = (heapframe *)(pcre_stack_malloc)(sizeof(heapframe));\
315      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;\
320      newframe->Xmarkptr = markptr;\
321    newframe->Xoffset_top = rc;\    newframe->Xoffset_top = rc;\
322    newframe->Xims = re;\    newframe->Xims = re;\
323    newframe->Xeptrb = rf;\    newframe->Xeptrb = rf;\
# Line 303  argument of match(), which never changes Line 333  argument of match(), which never changes
333    
334  #define RRETURN(ra)\  #define RRETURN(ra)\
335    {\    {\
336    heapframe *newframe = frame;\    heapframe *oldframe = frame;\
337    frame = newframe->Xprevframe;\    frame = oldframe->Xprevframe;\
338    (pcre_stack_free)(newframe);\    (pcre_stack_free)(oldframe);\
339    if (frame != NULL)\    if (frame != NULL)\
340      {\      {\
341      rrc = ra;\      rrc = ra;\
# Line 325  typedef struct heapframe { Line 355  typedef struct heapframe {
355    USPTR Xeptr;    USPTR Xeptr;
356    const uschar *Xecode;    const uschar *Xecode;
357    USPTR Xmstart;    USPTR Xmstart;
358      USPTR Xmarkptr;
359    int Xoffset_top;    int Xoffset_top;
360    long int Xims;    long int Xims;
361    eptrblock *Xeptrb;    eptrblock *Xeptrb;
# Line 398  typedef struct heapframe { Line 429  typedef struct heapframe {
429    
430  /* This function is called recursively in many circumstances. Whenever it  /* This function is called recursively in many circumstances. Whenever it
431  returns a negative (error) response, the outer incarnation must also return the  returns a negative (error) response, the outer incarnation must also return the
432  same response.  same response. */
433    
434  Performance note: It might be tempting to extract commonly used fields from the  /* These macros pack up tests that are used for partial matching, and which
435  md structure (e.g. utf8, end_subject) into individual variables to improve  appears several times in the code. We set the "hit end" flag if the pointer is
436    at the end of the subject and also past the start of the subject (i.e.
437    something has been matched). For hard partial matching, we then return
438    immediately. The second one is used when we already know we are past the end of
439    the subject. */
440    
441    #define CHECK_PARTIAL()\
442      if (md->partial != 0 && eptr >= md->end_subject && \
443          eptr > md->start_used_ptr) \
444        { \
445        md->hitend = TRUE; \
446        if (md->partial > 1) MRRETURN(PCRE_ERROR_PARTIAL); \
447        }
448    
449    #define SCHECK_PARTIAL()\
450      if (md->partial != 0 && eptr > md->start_used_ptr) \
451        { \
452        md->hitend = TRUE; \
453        if (md->partial > 1) MRRETURN(PCRE_ERROR_PARTIAL); \
454        }
455    
456    
457    /* Performance note: It might be tempting to extract commonly used fields from
458    the md structure (e.g. utf8, end_subject) into individual variables to improve
459  performance. Tests using gcc on a SPARC disproved this; in the first case, it  performance. Tests using gcc on a SPARC disproved this; in the first case, it
460  made performance worse.  made performance worse.
461    
# Line 410  Arguments: Line 464  Arguments:
464     ecode       pointer to current position in compiled code     ecode       pointer to current position in compiled code
465     mstart      pointer to the current match start position (can be modified     mstart      pointer to the current match start position (can be modified
466                   by encountering \K)                   by encountering \K)
467       markptr     pointer to the most recent MARK name, or NULL
468     offset_top  current top pointer     offset_top  current top pointer
469     md          pointer to "static" info for the match     md          pointer to "static" info for the match
470     ims         current /i, /m, and /s options     ims         current /i, /m, and /s options
# Line 423  Arguments: Line 478  Arguments:
478    
479  Returns:       MATCH_MATCH if matched            )  these values are >= 0  Returns:       MATCH_MATCH if matched            )  these values are >= 0
480                 MATCH_NOMATCH if failed to match  )                 MATCH_NOMATCH if failed to match  )
481                   a negative MATCH_xxx value for PRUNE, SKIP, etc
482                 a negative PCRE_ERROR_xxx value if aborted by an error condition                 a negative PCRE_ERROR_xxx value if aborted by an error condition
483                   (e.g. stopped by repeated call or recursion limit)                   (e.g. stopped by repeated call or recursion limit)
484  */  */
485    
486  static int  static int
487  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, USPTR mstart,  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, USPTR mstart,
488    int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,    const uschar *markptr, int offset_top, match_data *md, unsigned long int ims,
489    int flags, unsigned int rdepth)    eptrblock *eptrb, int flags, unsigned int rdepth)
490  {  {
491  /* 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,
492  so they can be ordinary variables in all cases. Mark some of them with  so they can be ordinary variables in all cases. Mark some of them with
# Line 450  heap storage. Set up the top-level frame Line 506  heap storage. Set up the top-level frame
506  heap whenever RMATCH() does a "recursion". See the macro definitions above. */  heap whenever RMATCH() does a "recursion". See the macro definitions above. */
507    
508  #ifdef NO_RECURSE  #ifdef NO_RECURSE
509  heapframe *frame = (pcre_stack_malloc)(sizeof(heapframe));  heapframe *frame = (heapframe *)(pcre_stack_malloc)(sizeof(heapframe));
510    if (frame == NULL) RRETURN(PCRE_ERROR_NOMEMORY);
511  frame->Xprevframe = NULL;            /* Marks the top level */  frame->Xprevframe = NULL;            /* Marks the top level */
512    
513  /* Copy in the original argument variables */  /* Copy in the original argument variables */
# Line 458  frame->Xprevframe = NULL;            /* Line 515  frame->Xprevframe = NULL;            /*
515  frame->Xeptr = eptr;  frame->Xeptr = eptr;
516  frame->Xecode = ecode;  frame->Xecode = ecode;
517  frame->Xmstart = mstart;  frame->Xmstart = mstart;
518    frame->Xmarkptr = markptr;
519  frame->Xoffset_top = offset_top;  frame->Xoffset_top = offset_top;
520  frame->Xims = ims;  frame->Xims = ims;
521  frame->Xeptrb = eptrb;  frame->Xeptrb = eptrb;
# Line 473  HEAP_RECURSE: Line 531  HEAP_RECURSE:
531  #define eptr               frame->Xeptr  #define eptr               frame->Xeptr
532  #define ecode              frame->Xecode  #define ecode              frame->Xecode
533  #define mstart             frame->Xmstart  #define mstart             frame->Xmstart
534    #define markptr            frame->Xmarkptr
535  #define offset_top         frame->Xoffset_top  #define offset_top         frame->Xoffset_top
536  #define ims                frame->Xims  #define ims                frame->Xims
537  #define eptrb              frame->Xeptrb  #define eptrb              frame->Xeptrb
# Line 600  TAIL_RECURSE: Line 659  TAIL_RECURSE:
659  /* OK, now we can get on with the real code of the function. Recursive calls  /* OK, now we can get on with the real code of the function. Recursive calls
660  are specified by the macro RMATCH and RRETURN is used to return. When  are specified by the macro RMATCH and RRETURN is used to return. When
661  NO_RECURSE is *not* defined, these just turn into a recursive call to match()  NO_RECURSE is *not* defined, these just turn into a recursive call to match()
662  and a "return", respectively (possibly with some debugging if DEBUG is  and a "return", respectively (possibly with some debugging if PCRE_DEBUG is
663  defined). However, RMATCH isn't like a function call because it's quite a  defined). However, RMATCH isn't like a function call because it's quite a
664  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,
665  however, impact performance when true recursion is being used. */  however, impact performance when true recursion is being used. */
# Line 642  for (;;) Line 701  for (;;)
701    minimize = possessive = FALSE;    minimize = possessive = FALSE;
702    op = *ecode;    op = *ecode;
703    
   /* For partial matching, remember if we ever hit the end of the subject after  
   matching at least one subject character. */  
   
   if (md->partial &&  
       eptr >= md->end_subject &&  
       eptr > mstart)  
     md->hitend = TRUE;  
   
704    switch(op)    switch(op)
705      {      {
706        case OP_MARK:
707        markptr = ecode + 2;
708        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,
709          ims, eptrb, flags, RM55);
710    
711        /* A return of MATCH_SKIP_ARG means that matching failed at SKIP with an
712        argument, and we must check whether that argument matches this MARK's
713        argument. It is passed back in md->start_match_ptr (an overloading of that
714        variable). If it does match, we reset that variable to the current subject
715        position and return MATCH_SKIP. Otherwise, pass back the return code
716        unaltered. */
717    
718        if (rrc == MATCH_SKIP_ARG &&
719            strcmp((char *)markptr, (char *)(md->start_match_ptr)) == 0)
720          {
721          md->start_match_ptr = eptr;
722          RRETURN(MATCH_SKIP);
723          }
724    
725        if (md->mark == NULL) md->mark = markptr;
726        RRETURN(rrc);
727    
728      case OP_FAIL:      case OP_FAIL:
729      RRETURN(MATCH_NOMATCH);      MRRETURN(MATCH_NOMATCH);
730    
731        /* COMMIT overrides PRUNE, SKIP, and THEN */
732    
733        case OP_COMMIT:
734        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
735          ims, eptrb, flags, RM52);
736        if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE &&
737            rrc != MATCH_SKIP && rrc != MATCH_SKIP_ARG &&
738            rrc != MATCH_THEN)
739          RRETURN(rrc);
740        MRRETURN(MATCH_COMMIT);
741    
742        /* PRUNE overrides THEN */
743    
744      case OP_PRUNE:      case OP_PRUNE:
745      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
746        ims, eptrb, flags, RM51);        ims, eptrb, flags, RM51);
747      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
748        MRRETURN(MATCH_PRUNE);
749    
750        case OP_PRUNE_ARG:
751        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,
752          ims, eptrb, flags, RM56);
753        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
754        md->mark = ecode + 2;
755      RRETURN(MATCH_PRUNE);      RRETURN(MATCH_PRUNE);
756    
757      case OP_COMMIT:      /* SKIP overrides PRUNE and THEN */
     RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,  
       ims, eptrb, flags, RM52);  
     if (rrc != MATCH_NOMATCH) RRETURN(rrc);  
     RRETURN(MATCH_COMMIT);  
758    
759      case OP_SKIP:      case OP_SKIP:
760      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
761        ims, eptrb, flags, RM53);        ims, eptrb, flags, RM53);
762      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)
763          RRETURN(rrc);
764      md->start_match_ptr = eptr;   /* Pass back current position */      md->start_match_ptr = eptr;   /* Pass back current position */
765      RRETURN(MATCH_SKIP);      MRRETURN(MATCH_SKIP);
766    
767        case OP_SKIP_ARG:
768        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,
769          ims, eptrb, flags, RM57);
770        if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)
771          RRETURN(rrc);
772    
773        /* Pass back the current skip name by overloading md->start_match_ptr and
774        returning the special MATCH_SKIP_ARG return code. This will either be
775        caught by a matching MARK, or get to the top, where it is treated the same
776        as PRUNE. */
777    
778        md->start_match_ptr = ecode + 2;
779        RRETURN(MATCH_SKIP_ARG);
780    
781        /* For THEN (and THEN_ARG) we pass back the address of the bracket or
782        the alt that is at the start of the current branch. This makes it possible
783        to skip back past alternatives that precede the THEN within the current
784        branch. */
785    
786      case OP_THEN:      case OP_THEN:
787      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
788        ims, eptrb, flags, RM54);        ims, eptrb, flags, RM54);
789      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
790        md->start_match_ptr = ecode - GET(ecode, 1);
791        MRRETURN(MATCH_THEN);
792    
793        case OP_THEN_ARG:
794        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1+LINK_SIZE],
795          offset_top, md, ims, eptrb, flags, RM58);
796        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
797        md->start_match_ptr = ecode - GET(ecode, 1);
798        md->mark = ecode + LINK_SIZE + 2;
799      RRETURN(MATCH_THEN);      RRETURN(MATCH_THEN);
800    
801      /* Handle a capturing bracket. If there is space in the offset vector, save      /* Handle a capturing bracket. If there is space in the offset vector, save
# Line 699  for (;;) Line 817  for (;;)
817      number = GET2(ecode, 1+LINK_SIZE);      number = GET2(ecode, 1+LINK_SIZE);
818      offset = number << 1;      offset = number << 1;
819    
820  #ifdef DEBUG  #ifdef PCRE_DEBUG
821      printf("start bracket %d\n", number);      printf("start bracket %d\n", number);
822      printf("subject=");      printf("subject=");
823      pchars(eptr, 16, TRUE, md);      pchars(eptr, 16, TRUE, md);
# Line 714  for (;;) Line 832  for (;;)
832        save_capture_last = md->capture_last;        save_capture_last = md->capture_last;
833    
834        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));
835        md->offset_vector[md->offset_end - number] = eptr - md->start_subject;        md->offset_vector[md->offset_end - number] =
836            (int)(eptr - md->start_subject);
837    
838        flags = (op == OP_SCBRA)? match_cbegroup : 0;        flags = (op == OP_SCBRA)? match_cbegroup : 0;
839        do        do
840          {          {
841          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
842            ims, eptrb, flags, RM1);            ims, eptrb, flags, RM1);
843          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);          if (rrc != MATCH_NOMATCH &&
844                (rrc != MATCH_THEN || md->start_match_ptr != ecode))
845              RRETURN(rrc);
846          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
847          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
848          }          }
# Line 733  for (;;) Line 854  for (;;)
854        md->offset_vector[offset+1] = save_offset2;        md->offset_vector[offset+1] = save_offset2;
855        md->offset_vector[md->offset_end - number] = save_offset3;        md->offset_vector[md->offset_end - number] = save_offset3;
856    
857          if (rrc != MATCH_THEN) md->mark = markptr;
858        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
859        }        }
860    
# Line 772  for (;;) Line 894  for (;;)
894    
895          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
896            eptrb, flags, RM48);            eptrb, flags, RM48);
897            if (rrc == MATCH_NOMATCH) md->mark = markptr;
898          RRETURN(rrc);          RRETURN(rrc);
899          }          }
900    
# Line 780  for (;;) Line 903  for (;;)
903    
904        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
905          eptrb, flags, RM2);          eptrb, flags, RM2);
906        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);        if (rrc != MATCH_NOMATCH &&
907              (rrc != MATCH_THEN || md->start_match_ptr != ecode))
908            RRETURN(rrc);
909        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
910        }        }
911      /* Control never reaches here. */      /* Control never reaches here. */
# Line 807  for (;;) Line 932  for (;;)
932          cb.callout_number   = ecode[LINK_SIZE+2];          cb.callout_number   = ecode[LINK_SIZE+2];
933          cb.offset_vector    = md->offset_vector;          cb.offset_vector    = md->offset_vector;
934          cb.subject          = (PCRE_SPTR)md->start_subject;          cb.subject          = (PCRE_SPTR)md->start_subject;
935          cb.subject_length   = md->end_subject - md->start_subject;          cb.subject_length   = (int)(md->end_subject - md->start_subject);
936          cb.start_match      = mstart - md->start_subject;          cb.start_match      = (int)(mstart - md->start_subject);
937          cb.current_position = eptr - md->start_subject;          cb.current_position = (int)(eptr - md->start_subject);
938          cb.pattern_position = GET(ecode, LINK_SIZE + 3);          cb.pattern_position = GET(ecode, LINK_SIZE + 3);
939          cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);          cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);
940          cb.capture_top      = offset_top/2;          cb.capture_top      = offset_top/2;
941          cb.capture_last     = md->capture_last;          cb.capture_last     = md->capture_last;
942          cb.callout_data     = md->callout_data;          cb.callout_data     = md->callout_data;
943          if ((rrc = (*pcre_callout)(&cb)) > 0) RRETURN(MATCH_NOMATCH);          if ((rrc = (*pcre_callout)(&cb)) > 0) MRRETURN(MATCH_NOMATCH);
944          if (rrc < 0) RRETURN(rrc);          if (rrc < 0) RRETURN(rrc);
945          }          }
946        ecode += _pcre_OP_lengths[OP_CALLOUT];        ecode += _pcre_OP_lengths[OP_CALLOUT];
