/[pcre]/code/branches/pcre16/pcre_exec.c
ViewVC logotype

Diff of /code/branches/pcre16/pcre_exec.c

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

revision 178 by ph10, Wed Jun 13 08:44:34 2007 UTC revision 428 by ph10, Mon Aug 31 17:10:26 2009 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) 1997-2007 University of Cambridge             Copyright (c) 1997-2009 University of Cambridge
10    
11  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
12  Redistribution and use in source and binary forms, with or without  Redistribution and use in source and binary forms, with or without
# Line 42  POSSIBILITY OF SUCH DAMAGE. Line 42  POSSIBILITY OF SUCH DAMAGE.
42  pattern matching using an NFA algorithm, trying to mimic Perl as closely as  pattern matching using an NFA algorithm, trying to mimic Perl as closely as
43  possible. There are also some static supporting functions. */  possible. There are also some static supporting functions. */
44    
45    #ifdef HAVE_CONFIG_H
46    #include "config.h"
47    #endif
48    
49  #define NLBLOCK md             /* Block containing newline information */  #define NLBLOCK md             /* Block containing newline information */
50  #define PSSTART start_subject  /* Field containing processed string start */  #define PSSTART start_subject  /* Field containing processed string start */
51  #define PSEND   end_subject    /* Field containing processed string end */  #define PSEND   end_subject    /* Field containing processed string end */
# Line 53  possible. There are also some static sup Line 57  possible. There are also some static sup
57  #undef min  #undef min
58  #undef max  #undef max
59    
 /* The chain of eptrblocks for tail recursions uses memory in stack workspace,  
 obtained at top level, the size of which is defined by EPTR_WORK_SIZE. */  
   
 #define EPTR_WORK_SIZE (1000)  
   