# Line 825  for (;;) Line 950  for (;;)
950    
951      /* Now see what the actual condition is */      /* Now see what the actual condition is */
952    
953      if (condcode == OP_RREF)         /* Recursion test */      if (condcode == OP_RREF || condcode == OP_NRREF)    /* Recursion test */
954        {        {
955        offset = GET2(ecode, LINK_SIZE + 2);     /* Recursion group number*/        if (md->recursive == NULL)                /* Not recursing => FALSE */
956        condition = md->recursive != NULL &&          {
957          (offset == RREF_ANY || offset == md->recursive->group_num);          condition = FALSE;
958        ecode += condition? 3 : GET(ecode, 1);          ecode += GET(ecode, 1);
959            }
960          else
961            {
962            int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/
963            condition =  (recno == RREF_ANY || recno == md->recursive->group_num);
964    
965            /* If the test is for recursion into a specific subpattern, and it is
966            false, but the test was set up by name, scan the table to see if the
967            name refers to any other numbers, and test them. The condition is true
968            if any one is set. */
969    
970            if (!condition && condcode == OP_NRREF && recno != RREF_ANY)
971              {
972              uschar *slotA = md->name_table;
973              for (i = 0; i < md->name_count; i++)
974                {
975                if (GET2(slotA, 0) == recno) break;
976                slotA += md->name_entry_size;
977                }
978    
979              /* Found a name for the number - there can be only one; duplicate
980              names for different numbers are allowed, but not vice versa. First
981              scan down for duplicates. */
982    
983              if (i < md->name_count)
984                {
985                uschar *slotB = slotA;
986                while (slotB > md->name_table)
987                  {
988                  slotB -= md->name_entry_size;
989                  if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)
990                    {
991                    condition = GET2(slotB, 0) == md->recursive->group_num;
992                    if (condition) break;
993                    }
994                  else break;
995                  }
996    
997                /* Scan up for duplicates */
998    
999                if (!condition)
1000                  {
1001                  slotB = slotA;
1002                  for (i++; i < md->name_count; i++)
1003                    {
1004                    slotB += md->name_entry_size;
1005                    if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)
1006                      {
1007                      condition = GET2(slotB, 0) == md->recursive->group_num;
1008                      if (condition) break;
1009                      }
1010                    else break;
1011                    }
1012                  }
1013                }
1014              }
1015    
1016            /* Chose branch according to the condition */
1017    
1018            ecode += condition? 3 : GET(ecode, 1);
1019            }
1020        }        }
1021    
1022      else if (condcode == OP_CREF)    /* Group used test */      else if (condcode == OP_CREF || condcode == OP_NCREF)  /* Group used test */
1023        {        {
1024        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */
1025        condition = offset < offset_top && md->offset_vector[offset] >= 0;        condition = offset < offset_top && md->offset_vector[offset] >= 0;
1026    
1027          /* If the numbered capture is unset, but the reference was by name,
1028          scan the table to see if the name refers to any other numbers, and test
1029          them. The condition is true if any one is set. This is tediously similar
1030          to the code above, but not close enough to try to amalgamate. */
1031    
1032          if (!condition && condcode == OP_NCREF)
1033            {
1034            int refno = offset >> 1;
1035            uschar *slotA = md->name_table;
1036    
1037            for (i = 0; i < md->name_count; i++)
1038              {
1039              if (GET2(slotA, 0) == refno) break;
1040              slotA += md->name_entry_size;
1041              }
1042    
1043            /* Found a name for the number - there can be only one; duplicate names
1044            for different numbers are allowed, but not vice versa. First scan down
1045            for duplicates. */
1046    
1047            if (i < md->name_count)
1048              {
1049              uschar *slotB = slotA;
1050              while (slotB > md->name_table)
1051                {
1052                slotB -= md->name_entry_size;
1053                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)
1054                  {
1055                  offset = GET2(slotB, 0) << 1;
1056                  condition = offset < offset_top &&
1057                    md->offset_vector[offset] >= 0;
1058                  if (condition) break;
1059                  }
1060                else break;
1061                }
1062    
1063              /* Scan up for duplicates */
1064    
1065              if (!condition)
1066                {
1067                slotB = slotA;
1068                for (i++; i < md->name_count; i++)
1069                  {
1070                  slotB += md->name_entry_size;
1071                  if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)
1072                    {
1073                    offset = GET2(slotB, 0) << 1;
1074                    condition = offset < offset_top &&
1075                      md->offset_vector[offset] >= 0;
1076                    if (condition) break;
1077                    }
1078                  else break;
1079                  }
1080                }
1081              }
1082            }
1083    
1084          /* Chose branch according to the condition */
1085    
1086        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 3 : GET(ecode, 1);
1087        }        }
1088    
# Line 860  for (;;) Line 1106  for (;;)
1106          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
1107          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
1108          }          }
1109        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)        else if (rrc != MATCH_NOMATCH &&
1110                  (rrc != MATCH_THEN || md->start_match_ptr != ecode))
1111          {          {
1112          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
1113          }          }
# Line 897  for (;;) Line 1144  for (;;)
1144      break;      break;
1145    
1146    
1147        /* Before OP_ACCEPT there may be any number of OP_CLOSE opcodes,
1148        to close any currently open capturing brackets. */
1149    
1150        case OP_CLOSE:
1151        number = GET2(ecode, 1);
1152        offset = number << 1;
1153    
1154    #ifdef PCRE_DEBUG
1155          printf("end bracket %d at *ACCEPT", number);
1156          printf("\n");
1157    #endif
1158    
1159        md->capture_last = number;
1160        if (offset >= md->offset_max) md->offset_overflow = TRUE; else
1161          {
1162          md->offset_vector[offset] =
1163            md->offset_vector[md->offset_end - number];
1164          md->offset_vector[offset+1] = (int)(eptr - md->start_subject);
1165          if (offset_top <= offset) offset_top = offset + 2;
1166          }
1167        ecode += 3;
1168        break;
1169    
1170    
1171      /* End of the pattern, either real or forced. If we are in a top-level      /* End of the pattern, either real or forced. If we are in a top-level
1172      recursion, we should restore the offsets appropriately and continue from      recursion, we should restore the offsets appropriately and continue from
1173      after the call. */      after the call. */
# Line 910  for (;;) Line 1181  for (;;)
1181        md->recursive = rec->prevrec;        md->recursive = rec->prevrec;
1182        memmove(md->offset_vector, rec->offset_save,        memmove(md->offset_vector, rec->offset_save,
1183          rec->saved_max * sizeof(int));          rec->saved_max * sizeof(int));
1184        mstart = rec->save_start;        offset_top = rec->save_offset_top;
1185        ims = original_ims;        ims = original_ims;
1186        ecode = rec->after_call;        ecode = rec->after_call;
1187        break;        break;
1188        }        }
1189    
1190      /* Otherwise, if PCRE_NOTEMPTY is set, fail if we have matched an empty      /* Otherwise, if we have matched an empty string, fail if PCRE_NOTEMPTY is
1191      string - backtracking will then try other alternatives, if any. */      set, or if PCRE_NOTEMPTY_ATSTART is set and we have matched at the start of
1192        the subject. In both cases, backtracking will then try other alternatives,
1193        if any. */
1194    
1195        if (eptr == mstart &&
1196            (md->notempty ||
1197              (md->notempty_atstart &&
1198                mstart == md->start_subject + md->start_offset)))
1199          MRRETURN(MATCH_NOMATCH);
1200    
1201        /* Otherwise, we have a match. */
1202    
     if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);  
1203      md->end_match_ptr = eptr;           /* Record where we ended */      md->end_match_ptr = eptr;           /* Record where we ended */
1204      md->end_offset_top = offset_top;    /* and how many extracts were taken */      md->end_offset_top = offset_top;    /* and how many extracts were taken */
1205      md->start_match_ptr = mstart;       /* and the start (\K can modify) */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
1206      RRETURN(MATCH_MATCH);  
1207        /* For some reason, the macros don't work properly if an expression is
1208        given as the argument to MRRETURN when the heap is in use. */
1209    
1210        rrc = (op == OP_END)? MATCH_MATCH : MATCH_ACCEPT;
1211        MRRETURN(rrc);
1212    
1213      /* Change option settings */      /* Change option settings */
1214    
# Line 945  for (;;) Line 1230  for (;;)
1230        {        {
1231        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
1232          RM4);          RM4);
1233        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
1234        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);          {
1235            mstart = md->start_match_ptr;   /* In case \K reset it */
1236            break;
1237            }
1238          if (rrc != MATCH_NOMATCH &&
1239              (rrc != MATCH_THEN || md->start_match_ptr != ecode))
1240            RRETURN(rrc);
1241        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1242        }        }
1243      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
1244      if (*ecode == OP_KET) RRETURN(MATCH_NOMATCH);      if (*ecode == OP_KET) MRRETURN(MATCH_NOMATCH);
1245    
1246      /* If checking an assertion for a condition, return MATCH_MATCH. */      /* If checking an assertion for a condition, return MATCH_MATCH. */
1247    
# Line 964  for (;;) Line 1255  for (;;)
1255      offset_top = md->end_offset_top;      offset_top = md->end_offset_top;
1256      continue;      continue;
1257    
1258      /* Negative assertion: all branches must fail to match */      /* Negative assertion: all branches must fail to match. Encountering SKIP,
1259        PRUNE, or COMMIT means we must assume failure without checking subsequent
1260        branches. */
1261    
1262      case OP_ASSERT_NOT:      case OP_ASSERT_NOT:
1263      case OP_ASSERTBACK_NOT:      case OP_ASSERTBACK_NOT:
# Line 972  for (;;) Line 1265  for (;;)
1265        {        {
1266        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
1267          RM5);          RM5);
1268        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) MRRETURN(MATCH_NOMATCH);
1269        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)
1270            {
1271            do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1272            break;
1273            }
1274          if (rrc != MATCH_NOMATCH &&
1275              (rrc != MATCH_THEN || md->start_match_ptr != ecode))
1276            RRETURN(rrc);
1277        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1278        }        }
1279      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 996  for (;;) Line 1296  for (;;)
1296        while (i-- > 0)        while (i-- > 0)
1297          {          {
1298          eptr--;          eptr--;
1299          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);          if (eptr < md->start_subject) MRRETURN(MATCH_NOMATCH);
1300          BACKCHAR(eptr);          BACKCHAR(eptr);
1301          }          }
1302        }        }
# Line 1007  for (;;) Line 1307  for (;;)
1307    
1308        {        {
1309        eptr -= GET(ecode, 1);        eptr -= GET(ecode, 1);
1310        if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);        if (eptr < md->start_subject) MRRETURN(MATCH_NOMATCH);
1311        }        }
1312    
1313      /* Skip to next op code */      /* Save the earliest consulted character, then skip to next op code */
1314    
1315        if (eptr < md->start_used_ptr) md->start_used_ptr = eptr;
1316      ecode += 1 + LINK_SIZE;      ecode += 1 + LINK_SIZE;
1317      break;      break;
1318    
# Line 1027  for (;;) Line 1328  for (;;)
1328        cb.callout_number   = ecode[1];        cb.callout_number   = ecode[1];
1329        cb.offset_vector    = md->offset_vector;        cb.offset_vector    = md->offset_vector;
1330        cb.subject          = (PCRE_SPTR)md->start_subject;        cb.subject          = (PCRE_SPTR)md->start_subject;
1331        cb.subject_length   = md->end_subject - md->start_subject;        cb.subject_length   = (int)(md->end_subject - md->start_subject);
1332        cb.start_match      = mstart - md->start_subject;        cb.start_match      = (int)(mstart - md->start_subject);
1333        cb.current_position = eptr - md->start_subject;        cb.current_position = (int)(eptr - md->start_subject);
1334        cb.pattern_position = GET(ecode, 2);        cb.pattern_position = GET(ecode, 2);
1335        cb.next_item_length = GET(ecode, 2 + LINK_SIZE);        cb.next_item_length = GET(ecode, 2 + LINK_SIZE);
1336        cb.capture_top      = offset_top/2;        cb.capture_top      = offset_top/2;
1337        cb.capture_last     = md->capture_last;        cb.capture_last     = md->capture_last;
1338        cb.callout_data     = md->callout_data;        cb.callout_data     = md->callout_data;
1339        if ((rrc = (*pcre_callout)(&cb)) > 0) RRETURN(MATCH_NOMATCH);        if ((rrc = (*pcre_callout)(&cb)) > 0) MRRETURN(MATCH_NOMATCH);
1340        if (rrc < 0) RRETURN(rrc);        if (rrc < 0) RRETURN(rrc);
1341        }        }
1342      ecode += 2 + 2*LINK_SIZE;      ecode += 2 + 2*LINK_SIZE;
# Line 1090  for (;;) Line 1391  for (;;)
1391    
1392        memcpy(new_recursive.offset_save, md->offset_vector,        memcpy(new_recursive.offset_save, md->offset_vector,
1393              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
1394        new_recursive.save_start = mstart;        new_recursive.save_offset_top = offset_top;
       mstart = eptr;  
1395    
1396        /* OK, now we can do the recursion. For each top-level alternative we        /* OK, now we can do the recursion. For each top-level alternative we
1397        restore the offset and recursion data. */        restore the offset and recursion data. */
# Line 1102  for (;;) Line 1402  for (;;)
1402          {          {
1403          RMATCH(eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,          RMATCH(eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,
1404            md, ims, eptrb, flags, RM6);            md, ims, eptrb, flags, RM6);
1405          if (rrc == MATCH_MATCH)          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
1406            {            {
1407            DPRINTF(("Recursion matched\n"));            DPRINTF(("Recursion matched\n"));
1408            md->recursive = new_recursive.prevrec;            md->recursive = new_recursive.prevrec;
1409            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
1410              (pcre_free)(new_recursive.offset_save);              (pcre_free)(new_recursive.offset_save);
1411            RRETURN(MATCH_MATCH);            MRRETURN(MATCH_MATCH);
1412            }            }
1413          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)          else if (rrc != MATCH_NOMATCH &&
1414                    (rrc != MATCH_THEN || md->start_match_ptr != ecode))
1415            {            {
1416            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1417            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
# Line 1129  for (;;) Line 1430  for (;;)
1430        md->recursive = new_recursive.prevrec;        md->recursive = new_recursive.prevrec;
1431        if (new_recursive.offset_save != stacksave)        if (new_recursive.offset_save != stacksave)
1432          (pcre_free)(new_recursive.offset_save);          (pcre_free)(new_recursive.offset_save);
1433        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
1434        }        }
1435      /* Control never reaches here */      /* Control never reaches here */
1436    
# Line 1138  for (;;) Line 1439  for (;;)
1439      a move back into the brackets. Friedl calls these "atomic" subpatterns.      a move back into the brackets. Friedl calls these "atomic" subpatterns.
1440      Check the alternative branches in turn - the matching won't pass the KET      Check the alternative branches in turn - the matching won't pass the KET
1441      for this kind of subpattern. If any one branch matches, we carry on as at      for this kind of subpattern. If any one branch matches, we carry on as at
1442      the end of a normal bracket, leaving the subject pointer. */      the end of a normal bracket, leaving the subject pointer, but resetting
1443        the start-of-match value in case it was changed by \K. */
1444    
1445      case OP_ONCE:      case OP_ONCE:
1446      prev = ecode;      prev = ecode;
# Line 1147  for (;;) Line 1449  for (;;)
1449      do      do
1450        {        {
1451        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM7);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM7);
1452        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH)  /* Note: _not_ MATCH_ACCEPT */
1453        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);          {
1454            mstart = md->start_match_ptr;
1455            break;
1456            }
1457          if (rrc != MATCH_NOMATCH &&
1458              (rrc != MATCH_THEN || md->start_match_ptr != ecode))
1459            RRETURN(rrc);
1460        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1461        }        }
1462      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 1266  for (;;) Line 1574  for (;;)
1574        }        }
1575      else saved_eptr = NULL;      else saved_eptr = NULL;
1576    
1577      /* 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 an atomic group, stop
1578      MATCH_MATCH, but record the current high water mark for use by positive      matching and return MATCH_MATCH, but record the current high water mark for
1579      assertions. Do this also for the "once" (atomic) groups. */      use by positive assertions. We also need to record the match start in case
1580        it was changed by \K. */
1581    
1582      if (*prev == OP_ASSERT || *prev == OP_ASSERT_NOT ||      if (*prev == OP_ASSERT || *prev == OP_ASSERT_NOT ||
1583          *prev == OP_ASSERTBACK || *prev == OP_ASSERTBACK_NOT ||          *prev == OP_ASSERTBACK || *prev == OP_ASSERTBACK_NOT ||
# Line 1276  for (;;) Line 1585  for (;;)
1585        {        {
1586        md->end_match_ptr = eptr;      /* For ONCE */        md->end_match_ptr = eptr;      /* For ONCE */
1587        md->end_offset_top = offset_top;        md->end_offset_top = offset_top;
1588        RRETURN(MATCH_MATCH);        md->start_match_ptr = mstart;
1589          MRRETURN(MATCH_MATCH);
1590        }        }
1591    
1592      /* 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 1290  for (;;) Line 1600  for (;;)
1600        number = GET2(prev, 1+LINK_SIZE);        number = GET2(prev, 1+LINK_SIZE);
1601        offset = number << 1;        offset = number << 1;
1602    
1603  #ifdef DEBUG  #ifdef PCRE_DEBUG
1604        printf("end bracket %d", number);        printf("end bracket %d", number);
1605        printf("\n");        printf("\n");
1606  #endif  #endif
# Line 1300  for (;;) Line 1610  for (;;)
1610          {          {
1611          md->offset_vector[offset] =          md->offset_vector[offset] =
1612            md->offset_vector[md->offset_end - number];            md->offset_vector[md->offset_end - number];
1613          md->offset_vector[offset+1] = eptr - md->start_subject;          md->offset_vector[offset+1] = (int)(eptr - md->start_subject);
1614          if (offset_top <= offset) offset_top = offset + 2;          if (offset_top <= offset) offset_top = offset + 2;
1615          }          }
1616    
# Line 1312  for (;;) Line 1622  for (;;)
1622          recursion_info *rec = md->recursive;          recursion_info *rec = md->recursive;
1623          DPRINTF(("Recursion (%d) succeeded - continuing\n", number));          DPRINTF(("Recursion (%d) succeeded - continuing\n", number));
1624          md->recursive = rec->prevrec;          md->recursive = rec->prevrec;
         mstart = rec->save_start;  
1625          memcpy(md->offset_vector, rec->offset_save,          memcpy(md->offset_vector, rec->offset_save,
1626            rec->saved_max * sizeof(int));            rec->saved_max * sizeof(int));
1627            offset_top = rec->save_offset_top;
1628          ecode = rec->after_call;          ecode = rec->after_call;
1629          ims = original_ims;          ims = original_ims;
1630          break;          break;
# Line 1371  for (;;) Line 1681  for (;;)
1681      /* Start of subject unless notbol, or after internal newline if multiline */      /* Start of subject unless notbol, or after internal newline if multiline */
1682    
1683      case OP_CIRC:      case OP_CIRC:
1684      if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH);      if (md->notbol && eptr == md->start_subject) MRRETURN(MATCH_NOMATCH);
1685      if ((ims & PCRE_MULTILINE) != 0)      if ((ims & PCRE_MULTILINE) != 0)
1686        {        {
1687        if (eptr != md->start_subject &&        if (eptr != md->start_subject &&
1688            (eptr == md->end_subject || !WAS_NEWLINE(eptr)))            (eptr == md->end_subject || !WAS_NEWLINE(eptr)))
1689          RRETURN(MATCH_NOMATCH);          MRRETURN(MATCH_NOMATCH);
1690        ecode++;        ecode++;
1691        break;        break;
1692        }        }
# Line 1385  for (;;) Line 1695  for (;;)
1695      /* Start of subject assertion */      /* Start of subject assertion */
1696    
1697      case OP_SOD:      case OP_SOD:
1698      if (eptr != md->start_subject) RRETURN(MATCH_NOMATCH);      if (eptr != md->start_subject) MRRETURN(MATCH_NOMATCH);
1699      ecode++;      ecode++;
1700      break;      break;
1701    
1702      /* Start of match assertion */      /* Start of match assertion */
1703    
1704      case OP_SOM:      case OP_SOM:
1705      if (eptr != md->start_subject + md->start_offset) RRETURN(MATCH_NOMATCH);      if (eptr != md->start_subject + md->start_offset) MRRETURN(MATCH_NOMATCH);
1706      ecode++;      ecode++;
1707      break;      break;
1708    
# Line 1410  for (;;) Line 1720  for (;;)
1720      if ((ims & PCRE_MULTILINE) != 0)      if ((ims & PCRE_MULTILINE) != 0)
1721        {        {
1722        if (eptr < md->end_subject)        if (eptr < md->end_subject)
1723          { if (!IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); }          { if (!IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH); }
1724        else        else
1725          { if (md->noteol) RRETURN(MATCH_NOMATCH); }          {
1726            if (md->noteol) MRRETURN(MATCH_NOMATCH);
1727            SCHECK_PARTIAL();
1728            }
1729        ecode++;        ecode++;
1730        break;        break;
1731        }        }
1732      else      else  /* Not multiline */
1733        {        {
1734        if (md->noteol) RRETURN(MATCH_NOMATCH);        if (md->noteol) MRRETURN(MATCH_NOMATCH);
1735        if (!md->endonly)        if (!md->endonly) goto ASSERT_NL_OR_EOS;
         {  
         if (eptr != md->end_subject &&  
             (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))  
           RRETURN(MATCH_NOMATCH);  
         ecode++;  
         break;  
         }  
1736        }        }
1737    
1738      /* ... else fall through for endonly */      /* ... else fall through for endonly */
1739    
1740      /* End of subject assertion (\z) */      /* End of subject assertion (\z) */
1741    
1742      case OP_EOD:      case OP_EOD:
1743      if (eptr < md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr < md->end_subject) MRRETURN(MATCH_NOMATCH);
1744        SCHECK_PARTIAL();
1745      ecode++;      ecode++;
1746      break;      break;
1747    
1748      /* End of subject or ending \n assertion (\Z) */      /* End of subject or ending \n assertion (\Z) */
1749    
1750      case OP_EODN:      case OP_EODN:
1751      if (eptr != md->end_subject &&      ASSERT_NL_OR_EOS:
1752        if (eptr < md->end_subject &&
1753          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))
1754        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
1755    
1756        /* Either at end of string or \n before end. */
1757    
1758        SCHECK_PARTIAL();
1759      ecode++;      ecode++;
1760      break;      break;
1761    
# Line 1454  for (;;) Line 1767  for (;;)
1767    
1768        /* Find out if the previous and current characters are "word" characters.        /* Find out if the previous and current characters are "word" characters.
1769        It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to        It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to
1770        be "non-word" characters. */        be "non-word" characters. Remember the earliest consulted character for
1771          partial matching. */
1772    
1773  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
1774        if (utf8)        if (utf8)
1775          {          {
1776            /* Get status of previous character */
1777    
1778          if (eptr == md->start_subject) prev_is_word = FALSE; else          if (eptr == md->start_subject) prev_is_word = FALSE; else
1779            {            {
1780            USPTR lastptr = eptr - 1;            USPTR lastptr = eptr - 1;
1781            while((*lastptr & 0xc0) == 0x80) lastptr--;            while((*lastptr & 0xc0) == 0x80) lastptr--;
1782              if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;
1783            GETCHAR(c, lastptr);            GETCHAR(c, lastptr);
1784    #ifdef SUPPORT_UCP
1785              if (md->use_ucp)
1786                {
1787                if (c == '_') prev_is_word = TRUE; else
1788                  {
1789                  int cat = UCD_CATEGORY(c);
1790                  prev_is_word = (cat == ucp_L || cat == ucp_N);
1791                  }
1792                }
1793              else
1794    #endif
1795            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
1796            }            }
1797          if (eptr >= md->end_subject) cur_is_word = FALSE; else  
1798            /* Get status of next character */
1799    
1800            if (eptr >= md->end_subject)
1801              {
1802              SCHECK_PARTIAL();
1803              cur_is_word = FALSE;
1804              }
1805            else
1806            {            {
1807            GETCHAR(c, eptr);            GETCHAR(c, eptr);
1808    #ifdef SUPPORT_UCP
1809              if (md->use_ucp)
1810                {
1811                if (c == '_') cur_is_word = TRUE; else
1812                  {
1813                  int cat = UCD_CATEGORY(c);
1814                  cur_is_word = (cat == ucp_L || cat == ucp_N);
1815                  }
1816                }
1817              else
1818    #endif
1819            cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;            cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
1820            }            }
1821          }          }
1822        else        else
1823  #endif  #endif
1824    
1825        /* More streamlined when not in UTF-8 mode */        /* Not in UTF-8 mode, but we may still have PCRE_UCP set, and for
1826          consistency with the behaviour of \w we do use it in this case. */
1827    
1828          {          {
1829          prev_is_word = (eptr != md->start_subject) &&          /* Get status of previous character */
1830            ((md->ctypes[eptr[-1]] & ctype_word) != 0);  
1831          cur_is_word = (eptr < md->end_subject) &&          if (eptr == md->start_subject) prev_is_word = FALSE; else
1832            ((md->ctypes[*eptr] & ctype_word) != 0);            {
1833              if (eptr <= md->start_used_ptr) md->start_used_ptr = eptr - 1;
1834    #ifdef SUPPORT_UCP
1835              if (md->use_ucp)
1836                {
1837                c = eptr[-1];
1838                if (c == '_') prev_is_word = TRUE; else
1839                  {
1840                  int cat = UCD_CATEGORY(c);
1841                  prev_is_word = (cat == ucp_L || cat == ucp_N);
1842                  }
1843                }
1844              else
1845    #endif
1846              prev_is_word = ((md->ctypes[eptr[-1]] & ctype_word) != 0);
1847              }
1848    
1849            /* Get status of next character */
1850    
1851            if (eptr >= md->end_subject)
1852              {
1853              SCHECK_PARTIAL();
1854              cur_is_word = FALSE;
1855              }
1856            else
1857    #ifdef SUPPORT_UCP
1858            if (md->use_ucp)
1859              {
1860              c = *eptr;
1861              if (c == '_') cur_is_word = TRUE; else
1862                {
1863                int cat = UCD_CATEGORY(c);
1864                cur_is_word = (cat == ucp_L || cat == ucp_N);
1865                }
1866              }
1867            else
1868    #endif
1869            cur_is_word = ((md->ctypes[*eptr] & ctype_word) != 0);
1870          }          }
1871    
1872        /* Now see if the situation is what we want */        /* Now see if the situation is what we want */
1873    
1874        if ((*ecode++ == OP_WORD_BOUNDARY)?        if ((*ecode++ == OP_WORD_BOUNDARY)?
1875             cur_is_word == prev_is_word : cur_is_word != prev_is_word)             cur_is_word == prev_is_word : cur_is_word != prev_is_word)
1876          RRETURN(MATCH_NOMATCH);          MRRETURN(MATCH_NOMATCH);
1877        }        }
1878      break;      break;
1879    
1880      /* Match a single character type; inline for speed */      /* Match a single character type; inline for speed */
1881    
1882      case OP_ANY:      case OP_ANY:
1883      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);      if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);
1884      /* Fall through */      /* Fall through */
1885    
1886      case OP_ALLANY:      case OP_ALLANY:
1887      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr++ >= md->end_subject)
1888          {
1889          SCHECK_PARTIAL();
1890          MRRETURN(MATCH_NOMATCH);
1891          }
1892      if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;      if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
1893      ecode++;      ecode++;
1894      break;      break;
# Line 1508  for (;;) Line 1897  for (;;)
1897      any byte, even newline, independent of the setting of PCRE_DOTALL. */      any byte, even newline, independent of the setting of PCRE_DOTALL. */
1898    
1899      case OP_ANYBYTE:      case OP_ANYBYTE:
1900      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr++ >= md->end_subject)
1901          {
1902          SCHECK_PARTIAL();
1903          MRRETURN(MATCH_NOMATCH);
1904          }
1905      ecode++;      ecode++;
1906      break;      break;
1907    
1908      case OP_NOT_DIGIT:      case OP_NOT_DIGIT:
1909      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1910          {
1911          SCHECK_PARTIAL();
1912          MRRETURN(MATCH_NOMATCH);
1913          }
1914      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1915      if (      if (
1916  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1521  for (;;) Line 1918  for (;;)
1918  #endif  #endif
1919         (md->ctypes[c] & ctype_digit) != 0         (md->ctypes[c] & ctype_digit) != 0
1920         )         )
1921        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
1922      ecode++;      ecode++;
1923      break;      break;
1924    
1925      case OP_DIGIT:      case OP_DIGIT:
1926      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1927          {
1928          SCHECK_PARTIAL();
1929          MRRETURN(MATCH_NOMATCH);
1930          }
1931      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1932      if (      if (
1933  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1534  for (;;) Line 1935  for (;;)
1935  #endif  #endif
1936         (md->ctypes[c] & ctype_digit) == 0         (md->ctypes[c] & ctype_digit) == 0
1937         )         )
1938        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
1939      ecode++;      ecode++;
1940      break;      break;
1941    
1942      case OP_NOT_WHITESPACE:      case OP_NOT_WHITESPACE:
1943      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1944          {
1945          SCHECK_PARTIAL();
1946          MRRETURN(MATCH_NOMATCH);
1947          }
1948      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1949      if (      if (
1950  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1547  for (;;) Line 1952  for (;;)
1952  #endif  #endif
1953         (md->ctypes[c] & ctype_space) != 0         (md->ctypes[c] & ctype_space) != 0
1954         )         )
1955        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
1956      ecode++;      ecode++;
1957      break;      break;
1958    
1959      case OP_WHITESPACE:      case OP_WHITESPACE:
1960      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1961          {
1962          SCHECK_PARTIAL();
1963          MRRETURN(MATCH_NOMATCH);
1964          }
1965      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1966      if (      if (
1967  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1560  for (;;) Line 1969  for (;;)
1969  #endif  #endif
1970         (md->ctypes[c] & ctype_space) == 0         (md->ctypes[c] & ctype_space) == 0
1971         )         )
1972        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
1973      ecode++;      ecode++;
1974      break;      break;
1975    
1976      case OP_NOT_WORDCHAR:      case OP_NOT_WORDCHAR:
1977      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1978          {
1979          SCHECK_PARTIAL();
1980          MRRETURN(MATCH_NOMATCH);
1981          }
1982      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1983      if (      if (
1984  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1573  for (;;) Line 1986  for (;;)
1986  #endif  #endif
1987         (md->ctypes[c] & ctype_word) != 0         (md->ctypes[c] & ctype_word) != 0
1988         )         )
1989        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
1990      ecode++;      ecode++;
1991      break;      break;
1992    
1993      case OP_WORDCHAR:      case OP_WORDCHAR:
1994      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1995          {
1996          SCHECK_PARTIAL();
1997          MRRETURN(MATCH_NOMATCH);
1998          }
1999      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2000      if (      if (
2001  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1586  for (;;) Line 2003  for (;;)
2003  #endif  #endif
2004         (md->ctypes[c] & ctype_word) == 0         (md->ctypes[c] & ctype_word) == 0
2005         )         )
2006        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
2007      ecode++;      ecode++;
2008      break;      break;
2009    
2010      case OP_ANYNL:      case OP_ANYNL:
2011      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2012          {
2013          SCHECK_PARTIAL();
2014          MRRETURN(MATCH_NOMATCH);
2015          }
2016      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2017      switch(c)      switch(c)
2018        {        {
2019        default: RRETURN(MATCH_NOMATCH);        default: MRRETURN(MATCH_NOMATCH);
2020    
2021        case 0x000d:        case 0x000d:
2022        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
2023        break;        break;
# Line 1608  for (;;) Line 2030  for (;;)
2030        case 0x0085:        case 0x0085:
2031        case 0x2028:        case 0x2028:
2032        case 0x2029:        case 0x2029:
2033        if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);        if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);
2034        break;        break;
2035        }        }
2036      ecode++;      ecode++;
2037      break;      break;
2038    
2039      case OP_NOT_HSPACE:      case OP_NOT_HSPACE:
2040      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2041          {
2042          SCHECK_PARTIAL();
2043          MRRETURN(MATCH_NOMATCH);
2044          }
2045      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2046      switch(c)      switch(c)
2047        {        {
# Line 1639  for (;;) Line 2065  for (;;)
2065        case 0x202f:    /* NARROW NO-BREAK SPACE */        case 0x202f:    /* NARROW NO-BREAK SPACE */
2066        case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */        case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
2067        case 0x3000:    /* IDEOGRAPHIC SPACE */        case 0x3000:    /* IDEOGRAPHIC SPACE */
2068        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
2069        }        }
2070      ecode++;      ecode++;
2071      break;      break;
2072    
2073      case OP_HSPACE:      case OP_HSPACE:
2074      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2075          {
2076          SCHECK_PARTIAL();
2077          MRRETURN(MATCH_NOMATCH);
2078          }
2079      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2080      switch(c)      switch(c)
2081        {        {
2082        default: RRETURN(MATCH_NOMATCH);        default: MRRETURN(MATCH_NOMATCH);
2083        case 0x09:      /* HT */        case 0x09:      /* HT */
2084        case 0x20:      /* SPACE */        case 0x20:      /* SPACE */
2085        case 0xa0:      /* NBSP */        case 0xa0:      /* NBSP */
# Line 1675  for (;;) Line 2105  for (;;)
2105      break;      break;
2106    
2107      case OP_NOT_VSPACE:      case OP_NOT_VSPACE:
2108      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2109          {
2110          SCHECK_PARTIAL();
2111          MRRETURN(MATCH_NOMATCH);
2112          }
2113      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2114      switch(c)      switch(c)
2115        {        {
# Line 1687  for (;;) Line 2121  for (;;)
2121        case 0x85:      /* NEL */        case 0x85:      /* NEL */
2122        case 0x2028:    /* LINE SEPARATOR */        case 0x2028:    /* LINE SEPARATOR */
2123        case 0x2029:    /* PARAGRAPH SEPARATOR */        case 0x2029:    /* PARAGRAPH SEPARATOR */
2124        RRETURN(MATCH_NOMATCH);        MRRETURN(MATCH_NOMATCH);
2125        }        }
2126      ecode++;      ecode++;
2127      break;      break;
2128    
2129      case OP_VSPACE:      case OP_VSPACE:
2130      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2131          {
2132          SCHECK_PARTIAL();
2133          MRRETURN(MATCH_NOMATCH);
2134          }
2135      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2136      switch(c)      switch(c)
2137        {        {
2138        default: RRETURN(MATCH_NOMATCH);        default: MRRETURN(MATCH_NOMATCH);
2139        case 0x0a:      /* LF */        case 0x0a:      /* LF */
2140        case 0x0b:      /* VT */        case 0x0b:      /* VT */
2141        case 0x0c:      /* FF */        case 0x0c:      /* FF */
# Line 1716  for (;;) Line 2154  for (;;)
2154    
2155      case OP_PROP:      case OP_PROP:
2156      case OP_NOTPROP:      case OP_NOTPROP:
2157      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2158          {
2159          SCHECK_PARTIAL();
2160          MRRETURN(MATCH_NOMATCH);
2161          }
2162      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2163        {        {
2164        const ucd_record *prop = GET_UCD(c);        const ucd_record *prop = GET_UCD(c);
# Line 1724  for (;;) Line 2166  for (;;)
2166        switch(ecode[1])        switch(ecode[1])
2167          {          {
2168          case PT_ANY:          case PT_ANY:
2169          if (op == OP_NOTPROP) RRETURN(MATCH_NOMATCH);          if (op == OP_NOTPROP) MRRETURN(MATCH_NOMATCH);
2170          break;          break;
2171    
2172          case PT_LAMP:          case PT_LAMP:
2173          if ((prop->chartype == ucp_Lu ||          if ((prop->chartype == ucp_Lu ||
2174               prop->chartype == ucp_Ll ||               prop->chartype == ucp_Ll ||
2175               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))
2176            RRETURN(MATCH_NOMATCH);            MRRETURN(MATCH_NOMATCH);
2177           break;          break;
2178    
2179          case PT_GC:          case PT_GC:
2180          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))
2181            RRETURN(MATCH_NOMATCH);            MRRETURN(MATCH_NOMATCH);
2182          break;          break;
2183    
2184          case PT_PC:          case PT_PC:
2185          if ((ecode[2] != prop->chartype) == (op == OP_PROP))          if ((ecode[2] != prop->chartype) == (op == OP_PROP))
2186            RRETURN(MATCH_NOMATCH);            MRRETURN(MATCH_NOMATCH);
2187          break;          break;
2188    
2189          case PT_SC:          case PT_SC:
2190          if ((ecode[2] != prop->script) == (op == OP_PROP))          if ((ecode[2] != prop->script) == (op == OP_PROP))
2191            RRETURN(MATCH_NOMATCH);            MRRETURN(MATCH_NOMATCH);
2192            break;
2193    
2194            /* These are specials */
2195    
2196            case PT_ALNUM:
2197            if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||
2198                 _pcre_ucp_gentype[prop->chartype] == ucp_N) == (op == OP_NOTPROP))
2199              MRRETURN(MATCH_NOMATCH);
2200          break;          break;
2201    
2202            case PT_SPACE:    /* Perl space */
2203            if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||
2204                 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
2205                   == (op == OP_NOTPROP))
2206              MRRETURN(MATCH_NOMATCH);
2207            break;
2208    
2209            case PT_PXSPACE:  /* POSIX space */
2210            if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||
2211                 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
2212                 c == CHAR_FF || c == CHAR_CR)
2213                   == (op == OP_NOTPROP))
2214              MRRETURN(MATCH_NOMATCH);
2215            break;
2216    
2217            case PT_WORD:
2218            if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||
2219                 _pcre_ucp_gentype[prop->chartype] == ucp_N ||
2220                 c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))
2221              MRRETURN(MATCH_NOMATCH);
2222            break;
2223    
2224            /* This should never occur */
2225    
2226          default:          default:
2227          RRETURN(PCRE_ERROR_INTERNAL);          RRETURN(PCRE_ERROR_INTERNAL);
2228          }          }
# Line 1761  for (;;) Line 2235  for (;;)
2235      is in the binary; otherwise a compile-time error occurs. */      is in the binary; otherwise a compile-time error occurs. */
2236    
2237      case OP_EXTUNI:      case OP_EXTUNI:
2238      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2239          {
2240          SCHECK_PARTIAL();
2241          MRRETURN(MATCH_NOMATCH);
2242          }
2243      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2244        {        {
2245        int category = UCD_CATEGORY(c);        int category = UCD_CATEGORY(c);
2246        if (category == ucp_M) RRETURN(MATCH_NOMATCH);        if (category == ucp_M) MRRETURN(MATCH_NOMATCH);
2247        while (eptr < md->end_subject)        while (eptr < md->end_subject)
2248          {          {
2249          int len = 1;          int len = 1;
# Line 1792  for (;;) Line 2270  for (;;)
2270      loops). */      loops). */
2271    
2272      case OP_REF:      case OP_REF:
2273        {      offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
2274        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */      ecode += 3;
       ecode += 3;  
2275    
2276        /* If the reference is unset, there are two possibilities:      /* If the reference is unset, there are two possibilities:
2277    
2278        (a) In the default, Perl-compatible state, set the length to be longer      (a) In the default, Perl-compatible state, set the length negative;
2279        than the amount of subject left; this ensures that every attempt at a      this ensures that every attempt at a match fails. We can't just fail
2280        match fails. We can't just fail here, because of the possibility of      here, because of the possibility of quantifiers with zero minima.
       quantifiers with zero minima.  
2281    
2282        (b) If the JavaScript compatibility flag is set, set the length to zero      (b) If the JavaScript compatibility flag is set, set the length to zero
2283        so that the back reference matches an empty string.      so that the back reference matches an empty string.
2284    
2285        Otherwise, set the length to the length of what was matched by the      Otherwise, set the length to the length of what was matched by the
2286        referenced subpattern. */      referenced subpattern. */
2287    
2288        if (offset >= offset_top || md->offset_vector[offset] < 0)      if (offset >= offset_top || md->offset_vector[offset] < 0)
2289          length = (md->jscript_compat)? 0 : md->end_subject - eptr + 1;        length = (md->jscript_compat)? 0 : -1;
2290        else      else
2291          length = md->offset_vector[offset+1] - md->offset_vector[offset];        length = md->offset_vector[offset+1] - md->offset_vector[offset];
2292    
2293        /* Set up for repetition, or handle the non-repeated case */      /* Set up for repetition, or handle the non-repeated case */
2294    
2295        switch (*ecode)      switch (*ecode)
2296          {        {
2297          case OP_CRSTAR:        case OP_CRSTAR:
2298          case OP_CRMINSTAR:        case OP_CRMINSTAR:
2299          case OP_CRPLUS:        case OP_CRPLUS:
2300          case OP_CRMINPLUS:        case OP_CRMINPLUS:
2301          case OP_CRQUERY:        case OP_CRQUERY:
2302          case OP_CRMINQUERY:        case OP_CRMINQUERY:
2303          c = *ecode++ - OP_CRSTAR;        c = *ecode++ - OP_CRSTAR;
2304          minimize = (c & 1) != 0;        minimize = (c & 1) != 0;
2305          min = rep_min[c];                 /* Pick up values from tables; */        min = rep_min[c];                 /* Pick up values from tables; */
2306          max = rep_max[c];                 /* zero for max => infinity */        max = rep_max[c];                 /* zero for max => infinity */
2307          if (max == 0) max = INT_MAX;        if (max == 0) max = INT_MAX;
2308          break;        break;
2309    
2310          case OP_CRRANGE:        case OP_CRRANGE:
2311          case OP_CRMINRANGE:        case OP_CRMINRANGE:
2312          minimize = (*ecode == OP_CRMINRANGE);        minimize = (*ecode == OP_CRMINRANGE);
2313          min = GET2(ecode, 1);        min = GET2(ecode, 1);
2314          max = GET2(ecode, 3);        max = GET2(ecode, 3);
2315          if (max == 0) max = INT_MAX;        if (max == 0) max = INT_MAX;
2316          ecode += 5;        ecode += 5;
2317          break;        break;
2318    
2319          default:               /* No repeat follows */        default:               /* No repeat follows */
2320          if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH);        if ((length = match_ref(offset, eptr, length, md, ims)) < 0)
2321          eptr += length;          {
2322          continue;              /* With the main loop */          CHECK_PARTIAL();
2323            MRRETURN(MATCH_NOMATCH);
2324          }          }
2325          eptr += length;
2326          continue;              /* With the main loop */
2327          }
2328    
2329        /* If the length of the reference is zero, just continue with the      /* Handle repeated back references. If the length of the reference is
2330        main loop. */      zero, just continue with the main loop. */
2331    
2332        if (length == 0) continue;      if (length == 0) continue;
2333    
2334        /* First, ensure the minimum number of matches are present. We get back      /* First, ensure the minimum number of matches are present. We get back
2335        the length of the reference string explicitly rather than passing the      the length of the reference string explicitly rather than passing the
2336        address of eptr, so that eptr can be a register variable. */      address of eptr, so that eptr can be a register variable. */
2337    
2338        for (i = 1; i <= min; i++)      for (i = 1; i <= min; i++)
2339          {
2340          int slength;
2341          if ((slength = match_ref(offset, eptr, length, md, ims)) < 0)
2342          {          {
2343          if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH);          CHECK_PARTIAL();
2344          eptr += length;          MRRETURN(MATCH_NOMATCH);
2345          }          }
2346          eptr += slength;
2347          }
2348    
2349        /* If min = max, continue at the same level without recursion.      /* If min = max, continue at the same level without recursion.
2350        They are not both allowed to be zero. */      They are not both allowed to be zero. */
2351    
2352        if (min == max) continue;      if (min == max) continue;
2353    
2354        /* If minimizing, keep trying and advancing the pointer */      /* If minimizing, keep trying and advancing the pointer */
2355    
2356        if (minimize)      if (minimize)
2357          {
2358          for (fi = min;; fi++)
2359          {          {
2360          for (fi = min;; fi++)          int slength;
2361            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);
2362            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2363            if (fi >= max) MRRETURN(MATCH_NOMATCH);
2364            if ((slength = match_ref(offset, eptr, length, md, ims)) < 0)
2365            {            {
2366            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);            CHECK_PARTIAL();
2367            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            MRRETURN(MATCH_NOMATCH);
           if (fi >= max || !match_ref(offset, eptr, length, md, ims))  
             RRETURN(MATCH_NOMATCH);  
           eptr += length;  
2368            }            }
2369          /* Control never gets here */          eptr += slength;
2370          }          }
2371          /* Control never gets here */
2372          }
2373    
2374        /* If maximizing, find the longest string and work backwards */      /* If maximizing, find the longest string and work backwards */
2375    
2376        else      else
2377          {
2378          pp = eptr;
2379          for (i = min; i < max; i++)
2380          {          {
2381          pp = eptr;          int slength;
2382          for (i = min; i < max; i++)          if ((slength = match_ref(offset, eptr, length, md, ims)) < 0)
           {  
           if (!match_ref(offset, eptr, length, md, ims)) break;  
           eptr += length;  
           }  
         while (eptr >= pp)  
2383            {            {
2384            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM15);            CHECK_PARTIAL();
2385            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            break;
           eptr -= length;  
2386            }            }
2387          RRETURN(MATCH_NOMATCH);          eptr += slength;
2388          }          }
2389          while (eptr >= pp)
2390            {
2391            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM15);
2392            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2393            eptr -= length;
2394            }
2395          MRRETURN(MATCH_NOMATCH);
2396        }        }
2397      /* Control never gets here */      /* Control never gets here */
2398    
   
   
2399      /* Match a bit-mapped character class, possibly repeatedly. This op code is      /* Match a bit-mapped character class, possibly repeatedly. This op code is
2400      used when all the characters in the class have values in the range 0-255,      used when all the characters in the class have values in the range 0-255,
2401      and either the matching is caseful, or the characters are in the range      and either the matching is caseful, or the characters are in the range
# Line 1958  for (;;) Line 2450  for (;;)
2450          {          {
2451          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2452            {            {
2453            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
2454                {
2455                SCHECK_PARTIAL();
2456                MRRETURN(MATCH_NOMATCH);
2457                }
2458            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
2459            if (c > 255)            if (c > 255)
2460              {              {
2461              if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);              if (op == OP_CLASS) MRRETURN(MATCH_NOMATCH);
2462              }              }
2463            else            else
2464              {              {
2465              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);              if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);
2466              }              }
2467            }            }
2468          }          }
# Line 1976  for (;;) Line 2472  for (;;)
2472          {          {
2473          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2474            {            {
2475            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
2476                {
2477                SCHECK_PARTIAL();
2478                MRRETURN(MATCH_NOMATCH);
2479                }
2480            c = *eptr++;            c = *eptr++;
2481            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);            if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);
2482            }            }
2483          }          }
2484    
# Line 2000  for (;;) Line 2500  for (;;)
2500              {              {
2501              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);
2502              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2503              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
2504                if (eptr >= md->end_subject)
2505                  {
2506                  SCHECK_PARTIAL();
2507                  MRRETURN(MATCH_NOMATCH);
2508                  }
2509              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
2510              if (c > 255)              if (c > 255)
2511                {                {
2512                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);                if (op == OP_CLASS) MRRETURN(MATCH_NOMATCH);
2513                }                }
2514              else              else
2515                {                {
2516                if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);                if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);
2517                }                }
2518              }              }
2519            }            }
# Line 2020  for (;;) Line 2525  for (;;)
2525              {              {
2526              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);
2527              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2528              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
2529                if (eptr >= md->end_subject)
2530                  {
2531                  SCHECK_PARTIAL();
2532                  MRRETURN(MATCH_NOMATCH);
2533                  }
2534              c = *eptr++;              c = *eptr++;
2535              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);              if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);
2536              }              }
2537            }            }
2538          /* Control never gets here */          /* Control never gets here */
# Line 2041  for (;;) Line 2551  for (;;)
2551            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2552              {              {
2553              int len = 1;              int len = 1;
2554              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
2555                  {
2556                  SCHECK_PARTIAL();
2557                  break;
2558                  }
2559              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
2560              if (c > 255)              if (c > 255)
2561                {                {
# Line 2067  for (;;) Line 2581  for (;;)
2581            {            {
2582            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2583              {              {
2584              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
2585                  {
2586                  SCHECK_PARTIAL();
2587                  break;
2588                  }
2589              c = *eptr;              c = *eptr;
2590              if ((data[c/8] & (1 << (c&7))) == 0) break;              if ((data[c/8] & (1 << (c&7))) == 0) break;
2591              eptr++;              eptr++;
# Line 2080  for (;;) Line 2598  for (;;)
2598              }              }
2599            }            }
2600    
2601          RRETURN(MATCH_NOMATCH);          MRRETURN(MATCH_NOMATCH);
2602          }          }
2603        }        }
2604      /* Control never gets here */      /* Control never gets here */
# Line 2129  for (;;) Line 2647  for (;;)
2647    
2648        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2649          {          {
2650          if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);          if (eptr >= md->end_subject)
2651              {
2652              SCHECK_PARTIAL();
2653              MRRETURN(MATCH_NOMATCH);
2654              }
2655          GETCHARINCTEST(c, eptr);          GETCHARINCTEST(c, eptr);
2656          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);          if (!_pcre_xclass(c, data)) MRRETURN(MATCH_NOMATCH);
2657          }          }
2658    
2659        /* 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 2148  for (;;) Line 2670  for (;;)
2670            {            {
2671            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);
2672            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2673            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max) MRRETURN(MATCH_NOMATCH);
2674              if (eptr >= md->end_subject)
2675                {
2676                SCHECK_PARTIAL();
2677                MRRETURN(MATCH_NOMATCH);
2678                }
2679            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
2680            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);            if (!_pcre_xclass(c, data)) MRRETURN(MATCH_NOMATCH);
2681            }            }
2682          /* Control never gets here */          /* Control never gets here */
2683          }          }
# Line 2163  for (;;) Line 2690  for (;;)
2690          for (i = min; i < max; i++)          for (i = min; i < max; i++)
2691            {            {
2692            int len = 1;            int len = 1;
2693            if (eptr >= md->end_subject) break;            if (eptr >= md->end_subject)
2694                {
2695                SCHECK_PARTIAL();
2696                break;
2697                }
2698            GETCHARLENTEST(c, eptr, len);            GETCHARLENTEST(c, eptr, len);
2699            if (!_pcre_xclass(c, data)) break;            if (!_pcre_xclass(c, data)) break;
2700            eptr += len;            eptr += len;
# Line 2175  for (;;) Line 2706  for (;;)
2706            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
2707            if (utf8) BACKCHAR(eptr);            if (utf8) BACKCHAR(eptr);
2708            }            }
2709          RRETURN(MATCH_NOMATCH);          MRRETURN(MATCH_NOMATCH);
2710          }          }
2711    
2712        /* Control never gets here */        /* Control never gets here */
# Line 2191  for (;;) Line 2722  for (;;)
2722        length = 1;        length = 1;
2723        ecode++;        ecode++;
2724        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
2725        if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);        if (length > md->end_subject - eptr)
2726        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);          {
2727            CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
2728            MRRETURN(MATCH_NOMATCH);
2729            }
2730          while (length-- > 0) if (*ecode++ != *eptr++) MRRETURN(MATCH_NOMATCH);
2731        }        }
2732      else      else
2733  #endif  #endif
2734    
2735      /* Non-UTF-8 mode */      /* Non-UTF-8 mode */
2736        {        {
2737        if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH);        if (md->end_subject - eptr < 1)
2738        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);          {
2739            SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
2740            MRRETURN(MATCH_NOMATCH);
2741            }
2742          if (ecode[1] != *eptr++) MRRETURN(MATCH_NOMATCH);
2743        ecode += 2;        ecode += 2;
2744        }        }
2745      break;      break;
# Line 2215  for (;;) Line 2754  for (;;)
2754        ecode++;        ecode++;
2755        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
2756    
2757        if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);        if (length > md->end_subject - eptr)
2758            {
2759            CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
2760            MRRETURN(MATCH_NOMATCH);
2761            }
2762    
2763        /* 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
2764        can use the fast lookup table. */        can use the fast lookup table. */
2765    
2766        if (fc < 128)        if (fc < 128)
2767          {          {
2768          if (md->lcc[*ecode++] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);          if (md->lcc[*ecode++] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);
2769          }          }
2770    
2771        /* Otherwise we must pick up the subject character */        /* Otherwise we must pick up the subject character */
# Line 2241  for (;;) Line 2784  for (;;)
2784  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2785            if (dc != UCD_OTHERCASE(fc))            if (dc != UCD_OTHERCASE(fc))
2786  #endif  #endif
2787              RRETURN(MATCH_NOMATCH);              MRRETURN(MATCH_NOMATCH);
2788            }            }
2789          }          }
2790        }        }
# Line 2250  for (;;) Line 2793  for (;;)
2793    
2794      /* Non-UTF-8 mode */      /* Non-UTF-8 mode */
2795        {        {
2796        if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH);        if (md->end_subject - eptr < 1)
2797        if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);          {
2798            SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
2799            MRRETURN(MATCH_NOMATCH);
2800            }
2801          if (md->lcc[ecode[1]] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);
2802        ecode += 2;        ecode += 2;
2803        }        }
2804      break;      break;
# Line 2304  for (;;) Line 2851  for (;;)
2851      case OP_MINQUERY:      case OP_MINQUERY:
2852      c = *ecode++ - OP_STAR;      c = *ecode++ - OP_STAR;
2853      minimize = (c & 1) != 0;      minimize = (c & 1) != 0;
2854    
2855      min = rep_min[c];                 /* Pick up values from tables; */      min = rep_min[c];                 /* Pick up values from tables; */
2856      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
2857      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
2858    
2859      /* Common code for all repeated single-character matches. We can give      /* Common code for all repeated single-character matches. */
     up quickly if there are fewer than the minimum number of characters left in  
     the subject. */  
2860    
2861      REPEATCHAR:      REPEATCHAR:
2862  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 2319  for (;;) Line 2865  for (;;)
2865        length = 1;        length = 1;
2866        charptr = ecode;        charptr = ecode;
2867        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
       if (min * length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
2868        ecode += length;        ecode += length;
2869    
2870        /* Handle multibyte character matching specially here. There is        /* Handle multibyte character matching specially here. There is
# Line 2337  for (;;) Line 2882  for (;;)
2882    
2883          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2884            {            {
2885            if (memcmp(eptr, charptr, length) == 0) eptr += length;            if (eptr <= md->end_subject - length &&
2886                memcmp(eptr, charptr, length) == 0) eptr += length;
2887  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2888            /* Need braces because of following else */            else if (oclength > 0 &&
2889            else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                     eptr <= md->end_subject - oclength &&
2890                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2891    #endif  /* SUPPORT_UCP */
2892            else            else
2893              {              {
2894              if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);              CHECK_PARTIAL();
2895              eptr += oclength;              MRRETURN(MATCH_NOMATCH);
2896              }              }
 #else   /* without SUPPORT_UCP */  
           else { RRETURN(MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2897            }            }
2898    
2899          if (min == max) continue;          if (min == max) continue;
# Line 2359  for (;;) Line 2904  for (;;)
2904              {              {
2905              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);
2906              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2907              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
2908              if (memcmp(eptr, charptr, length) == 0) eptr += length;              if (eptr <= md->end_subject - length &&
2909                  memcmp(eptr, charptr, length) == 0) eptr += length;
2910  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2911              /* Need braces because of following else */              else if (oclength > 0 &&
2912              else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                       eptr <= md->end_subject - oclength &&
2913                         memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2914    #endif  /* SUPPORT_UCP */
2915              else              else
2916                {                {
2917                if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);                CHECK_PARTIAL();
2918                eptr += oclength;                MRRETURN(MATCH_NOMATCH);
2919                }                }
 #else   /* without SUPPORT_UCP */  
             else { RRETURN (MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2920              }              }
2921            /* Control never gets here */            /* Control never gets here */
2922            }            }
# Line 2381  for (;;) Line 2926  for (;;)
2926            pp = eptr;            pp = eptr;
2927            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2928              {              {
2929              if (eptr > md->end_subject - length) break;              if (eptr <= md->end_subject - length &&
2930              if (memcmp(eptr, charptr, length) == 0) eptr += length;                  memcmp(eptr, charptr, length) == 0) eptr += length;
2931  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2932              else if (oclength == 0) break;              else if (oclength > 0 &&
2933                         eptr <= md->end_subject - oclength &&
2934                         memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2935    #endif  /* SUPPORT_UCP */
2936              else              else
2937                {                {
2938                if (memcmp(eptr, occhars, oclength) != 0) break;                CHECK_PARTIAL();
2939                eptr += oclength;                break;
2940                }                }
 #else   /* without SUPPORT_UCP */  
             else break;  
 #endif  /* SUPPORT_UCP */  
2941              }              }
2942    
2943            if (possessive) continue;            if (possessive) continue;
2944    
2945            for(;;)            for(;;)
2946             {              {
2947             RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);
2948             if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2949             if (eptr == pp) RRETURN(MATCH_NOMATCH);              if (eptr == pp) { MRRETURN(MATCH_NOMATCH); }
2950  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2951             eptr--;              eptr--;
2952             BACKCHAR(eptr);              BACKCHAR(eptr);
2953  #else   /* without SUPPORT_UCP */  #else   /* without SUPPORT_UCP */
2954             eptr -= length;              eptr -= length;
2955  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2956             }              }
2957            }            }
2958          /* Control never gets here */          /* Control never gets here */
2959          }          }
# Line 2420  for (;;) Line 2966  for (;;)
2966  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF8 */
2967    
2968      /* When not in UTF-8 mode, load a single-byte character. */      /* When not in UTF-8 mode, load a single-byte character. */
2969        {  
2970        if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);      fc = *ecode++;
       fc = *ecode++;  
       }  
2971    
2972      /* The value of fc at this point is always less than 256, though we may or      /* The value of fc at this point is always less than 256, though we may or
2973      may not be in UTF-8 mode. The code is duplicated for the caseless and      may not be in UTF-8 mode. The code is duplicated for the caseless and
# Line 2441  for (;;) Line 2985  for (;;)
2985        {        {
2986        fc = md->lcc[fc];        fc = md->lcc[fc];
2987        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2988          if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);          {
2989            if (eptr >= md->end_subject)
2990              {
2991              SCHECK_PARTIAL();
2992              MRRETURN(MATCH_NOMATCH);
2993              }
2994            if (fc != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);
2995            }
2996        if (min == max) continue;        if (min == max) continue;
2997        if (minimize)        if (minimize)
2998          {          {
# Line 2449  for (;;) Line 3000  for (;;)
3000            {            {
3001            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);
3002            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3003            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) MRRETURN(MATCH_NOMATCH);
3004                fc != md->lcc[*eptr++])            if (eptr >= md->end_subject)
3005              RRETURN(MATCH_NOMATCH);              {
3006                SCHECK_PARTIAL();
3007                MRRETURN(MATCH_NOMATCH);
3008                }
3009              if (fc != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);
3010            }            }
3011          /* Control never gets here */          /* Control never gets here */
3012          }          }
# Line 2460  for (;;) Line 3015  for (;;)
3015          pp = eptr;          pp = eptr;
3016          for (i = min; i < max; i++)          for (i = min; i < max; i++)
3017            {            {
3018            if (eptr >= md->end_subject || fc != md->lcc[*eptr]) break;            if (eptr >= md->end_subject)
3019                {
3020                SCHECK_PARTIAL();
3021                break;
3022                }
3023              if (fc != md->lcc[*eptr]) break;
3024            eptr++;            eptr++;
3025            }            }
3026    
3027          if (possessive) continue;          if (possessive) continue;
3028    
3029          while (eptr >= pp)          while (eptr >= pp)
3030            {            {
3031            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);
3032            eptr--;            eptr--;
3033            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3034            }            }
3035          RRETURN(MATCH_NOMATCH);          MRRETURN(MATCH_NOMATCH);
3036          }          }
3037        /* Control never gets here */        /* Control never gets here */
3038        }        }
# Line 2479  for (;;) Line 3041  for (;;)
3041    
3042      else      else
3043        {        {
3044        for (i = 1; i <= min; i++) if (fc != *eptr++) RRETURN(MATCH_NOMATCH);        for (i = 1; i <= min; i++)
3045            {
3046            if (eptr >= md->end_subject)
3047              {
3048              SCHECK_PARTIAL();
3049              MRRETURN(MATCH_NOMATCH);
3050              }
3051            if (fc != *eptr++) MRRETURN(MATCH_NOMATCH);
3052            }
3053    
3054        if (min == max) continue;        if (min == max) continue;
3055    
3056        if (minimize)        if (minimize)
3057          {          {
3058          for (fi = min;; fi++)          for (fi = min;; fi++)
3059            {            {
3060            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);
3061            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3062            if (fi >= max || eptr >= md->end_subject || fc != *eptr++)            if (fi >= max) MRRETURN(MATCH_NOMATCH);
3063              RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3064                {
3065                SCHECK_PARTIAL();
3066                MRRETURN(MATCH_NOMATCH);
3067                }
3068              if (fc != *eptr++) MRRETURN(MATCH_NOMATCH);
3069            }            }
3070          /* Control never gets here */          /* Control never gets here */
3071          }          }
# Line 2497  for (;;) Line 3074  for (;;)
3074          pp = eptr;          pp = eptr;
3075          for (i = min; i < max; i++)          for (i = min; i < max; i++)
3076            {            {
3077            if (eptr >= md->end_subject || fc != *eptr) break;            if (eptr >= md->end_subject)
3078                {
3079                SCHECK_PARTIAL();
3080                break;
3081                }
3082              if (fc != *eptr) break;
3083            eptr++;            eptr++;
3084            }            }
3085          if (possessive) continue;          if (possessive) continue;
3086    
3087          while (eptr >= pp)          while (eptr >= pp)
3088            {            {
3089            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27);
3090            eptr--;            eptr--;
3091            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3092            }            }
3093          RRETURN(MATCH_NOMATCH);          MRRETURN(MATCH_NOMATCH);
3094          }          }
3095        }        }
3096      /* Control never gets here */      /* Control never gets here */
# Line 2516  for (;;) Line 3099  for (;;)
3099      checking can be multibyte. */      checking can be multibyte. */
3100    
3101      case OP_NOT:      case OP_NOT:
3102      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
3103          {
3104          SCHECK_PARTIAL();
3105          MRRETURN(MATCH_NOMATCH);
3106          }
3107      ecode++;      ecode++;
3108      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
3109      if ((ims & PCRE_CASELESS) != 0)      if ((ims & PCRE_CASELESS) != 0)
# Line 2525  for (;;) Line 3112  for (;;)
3112        if (c < 256)        if (c < 256)
3113  #endif  #endif
3114        c = md->lcc[c];        c = md->lcc[c];
3115        if (md->lcc[*ecode++] == c) RRETURN(MATCH_NOMATCH);        if (md->lcc[*ecode++] == c) MRRETURN(MATCH_NOMATCH);
3116        }        }
3117      else      else
3118        {        {
3119        if (*ecode++ == c) RRETURN(MATCH_NOMATCH);        if (*ecode++ == c) MRRETURN(MATCH_NOMATCH);
3120        }        }
3121      break;      break;
3122    
# Line 2593  for (;;) Line 3180  for (;;)
3180      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
3181      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
3182    
3183      /* Common code for all repeated single-byte matches. We can give up quickly      /* Common code for all repeated single-byte matches. */
     if there are fewer than the minimum number of bytes left in the  
     subject. */  
3184    
3185      REPEATNOTCHAR:      REPEATNOTCHAR:
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
3186      fc = *ecode++;      fc = *ecode++;
3187    
3188      /* The code is duplicated for the caseless and caseful cases, for speed,      /* The code is duplicated for the caseless and caseful cases, for speed,
# Line 2623  for (;;) Line 3207  for (;;)
3207          register unsigned int d;          register unsigned int d;
3208          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3209            {            {
3210              if (eptr >= md->end_subject)
3211                {
3212                SCHECK_PARTIAL();
3213                MRRETURN(MATCH_NOMATCH);
3214                }
3215            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3216            if (d < 256) d = md->lcc[d];            if (d < 256) d = md->lcc[d];
3217            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) MRRETURN(MATCH_NOMATCH);
3218            }            }
3219          }          }
3220        else        else
# Line 2634  for (;;) Line 3223  for (;;)
3223        /* Not UTF-8 mode */        /* Not UTF-8 mode */
3224          {          {
3225          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3226            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);            {
3227              if (eptr >= md->end_subject)
3228                {
3229                SCHECK_PARTIAL();
3230                MRRETURN(MATCH_NOMATCH);
3231                }
3232              if (fc == md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);
3233              }
3234          }          }
3235    
3236        if (min == max) continue;        if (min == max) continue;
# Line 2650  for (;;) Line 3246  for (;;)
3246              {              {
3247              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);
3248              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3249              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
3250                if (eptr >= md->end_subject)
3251                  {
3252                  SCHECK_PARTIAL();
3253                  MRRETURN(MATCH_NOMATCH);
3254                  }
3255              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3256              if (d < 256) d = md->lcc[d];              if (d < 256) d = md->lcc[d];
3257              if (fc == d) RRETURN(MATCH_NOMATCH);              if (fc == d) MRRETURN(MATCH_NOMATCH);
   
3258              }              }
3259            }            }
3260          else          else
# Line 2665  for (;;) Line 3265  for (;;)
3265              {              {
3266              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);
3267              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3268              if (fi >= max || eptr >= md->end_subject || fc == md->lcc[*eptr++])              if (fi >= max) MRRETURN(MATCH_NOMATCH);
3269                RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3270                  {
3271                  SCHECK_PARTIAL();
3272                  MRRETURN(MATCH_NOMATCH);
3273                  }
3274                if (fc == md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);
3275              }              }
3276            }            }
3277          /* Control never gets here */          /* Control never gets here */
# Line 2686  for (;;) Line 3291  for (;;)
3291            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3292              {              {
3293              int len = 1;              int len = 1;
3294              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
3295                  {
3296                  SCHECK_PARTIAL();
3297                  break;
3298                  }
3299              GETCHARLEN(d, eptr, len);              GETCHARLEN(d, eptr, len);
3300              if (d < 256) d = md->lcc[d];              if (d < 256) d = md->lcc[d];
3301              if (fc == d) break;              if (fc == d) break;
# Line 2707  for (;;) Line 3316  for (;;)
3316            {            {
3317            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3318              {              {
3319              if (eptr >= md->end_subject || fc == md->lcc[*eptr]) break;              if (eptr >= md->end_subject)
3320                  {
3321                  SCHECK_PARTIAL();
3322                  break;
3323                  }
3324                if (fc == md->lcc[*eptr]) break;
3325              eptr++;              eptr++;
3326              }              }
3327            if (possessive) continue;            if (possessive) continue;
# Line 2719  for (;;) Line 3333  for (;;)
3333              }              }
3334            }            }
3335    
3336          RRETURN(MATCH_NOMATCH);          MRRETURN(MATCH_NOMATCH);
3337          }          }
3338        /* Control never gets here */        /* Control never gets here */
3339        }        }
# Line 2735  for (;;) Line 3349  for (;;)
3349          register unsigned int d;          register unsigned int d;
3350          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3351            {            {
3352              if (eptr >= md->end_subject)
3353                {
3354                SCHECK_PARTIAL();
3355                MRRETURN(MATCH_NOMATCH);
3356                }
3357            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3358            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) MRRETURN(MATCH_NOMATCH);
3359            }            }
3360          }          }
3361        else        else
# Line 2744  for (;;) Line 3363  for (;;)
3363        /* Not UTF-8 mode */        /* Not UTF-8 mode */
3364          {          {
3365          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3366            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);            {
3367              if (eptr >= md->end_subject)
3368                {
3369                SCHECK_PARTIAL();
3370                MRRETURN(MATCH_NOMATCH);
3371                }
3372              if (fc == *eptr++) MRRETURN(MATCH_NOMATCH);
3373              }
3374          }          }
3375    
3376        if (min == max) continue;        if (min == max) continue;
# Line 2760  for (;;) Line 3386  for (;;)
3386              {              {
3387              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);
3388              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3389              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
3390                if (eptr >= md->end_subject)
3391                  {
3392                  SCHECK_PARTIAL();
3393                  MRRETURN(MATCH_NOMATCH);
3394                  }
3395              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3396              if (fc == d) RRETURN(MATCH_NOMATCH);              if (fc == d) MRRETURN(MATCH_NOMATCH);
3397              }              }
3398            }            }
3399          else          else
# Line 2773  for (;;) Line 3404  for (;;)
3404              {              {
3405              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);
3406              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3407              if (fi >= max || eptr >= md->end_subject || fc == *eptr++)              if (fi >= max) MRRETURN(MATCH_NOMATCH);
3408                RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3409                  {
3410                  SCHECK_PARTIAL();
3411                  MRRETURN(MATCH_NOMATCH);
3412                  }
3413                if (fc == *eptr++) MRRETURN(MATCH_NOMATCH);
3414              }              }
3415            }            }
3416          /* Control never gets here */          /* Control never gets here */
# Line 2794  for (;;) Line 3430  for (;;)
3430            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3431              {              {
3432              int len = 1;              int len = 1;
3433              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
3434                  {
3435                  SCHECK_PARTIAL();
3436                  break;
3437                  }
3438              GETCHARLEN(d, eptr, len);              GETCHARLEN(d, eptr, len);
3439              if (fc == d) break;              if (fc == d) break;
3440              eptr += len;              eptr += len;
# Line 2814  for (;;) Line 3454  for (;;)
3454            {            {
3455            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3456              {              {
3457              if (eptr >= md->end_subject || fc == *eptr) break;              if (eptr >= md->end_subject)
3458                  {
3459                  SCHECK_PARTIAL();
3460                  break;
3461                  }
3462                if (fc == *eptr) break;
3463              eptr++;              eptr++;
3464              }              }
3465            if (possessive) continue;            if (possessive) continue;
# Line 2826  for (;;) Line 3471  for (;;)
3471              }              }
3472            }            }
3473    
3474          RRETURN(MATCH_NOMATCH);          MRRETURN(MATCH_NOMATCH);
3475          }          }
3476        }        }
3477      /* Control never gets here */      /* Control never gets here */
# Line 2908  for (;;) Line 3553  for (;;)
3553    
3554      /* First, ensure the minimum number of matches are present. Use inline      /* First, ensure the minimum number of matches are present. Use inline
3555      code for maximizing the speed, and do the type test once at the start      code for maximizing the speed, and do the type test once at the start
3556      (i.e. keep it out of the loop). Also we can test that there are at least      (i.e. keep it out of the loop). Separate the UTF-8 code completely as that
     the minimum number of bytes before we start. This isn't as effective in  
     UTF-8 mode, but it does no harm. Separate the UTF-8 code completely as that  
3557      is tidier. Also separate the UCP code, which can be the same for both UTF-8      is tidier. Also separate the UCP code, which can be the same for both UTF-8
3558      and single-bytes. */      and single-bytes. */
3559    
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
3560      if (min > 0)      if (min > 0)
3561        {        {
3562  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 2923  for (;;) Line 3565  for (;;)
3565          switch(prop_type)          switch(prop_type)
3566            {            {
3567            case PT_ANY:            case PT_ANY:
3568            if (prop_fail_result) RRETURN(MATCH_NOMATCH);            if (prop_fail_result) MRRETURN(MATCH_NOMATCH);
3569            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3570              {              {
3571              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3572                  {
3573                  SCHECK_PARTIAL();
3574                  MRRETURN(MATCH_NOMATCH);
3575                  }
3576              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3577              }              }
3578            break;            break;
# Line 2934  for (;;) Line 3580  for (;;)
3580            case PT_LAMP:            case PT_LAMP:
3581            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3582              {              {
3583              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3584                  {
3585                  SCHECK_PARTIAL();
3586                  MRRETURN(MATCH_NOMATCH);
3587                  }
3588              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3589              prop_chartype = UCD_CHARTYPE(c);              prop_chartype = UCD_CHARTYPE(c);
3590              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
3591                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
3592                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
3593                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
3594              }              }
3595            break;            break;
3596    
3597            case PT_GC:            case PT_GC:
3598            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3599              {              {
3600              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3601                  {
3602                  SCHECK_PARTIAL();
3603                  MRRETURN(MATCH_NOMATCH);
3604                  }
3605              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3606              prop_category = UCD_CATEGORY(c);              prop_category = UCD_CATEGORY(c);
3607              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
3608                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
3609              }              }
3610            break;            break;
3611    
3612            case PT_PC:            case PT_PC:
3613            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3614              {              {
3615              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3616                  {
3617                  SCHECK_PARTIAL();
3618                  MRRETURN(MATCH_NOMATCH);
3619                  }
3620              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3621              prop_chartype = UCD_CHARTYPE(c);              prop_chartype = UCD_CHARTYPE(c);
3622              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
3623                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
3624              }              }
3625            break;            break;
3626    
3627            case PT_SC:            case PT_SC:
3628            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3629              {              {
3630              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3631                  {
3632                  SCHECK_PARTIAL();
3633                  MRRETURN(MATCH_NOMATCH);
3634                  }
3635              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3636              prop_script = UCD_SCRIPT(c);              prop_script = UCD_SCRIPT(c);
3637              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
3638                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
3639                }
3640              break;
3641    
3642              case PT_ALNUM:
3643              for (i = 1; i <= min; i++)
3644                {
3645                if (eptr >= md->end_subject)
3646                  {
3647                  SCHECK_PARTIAL();
3648                  MRRETURN(MATCH_NOMATCH);
3649                  }
3650                GETCHARINCTEST(c, eptr);
3651                prop_category = UCD_CATEGORY(c);
3652                if ((prop_category == ucp_L || prop_category == ucp_N)
3653                       == prop_fail_result)
3654                  MRRETURN(MATCH_NOMATCH);
3655                }
3656              break;
3657    
3658              case PT_SPACE:    /* Perl space */
3659              for (i = 1; i <= min; i++)
3660                {
3661                if (eptr >= md->end_subject)
3662                  {
3663                  SCHECK_PARTIAL();
3664                  MRRETURN(MATCH_NOMATCH);
3665                  }
3666                GETCHARINCTEST(c, eptr);
3667                prop_category = UCD_CATEGORY(c);
3668                if ((prop_category == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
3669                     c == CHAR_FF || c == CHAR_CR)
3670                       == prop_fail_result)
3671                  MRRETURN(MATCH_NOMATCH);
3672                }
3673              break;
3674    
3675              case PT_PXSPACE:  /* POSIX space */
3676              for (i = 1; i <= min; i++)
3677                {
3678                if (eptr >= md->end_subject)
3679                  {
3680                  SCHECK_PARTIAL();
3681                  MRRETURN(MATCH_NOMATCH);
3682                  }
3683                GETCHARINCTEST(c, eptr);
3684                prop_category = UCD_CATEGORY(c);
3685                if ((prop_category == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
3686                     c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)
3687                       == prop_fail_result)
3688                  MRRETURN(MATCH_NOMATCH);
3689              }              }
3690            break;            break;
3691    
3692              case PT_WORD:
3693              for (i = 1; i <= min; i++)
3694                {
3695                if (eptr >= md->end_subject)
3696                  {
3697                  SCHECK_PARTIAL();
3698                  MRRETURN(MATCH_NOMATCH);
3699                  }
3700                GETCHARINCTEST(c, eptr);
3701                prop_category = UCD_CATEGORY(c);
3702                if ((prop_category == ucp_L || prop_category == ucp_N ||
3703                     c == CHAR_UNDERSCORE)
3704                       == prop_fail_result)
3705                  MRRETURN(MATCH_NOMATCH);
3706                }
3707              break;
3708    
3709              /* This should not occur */
3710    
3711            default:            default:
3712            RRETURN(PCRE_ERROR_INTERNAL);            RRETURN(PCRE_ERROR_INTERNAL);
3713            }            }
# Line 2989  for (;;) Line 3720  for (;;)
3720          {          {
3721          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3722            {            {
3723              if (eptr >= md->end_subject)
3724                {
3725                SCHECK_PARTIAL();
3726                MRRETURN(MATCH_NOMATCH);
3727                }
3728            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3729            prop_category = UCD_CATEGORY(c);            prop_category = UCD_CATEGORY(c);
3730            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);            if (prop_category == ucp_M) MRRETURN(MATCH_NOMATCH);
3731            while (eptr < md->end_subject)            while (eptr < md->end_subject)
3732              {              {
3733              int len = 1;              int len = 1;
3734              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
3735                {                else { GETCHARLEN(c, eptr, len); }
               GETCHARLEN(c, eptr, len);  
               }  
3736              prop_category = UCD_CATEGORY(c);              prop_category = UCD_CATEGORY(c);
3737              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
3738              eptr += len;              eptr += len;
# Line 3017  for (;;) Line 3751  for (;;)
3751          case OP_ANY:          case OP_ANY:
3752          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3753            {            {
3754            if (eptr >= md->end_subject || IS_NEWLINE(eptr))            if (eptr >= md->end_subject)
3755              RRETURN(MATCH_NOMATCH);              {
3756                SCHECK_PARTIAL();
3757                MRRETURN(MATCH_NOMATCH);
3758                }
3759              if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);
3760            eptr++;            eptr++;
3761            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3762            }            }
# Line 3027  for (;;) Line 3765  for (;;)
3765          case OP_ALLANY:          case OP_ALLANY:
3766          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3767            {            {
3768            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3769                {
3770                SCHECK_PARTIAL();
3771                MRRETURN(MATCH_NOMATCH);
3772                }
3773            eptr++;            eptr++;
3774            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3775            }            }
3776          break;          break;
3777    
3778          case OP_ANYBYTE:          case OP_ANYBYTE:
3779            if (eptr > md->end_subject - min) MRRETURN(MATCH_NOMATCH);
3780          eptr += min;          eptr += min;
3781          break;          break;
3782    
3783          case OP_ANYNL:          case OP_ANYNL:
3784          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3785            {            {
3786            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3787                {
3788                SCHECK_PARTIAL();
3789                MRRETURN(MATCH_NOMATCH);
3790                }
3791            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3792            switch(c)            switch(c)
3793              {              {
3794              default: RRETURN(MATCH_NOMATCH);              default: MRRETURN(MATCH_NOMATCH);
3795    
3796              case 0x000d:              case 0x000d:
3797              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3798              break;              break;
# Line 3057  for (;;) Line 3805  for (;;)
3805              case 0x0085:              case 0x0085:
3806              case 0x2028:              case 0x2028:
3807              case 0x2029:              case 0x2029:
3808              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);              if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);
3809              break;              break;
3810              }              }
3811            }            }
# Line 3066  for (;;) Line 3814  for (;;)
3814          case OP_NOT_HSPACE:          case OP_NOT_HSPACE:
3815          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3816            {            {
3817            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3818                {
3819                SCHECK_PARTIAL();
3820                MRRETURN(MATCH_NOMATCH);
3821                }
3822            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3823            switch(c)            switch(c)
3824              {              {
# Line 3090  for (;;) Line 3842  for (;;)
3842              case 0x202f:    /* NARROW NO-BREAK SPACE */              case 0x202f:    /* NARROW NO-BREAK SPACE */
3843              case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */              case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3844              case 0x3000:    /* IDEOGRAPHIC SPACE */              case 0x3000:    /* IDEOGRAPHIC SPACE */
3845              RRETURN(MATCH_NOMATCH);              MRRETURN(MATCH_NOMATCH);
3846              }              }
3847            }            }
3848          break;          break;
# Line 3098  for (;;) Line 3850  for (;;)
3850          case OP_HSPACE:          case OP_HSPACE:
3851          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3852            {            {
3853            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3854                {
3855                SCHECK_PARTIAL();
3856                MRRETURN(MATCH_NOMATCH);
3857                }
3858            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3859            switch(c)            switch(c)
3860              {              {
3861              default: RRETURN(MATCH_NOMATCH);              default: MRRETURN(MATCH_NOMATCH);
3862              case 0x09:      /* HT */              case 0x09:      /* HT */
3863              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
3864              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
# Line 3130  for (;;) Line 3886  for (;;)
3886          case OP_NOT_VSPACE:          case OP_NOT_VSPACE:
3887          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3888            {            {
3889            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3890                {
3891                SCHECK_PARTIAL();
3892                MRRETURN(MATCH_NOMATCH);
3893                }
3894            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3895            switch(c)            switch(c)
3896              {              {
# Line 3142  for (;;) Line 3902  for (;;)
3902              case 0x85:      /* NEL */              case 0x85:      /* NEL */
3903              case 0x2028:    /* LINE SEPARATOR */              case 0x2028:    /* LINE SEPARATOR */
3904              case 0x2029:    /* PARAGRAPH SEPARATOR */              case 0x2029:    /* PARAGRAPH SEPARATOR */
3905              RRETURN(MATCH_NOMATCH);              MRRETURN(MATCH_NOMATCH);
3906              }              }
3907            }            }
3908          break;          break;
# Line 3150  for (;;) Line 3910  for (;;)
3910          case OP_VSPACE:          case OP_VSPACE:
3911          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3912            {            {
3913            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3914                {
3915                SCHECK_PARTIAL();
3916                MRRETURN(MATCH_NOMATCH);
3917                }
3918            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3919            switch(c)            switch(c)
3920              {              {
3921              default: RRETURN(MATCH_NOMATCH);              default: MRRETURN(MATCH_NOMATCH);
3922              case 0x0a:      /* LF */              case 0x0a:      /* LF */
3923              case 0x0b:      /* VT */              case 0x0b:      /* VT */
3924              case 0x0c:      /* FF */              case 0x0c:      /* FF */
# Line 3170  for (;;) Line 3934  for (;;)
3934          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
3935          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3936            {            {
3937            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3938                {
3939                SCHECK_PARTIAL();
3940                MRRETURN(MATCH_NOMATCH);
3941                }
3942            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3943            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)
3944              RRETURN(MATCH_NOMATCH);              MRRETURN(MATCH_NOMATCH);
3945            }            }
3946          break;          break;
3947    
3948          case OP_DIGIT:          case OP_DIGIT:
3949          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3950            {            {
3951            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3952               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)              {
3953              RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
3954                MRRETURN(MATCH_NOMATCH);
3955                }
3956              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)
3957                MRRETURN(MATCH_NOMATCH);
3958            /* 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 */
3959            }            }
3960          break;          break;
# Line 3190  for (;;) Line 3962  for (;;)
3962          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
3963          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3964            {            {
3965            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3966               (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0))              {
3967              RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
3968                MRRETURN(MATCH_NOMATCH);
3969                }
3970              if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)
3971                MRRETURN(MATCH_NOMATCH);
3972            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3973            }            }
3974          break;          break;
# Line 3200  for (;;) Line 3976  for (;;)
3976          case OP_WHITESPACE:          case OP_WHITESPACE:
3977          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3978            {            {
3979            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3980               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)              {
3981              RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
3982                MRRETURN(MATCH_NOMATCH);
3983                }
3984              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)
3985                MRRETURN(MATCH_NOMATCH);
3986            /* 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 */
3987            }            }
3988          break;          break;
# Line 3210  for (;;) Line 3990  for (;;)
3990          case OP_NOT_WORDCHAR:          case OP_NOT_WORDCHAR:
3991          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3992            {            {
3993            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3994               (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0))              {
3995              RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
3996                MRRETURN(MATCH_NOMATCH);
3997                }
3998              if (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0)
3999                MRRETURN(MATCH_NOMATCH);
4000            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
4001            }            }
4002          break;          break;
# Line 3220  for (;;) Line 4004  for (;;)
4004          case OP_WORDCHAR:          case OP_WORDCHAR:
4005          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4006            {            {
4007            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
4008               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)              {
4009              RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
4010                MRRETURN(MATCH_NOMATCH);
4011                }
4012              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)
4013                MRRETURN(MATCH_NOMATCH);
4014            /* 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 */
4015            }            }
4016          break;          break;
# Line 3235  for (;;) Line 4023  for (;;)
4023  #endif     /* SUPPORT_UTF8 */  #endif     /* SUPPORT_UTF8 */
4024    
4025        /* 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
4026        than OP_PROP and OP_NOTPROP. We can assume that there are the minimum        than OP_PROP and OP_NOTPROP. */
       number of bytes present, as this was tested above. */  
4027    
4028        switch(ctype)        switch(ctype)
4029          {          {
4030          case OP_ANY:          case OP_ANY:
4031          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4032            {            {
4033            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
4034                {
4035                SCHECK_PARTIAL();
4036                MRRETURN(MATCH_NOMATCH);
4037                }
4038              if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);
4039            eptr++;            eptr++;
4040            }            }
4041          break;          break;
4042    
4043          case OP_ALLANY:          case OP_ALLANY:
4044            if (eptr > md->end_subject - min)
4045              {
4046              SCHECK_PARTIAL();
4047              MRRETURN(MATCH_NOMATCH);
4048              }
4049          eptr += min;          eptr += min;
4050          break;          break;
4051    
4052          case OP_ANYBYTE:          case OP_ANYBYTE:
4053            if (eptr > md->end_subject - min)
4054              {
4055              SCHECK_PARTIAL();
4056              MRRETURN(MATCH_NOMATCH);
4057              }
4058          eptr += min;          eptr += min;
4059          break;          break;
4060    
         /* Because of the CRLF case, we can't assume the minimum number of  
         bytes are present in this case. */  
   
4061          case OP_ANYNL:          case OP_ANYNL:
4062          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4063            {            {
4064            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
4065                {
4066                SCHECK_PARTIAL();
4067                MRRETURN(MATCH_NOMATCH);
4068                }
4069            switch(*eptr++)            switch(*eptr++)
4070              {              {
4071              default: RRETURN(MATCH_NOMATCH);              default: MRRETURN(MATCH_NOMATCH);
4072    
4073              case 0x000d:              case 0x000d:
4074              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
4075              break;              break;
4076    
4077              case 0x000a:              case 0x000a:
4078              break;              break;
4079    
4080              case 0x000b:              case 0x000b:
4081              case 0x000c:              case 0x000c:
4082              case 0x0085:              case 0x0085:
4083              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);              if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);
4084              break;              break;
4085              }              }
4086            }            }
# Line 3284  for (;;) Line 4089  for (;;)
4089          case OP_NOT_HSPACE:          case OP_NOT_HSPACE:
4090          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4091            {            {
4092            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
4093                {
4094                SCHECK_PARTIAL();
4095                MRRETURN(MATCH_NOMATCH);
4096                }
4097            switch(*eptr++)            switch(*eptr++)
4098              {              {
4099              default: break;              default: break;
4100              case 0x09:      /* HT */              case 0x09:      /* HT */
4101              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4102              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
4103              RRETURN(MATCH_NOMATCH);              MRRETURN(MATCH_NOMATCH);
4104              }              }
4105            }            }
4106          break;          break;
# Line 3299  for (;;) Line 4108  for (;;)
4108          case OP_HSPACE:          case OP_HSPACE:
4109          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4110            {            {
4111            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
4112                {
4113                SCHECK_PARTIAL();
4114                MRRETURN(MATCH_NOMATCH);
4115                }
4116            switch(*eptr++)            switch(*eptr++)
4117              {              {
4118              default: RRETURN(MATCH_NOMATCH);              default: MRRETURN(MATCH_NOMATCH);
4119              case 0x09:      /* HT */              case 0x09:      /* HT */
4120              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
4121              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
# Line 3314  for (;;) Line 4127  for (;;)
4127          case OP_NOT_VSPACE:          case OP_NOT_VSPACE:
4128          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4129            {            {
4130            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
4131                {
4132                SCHECK_PARTIAL();
4133                MRRETURN(MATCH_NOMATCH);
4134                }
4135            switch(*eptr++)            switch(*eptr++)
4136              {              {
4137              default: break;              default: break;
# Line 3323  for (;;) Line 4140  for (;;)
4140              case 0x0c:      /* FF */              case 0x0c:      /* FF */
4141              case 0x0d:      /* CR */              case 0x0d:      /* CR */
4142              case 0x85:      /* NEL */              case 0x85:      /* NEL */
4143              RRETURN(MATCH_NOMATCH);              MRRETURN(MATCH_NOMATCH);
4144              }              }
4145            }            }
4146          break;          break;
# Line 3331  for (;;) Line 4148  for (;;)
4148          case OP_VSPACE:          case OP_VSPACE:
4149          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4150            {            {
4151            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
4152                {
4153                SCHECK_PARTIAL();
4154                MRRETURN(MATCH_NOMATCH);
4155                }
4156            switch(*eptr++)            switch(*eptr++)
4157              {              {
4158              default: RRETURN(MATCH_NOMATCH);              default: MRRETURN(MATCH_NOMATCH);
4159              case 0x0a:      /* LF */              case 0x0a:      /* LF */
4160              case 0x0b:      /* VT */              case 0x0b:      /* VT */
4161              case 0x0c:      /* FF */              case 0x0c:      /* FF */
# Line 3347  for (;;) Line 4168  for (;;)
4168    
4169          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
4170          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4171            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);            {
4172              if (eptr >= md->end_subject)
4173                {
4174                SCHECK_PARTIAL();
4175                MRRETURN(MATCH_NOMATCH);
4176                }
4177              if ((md->ctypes[*eptr++] & ctype_digit) != 0) MRRETURN(MATCH_NOMATCH);
4178              }
4179          break;          break;
4180    
4181          case OP_DIGIT:          case OP_DIGIT:
4182          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4183            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);            {
4184              if (eptr >= md->end_subject)
4185                {
4186                SCHECK_PARTIAL();
4187                MRRETURN(MATCH_NOMATCH);
4188                }
4189              if ((md->ctypes[*eptr++] & ctype_digit) == 0) MRRETURN(MATCH_NOMATCH);
4190              }
4191          break;          break;
4192    
4193          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
4194          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4195            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);            {
4196              if (eptr >= md->end_subject)
4197                {
4198                SCHECK_PARTIAL();
4199                MRRETURN(MATCH_NOMATCH);
4200                }
4201              if ((md->ctypes[*eptr++] & ctype_space) != 0) MRRETURN(MATCH_NOMATCH);
4202              }
4203          break;          break;
4204    
4205          case OP_WHITESPACE:          case OP_WHITESPACE:
4206          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4207            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);            {
4208              if (eptr >= md->end_subject)
4209                {
4210                SCHECK_PARTIAL();
4211                MRRETURN(MATCH_NOMATCH);
4212                }
4213              if ((md->ctypes[*eptr++] & ctype_space) == 0) MRRETURN(MATCH_NOMATCH);
4214              }
4215          break;          break;
4216    
4217          case OP_NOT_WORDCHAR:          case OP_NOT_WORDCHAR:
4218          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4219              {
4220              if (eptr >= md->end_subject)
4221                {
4222                SCHECK_PARTIAL();
4223                MRRETURN(MATCH_NOMATCH);
4224                }
4225            if ((md->ctypes[*eptr++] & ctype_word) != 0)            if ((md->ctypes[*eptr++] & ctype_word) != 0)
4226              RRETURN(MATCH_NOMATCH);              MRRETURN(MATCH_NOMATCH);
4227              }
4228          break;          break;
4229    
4230          case OP_WORDCHAR:          case OP_WORDCHAR:
4231          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4232              {
4233              if (eptr >= md->end_subject)
4234                {
4235                SCHECK_PARTIAL();
4236                MRRETURN(MATCH_NOMATCH);
4237                }
4238            if ((md->ctypes[*eptr++] & ctype_word) == 0)            if ((md->ctypes[*eptr++] & ctype_word) == 0)
4239              RRETURN(MATCH_NOMATCH);              MRRETURN(MATCH_NOMATCH);
4240              }
4241          break;          break;
4242    
4243          default:          default:
# Line 3402  for (;;) Line 4265  for (;;)
4265              {              {
4266              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);
4267              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4268              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
4269              GETCHARINC(c, eptr);              if (eptr >= md->end_subject)
4270              if (prop_fail_result) RRETURN(MATCH_NOMATCH);                {
4271                  SCHECK_PARTIAL();
4272                  MRRETURN(MATCH_NOMATCH);
4273                  }
4274                GETCHARINCTEST(c, eptr);
4275                if (prop_fail_result) MRRETURN(MATCH_NOMATCH);
4276              }              }
4277            /* Control never gets here */            /* Control never gets here */
4278    
# Line 3413  for (;;) Line 4281  for (;;)
4281              {              {
4282              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);
4283              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4284              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
4285              GETCHARINC(c, eptr);              if (eptr >= md->end_subject)
4286                  {
4287                  SCHECK_PARTIAL();
4288                  MRRETURN(MATCH_NOMATCH);
4289                  }
4290                GETCHARINCTEST(c, eptr);
4291              prop_chartype = UCD_CHARTYPE(c);              prop_chartype = UCD_CHARTYPE(c);
4292              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
4293                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
4294                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
4295                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4296                }
4297              /* Control never gets here */
4298    
4299              case PT_GC:
4300              for (fi = min;; fi++)
4301                {
4302                RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);
4303                if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4304                if (fi >= max) MRRETURN(MATCH_NOMATCH);
4305                if (eptr >= md->end_subject)
4306                  {
4307                  SCHECK_PARTIAL();
4308                  MRRETURN(MATCH_NOMATCH);
4309                  }
4310                GETCHARINCTEST(c, eptr);
4311                prop_category = UCD_CATEGORY(c);
4312                if ((prop_category == prop_value) == prop_fail_result)
4313                  MRRETURN(MATCH_NOMATCH);
4314                }
4315              /* Control never gets here */
4316    
4317              case PT_PC:
4318              for (fi = min;; fi++)
4319                {
4320                RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);
4321                if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4322                if (fi >= max) MRRETURN(MATCH_NOMATCH);
4323                if (eptr >= md->end_subject)
4324                  {
4325                  SCHECK_PARTIAL();
4326                  MRRETURN(MATCH_NOMATCH);
4327                  }
4328                GETCHARINCTEST(c, eptr);
4329                prop_chartype = UCD_CHARTYPE(c);
4330                if ((prop_chartype == prop_value) == prop_fail_result)
4331                  MRRETURN(MATCH_NOMATCH);
4332                }
4333              /* Control never gets here */
4334    
4335              case PT_SC:
4336              for (fi = min;; fi++)
4337                {
4338                RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);
4339                if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4340                if (fi >= max) MRRETURN(MATCH_NOMATCH);
4341                if (eptr >= md->end_subject)
4342                  {
4343                  SCHECK_PARTIAL();
4344                  MRRETURN(MATCH_NOMATCH);
4345                  }
4346                GETCHARINCTEST(c, eptr);
4347                prop_script = UCD_SCRIPT(c);
4348                if ((prop_script == prop_value) == prop_fail_result)
4349                  MRRETURN(MATCH_NOMATCH);
4350                }
4351              /* Control never gets here */
4352    
4353              case PT_ALNUM:
4354              for (fi = min;; fi++)
4355                {
4356                RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM59);
4357                if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4358                if (fi >= max) MRRETURN(MATCH_NOMATCH);
4359                if (eptr >= md->end_subject)
4360                  {
4361                  SCHECK_PARTIAL();
4362                  MRRETURN(MATCH_NOMATCH);
4363                  }
4364                GETCHARINCTEST(c, eptr);
4365                prop_category = UCD_CATEGORY(c);
4366                if ((prop_category == ucp_L || prop_category == ucp_N)
4367                       == prop_fail_result)
4368                  MRRETURN(MATCH_NOMATCH);
4369              }              }
4370            /* Control never gets here */            /* Control never gets here */
4371    
4372            case PT_GC:            case PT_SPACE:    /* Perl space */
4373            for (fi = min;; fi++)            for (fi = min;; fi++)
4374              {              {
4375              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM60);
4376              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4377              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
4378              GETCHARINC(c, eptr);              if (eptr >= md->end_subject)
4379                  {
4380                  SCHECK_PARTIAL();
4381                  MRRETURN(MATCH_NOMATCH);
4382                  }
4383                GETCHARINCTEST(c, eptr);
4384              prop_category = UCD_CATEGORY(c);              prop_category = UCD_CATEGORY(c);
4385              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
4386                RRETURN(MATCH_NOMATCH);                   c == CHAR_FF || c == CHAR_CR)
4387                       == prop_fail_result)
4388                  MRRETURN(MATCH_NOMATCH);
4389              }              }
4390            /* Control never gets here */            /* Control never gets here */
4391    
4392            case PT_PC:            case PT_PXSPACE:  /* POSIX space */
4393            for (fi = min;; fi++)            for (fi = min;; fi++)
4394              {              {
4395              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM61);
4396              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4397              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
4398              GETCHARINC(c, eptr);              if (eptr >= md->end_subject)
4399              prop_chartype = UCD_CHARTYPE(c);                {
4400              if ((prop_chartype == prop_value) == prop_fail_result)                SCHECK_PARTIAL();
4401                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4402                  }
4403                GETCHARINCTEST(c, eptr);
4404                prop_category = UCD_CATEGORY(c);
4405                if ((prop_category == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
4406                     c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)
4407                       == prop_fail_result)
4408                  MRRETURN(MATCH_NOMATCH);
4409              }              }
4410            /* Control never gets here */            /* Control never gets here */
4411    
4412            case PT_SC:            case PT_WORD:
4413            for (fi = min;; fi++)            for (fi = min;; fi++)
4414              {              {
4415              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM62);
4416              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4417              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) MRRETURN(MATCH_NOMATCH);
4418              GETCHARINC(c, eptr);              if (eptr >= md->end_subject)
4419              prop_script = UCD_SCRIPT(c);                {
4420              if ((prop_script == prop_value) == prop_fail_result)                SCHECK_PARTIAL();
4421                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4422                  }
4423                GETCHARINCTEST(c, eptr);
4424                prop_category = UCD_CATEGORY(c);
4425                if ((prop_category == ucp_L ||
4426                     prop_category == ucp_N ||
4427                     c == CHAR_UNDERSCORE)
4428                       == prop_fail_result)
4429                  MRRETURN(MATCH_NOMATCH);
4430              }              }
4431            /* Control never gets here */            /* Control never gets here */
4432    
4433              /* This should never occur */
4434    
4435            default:            default:
4436            RRETURN(PCRE_ERROR_INTERNAL);            RRETURN(PCRE_ERROR_INTERNAL);
4437            }            }
# Line 3476  for (;;) Line 4446  for (;;)
4446            {            {
4447            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);
4448            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4449            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max) MRRETURN(MATCH_NOMATCH);
4450              if (eptr >= md->end_subject)
4451                {
4452                SCHECK_PARTIAL();
4453                MRRETURN(MATCH_NOMATCH);
4454                }
4455            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
4456            prop_category = UCD_CATEGORY(c);            prop_category = UCD_CATEGORY(c);
4457            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);            if (prop_category == ucp_M) MRRETURN(MATCH_NOMATCH);
4458            while (eptr < md->end_subject)            while (eptr < md->end_subject)
4459              {              {
4460              int len = 1;              int len = 1;
4461              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
4462                {                else { GETCHARLEN(c, eptr, len); }
               GETCHARLEN(c, eptr, len);  
               }  
4463              prop_category = UCD_CATEGORY(c);              prop_category = UCD_CATEGORY(c);
4464              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
4465              eptr += len;              eptr += len;
# Line 3505  for (;;) Line 4478  for (;;)
4478            {            {
4479            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);
4480            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4481            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) MRRETURN(MATCH_NOMATCH);
4482                 (ctype == OP_ANY && IS_NEWLINE(eptr)))            if (eptr >= md->end_subject)
4483              RRETURN(MATCH_NOMATCH);              {
4484                SCHECK_PARTIAL();
4485                MRRETURN(MATCH_NOMATCH);
4486                }
4487              if (ctype == OP_ANY && IS_NEWLINE(eptr))
4488                MRRETURN(MATCH_NOMATCH);
4489            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4490            switch(ctype)            switch(ctype)
4491              {              {
# Line 3520  for (;;) Line 4497  for (;;)
4497              case OP_ANYNL:              case OP_ANYNL:
4498              switch(c)              switch(c)
4499                {                {
4500                default: RRETURN(MATCH_NOMATCH);                default: MRRETURN(MATCH_NOMATCH);
4501                case 0x000d:                case 0x000d:
4502                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
4503                break;                break;
# Line 3532  for (;;) Line 4509  for (;;)
4509                case 0x0085:                case 0x0085:
4510                case 0x2028:                case 0x2028:
4511                case 0x2029:                case 0x2029:
4512                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);                if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);
4513                break;                break;
4514                }                }
4515              break;              break;
# Line 3560  for (;;) Line 4537  for (;;)
4537                case 0x202f:    /* NARROW NO-BREAK SPACE */                case 0x202f:    /* NARROW NO-BREAK SPACE */
4538                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4539                case 0x3000:    /* IDEOGRAPHIC SPACE */                case 0x3000:    /* IDEOGRAPHIC SPACE */
4540                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4541                }                }
4542              break;              break;
4543    
4544              case OP_HSPACE:              case OP_HSPACE:
4545              switch(c)              switch(c)
4546                {                {
4547                default: RRETURN(MATCH_NOMATCH);                default: MRRETURN(MATCH_NOMATCH);
4548                case 0x09:      /* HT */                case 0x09:      /* HT */
4549                case 0x20:      /* SPACE */                case 0x20:      /* SPACE */
4550                case 0xa0:      /* NBSP */                case 0xa0:      /* NBSP */
# Line 3602  for (;;) Line 4579  for (;;)
4579                case 0x85:      /* NEL */                case 0x85:      /* NEL */
4580                case 0x2028:    /* LINE SEPARATOR */                case 0x2028:    /* LINE SEPARATOR */
4581                case 0x2029:    /* PARAGRAPH SEPARATOR */                case 0x2029:    /* PARAGRAPH SEPARATOR */
4582                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4583                }                }
4584              break;              break;
4585    
4586              case OP_VSPACE:              case OP_VSPACE:
4587              switch(c)              switch(c)
4588                {                {
4589                default: RRETURN(MATCH_NOMATCH);                default: MRRETURN(MATCH_NOMATCH);
4590                case 0x0a:      /* LF */                case 0x0a:      /* LF */
4591                case 0x0b:      /* VT */                case 0x0b:      /* VT */
4592                case 0x0c:      /* FF */                case 0x0c:      /* FF */
# Line 3623  for (;;) Line 4600  for (;;)
4600    
4601              case OP_NOT_DIGIT:              case OP_NOT_DIGIT:
4602              if (c < 256 && (md->ctypes[c] & ctype_digit) != 0)              if (c < 256 && (md->ctypes[c] & ctype_digit) != 0)
4603                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4604              break;              break;
4605    
4606              case OP_DIGIT:              case OP_DIGIT:
4607              if (c >= 256 || (md->ctypes[c] & ctype_digit) == 0)              if (c >= 256 || (md->ctypes[c] & ctype_digit) == 0)
4608                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4609              break;              break;
4610    
4611              case OP_NOT_WHITESPACE:              case OP_NOT_WHITESPACE:
4612              if (c < 256 && (md->ctypes[c] & ctype_space) != 0)              if (c < 256 && (md->ctypes[c] & ctype_space) != 0)
4613                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4614              break;              break;
4615    
4616              case OP_WHITESPACE:              case OP_WHITESPACE:
4617              if  (c >= 256 || (md->ctypes[c] & ctype_space) == 0)              if  (c >= 256 || (md->ctypes[c] & ctype_space) == 0)
4618                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4619              break;              break;
4620    
4621              case OP_NOT_WORDCHAR:              case OP_NOT_WORDCHAR:
4622              if (c < 256 && (md->ctypes[c] & ctype_word) != 0)              if (c < 256 && (md->ctypes[c] & ctype_word) != 0)
4623                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4624              break;              break;
4625    
4626              case OP_WORDCHAR:              case OP_WORDCHAR:
4627              if (c >= 256 || (md->ctypes[c] & ctype_word) == 0)              if (c >= 256 || (md->ctypes[c] & ctype_word) == 0)
4628                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4629              break;              break;
4630    
4631              default:              default:
# Line 3664  for (;;) Line 4641  for (;;)
4641            {            {
4642            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);
4643            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4644            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) MRRETURN(MATCH_NOMATCH);
4645                 (ctype == OP_ANY && IS_NEWLINE(eptr)))            if (eptr >= md->end_subject)
4646              RRETURN(MATCH_NOMATCH);              {
4647                SCHECK_PARTIAL();
4648                MRRETURN(MATCH_NOMATCH);
4649                }
4650              if (ctype == OP_ANY && IS_NEWLINE(eptr))
4651                MRRETURN(MATCH_NOMATCH);
4652            c = *eptr++;            c = *eptr++;
4653            switch(ctype)            switch(ctype)
4654              {              {
# Line 3679  for (;;) Line 4660  for (;;)
4660              case OP_ANYNL:              case OP_ANYNL:
4661              switch(c)              switch(c)
4662                {                {
4663                default: RRETURN(MATCH_NOMATCH);                default: MRRETURN(MATCH_NOMATCH);
4664                case 0x000d:                case 0x000d:
4665                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
4666                break;                break;
# Line 3690  for (;;) Line 4671  for (;;)
4671                case 0x000b:                case 0x000b:
4672                case 0x000c:                case 0x000c:
4673                case 0x0085:                case 0x0085:
4674                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);                if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);
4675                break;                break;
4676                }                }
4677              break;              break;
# Line 3702  for (;;) Line 4683  for (;;)
4683                case 0x09:      /* HT */                case 0x09:      /* HT */
4684                case 0x20:      /* SPACE */                case 0x20:      /* SPACE */
4685                case 0xa0:      /* NBSP */                case 0xa0:      /* NBSP */
4686                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4687                }                }
4688              break;              break;
4689    
4690              case OP_HSPACE:              case OP_HSPACE:
4691              switch(c)              switch(c)
4692                {                {
4693                default: RRETURN(MATCH_NOMATCH);                default: MRRETURN(MATCH_NOMATCH);
4694                case 0x09:      /* HT */                case 0x09:      /* HT */
4695                case 0x20:      /* SPACE */                case 0x20:      /* SPACE */
4696                case 0xa0:      /* NBSP */                case 0xa0:      /* NBSP */
# Line 3726  for (;;) Line 4707  for (;;)
4707                case 0x0c:      /* FF */                case 0x0c:      /* FF */
4708                case 0x0d:      /* CR */                case 0x0d:      /* CR */
4709                case 0x85:      /* NEL */                case 0x85:      /* NEL */
4710                RRETURN(MATCH_NOMATCH);                MRRETURN(MATCH_NOMATCH);
4711                }                }
4712              break;              break;
4713    
4714              case OP_VSPACE:              case OP_VSPACE:
4715