60  /* Flag bits for the match() function */  /* Flag bits for the match() function */
61    
62  #define match_condassert     0x01  /* Called to check a condition assertion */  #define match_condassert     0x01  /* Called to check a condition assertion */
63  #define match_cbegroup       0x02  /* Could-be-empty unlimited repeat group */  #define match_cbegroup       0x02  /* Could-be-empty unlimited repeat group */
 #define match_tail_recursed  0x04  /* Tail recursive call */  
64    
65  /* Non-error returns from the match() function. Error returns are externally  /* Non-error returns from the match() function. Error returns are externally
66  defined PCRE_ERROR_xxx codes, which are all negative. */  defined PCRE_ERROR_xxx codes, which are all negative. */
# Line 70  defined PCRE_ERROR_xxx codes, which are Line 68  defined PCRE_ERROR_xxx codes, which are
68  #define MATCH_MATCH        1  #define MATCH_MATCH        1
69  #define MATCH_NOMATCH      0  #define MATCH_NOMATCH      0
70    
71    /* Special internal returns from the match() function. Make them sufficiently
72    negative to avoid the external error codes. */
73    
74    #define MATCH_COMMIT       (-999)
75    #define MATCH_PRUNE        (-998)
76    #define MATCH_SKIP         (-997)
77    #define MATCH_THEN         (-996)
78    
79  /* 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.
80  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,
81  because the offset vector is always a multiple of 3 long. */  because the offset vector is always a multiple of 3 long. */
# Line 152  printf("\n"); Line 158  printf("\n");
158    
159  if (length > md->end_subject - eptr) return FALSE;  if (length > md->end_subject - eptr) return FALSE;
160    
161  /* Separate the caselesss case for speed */  /* Separate the caseless case for speed. In UTF-8 mode we can only do this
162    properly if Unicode properties are supported. Otherwise, we can check only
163    ASCII characters. */
164    
165  if ((ims & PCRE_CASELESS) != 0)  if ((ims & PCRE_CASELESS) != 0)
166    {    {
167    #ifdef SUPPORT_UTF8
168    #ifdef SUPPORT_UCP
169      if (md->utf8)
170        {
171        USPTR endptr = eptr + length;
172        while (eptr < endptr)
173          {
174          int c, d;
175          GETCHARINC(c, eptr);
176          GETCHARINC(d, p);
177          if (c != d && c != UCD_OTHERCASE(d)) return FALSE;
178          }
179        }
180      else
181    #endif
182    #endif
183    
184      /* The same code works when not in UTF-8 mode and in UTF-8 mode when there
185      is no UCP support. */
186    
187    while (length-- > 0)    while (length-- > 0)
188      if (md->lcc[*p++] != md->lcc[*eptr++]) return FALSE;      { if (md->lcc[*p++] != md->lcc[*eptr++]) return FALSE; }
189    }    }
190    
191    /* In the caseful case, we can just compare the bytes, whether or not we
192    are in UTF-8 mode. */
193    
194  else  else
195    { while (length-- > 0) if (*p++ != *eptr++) return FALSE; }    { while (length-- > 0) if (*p++ != *eptr++) return FALSE; }
196    
# Line 205  variable instead of being passed in the Line 237  variable instead of being passed in the
237  ****************************************************************************  ****************************************************************************
238  ***************************************************************************/  ***************************************************************************/
239    
240    /* Numbers for RMATCH calls. When this list is changed, the code at HEAP_RETURN
241  /* Numbers for RMATCH calls */  below must be updated in sync.  */
242    
243  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM6,  RM7,  RM8,  RM9,  RM10,  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM6,  RM7,  RM8,  RM9,  RM10,
244         RM11,  RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,         RM11,  RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,
245         RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,         RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,
246         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
247         RM41,  RM42, RM43, RM44, RM45, RM46, RM47 };         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
248           RM51,  RM52, RM53, RM54 };
249    
250  /* 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
251  versions and production versions. Note that the "rw" argument of RMATCH isn't  versions and production versions. Note that the "rw" argument of RMATCH isn't
# Line 290  typedef struct heapframe { Line 322  typedef struct heapframe {
322    
323    /* Function arguments that may change */    /* Function arguments that may change */
324    
325    const uschar *Xeptr;    USPTR Xeptr;
326    const uschar *Xecode;    const uschar *Xecode;
327    const uschar *Xmstart;    USPTR Xmstart;
328    int Xoffset_top;    int Xoffset_top;
329    long int Xims;    long int Xims;
330    eptrblock *Xeptrb;    eptrblock *Xeptrb;
# Line 301  typedef struct heapframe { Line 333  typedef struct heapframe {
333    
334    /* Function local variables */    /* Function local variables */
335    
336    const uschar *Xcallpat;    USPTR Xcallpat;
337    const uschar *Xcharptr;  #ifdef SUPPORT_UTF8
338    const uschar *Xdata;    USPTR Xcharptr;
339    const uschar *Xnext;  #endif
340    const uschar *Xpp;    USPTR Xdata;
341    const uschar *Xprev;    USPTR Xnext;
342    const uschar *Xsaved_eptr;    USPTR Xpp;
343      USPTR Xprev;
344      USPTR Xsaved_eptr;
345    
346    recursion_info Xnew_recursive;    recursion_info Xnew_recursive;
347    
# Line 328  typedef struct heapframe { Line 362  typedef struct heapframe {
362    uschar Xocchars[8];    uschar Xocchars[8];
363  #endif  #endif
364    
365      int Xcodelink;
366    int Xctype;    int Xctype;
367    unsigned int Xfc;    unsigned int Xfc;
368    int Xfi;    int Xfi;
# Line 363  typedef struct heapframe { Line 398  typedef struct heapframe {
398    
399  /* This function is called recursively in many circumstances. Whenever it  /* This function is called recursively in many circumstances. Whenever it
400  returns a negative (error) response, the outer incarnation must also return the  returns a negative (error) response, the outer incarnation must also return the
401  same response.  same response. */
402    
403    /* These macros pack up tests that are used for partial matching, and which
404    appears several times in the code. We set the "hit end" flag if the pointer is
405    at the end of the subject and also past the start of the subject (i.e.
406    something has been matched). For hard partial matching, we then return
407    immediately. The second one is used when we already know we are past the end of
408    the subject. */
409    
410  Performance note: It might be tempting to extract commonly used fields from the  #define CHECK_PARTIAL()\
411  md structure (e.g. utf8, end_subject) into individual variables to improve    if (md->partial && eptr >= md->end_subject && eptr > mstart)\
412        {\
413        md->hitend = TRUE;\
414        if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);\
415        }
416    
417    #define SCHECK_PARTIAL()\
418      if (md->partial && eptr > mstart)\
419        {\
420        md->hitend = TRUE;\
421        if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);\
422        }
423    
424    
425    /* Performance note: It might be tempting to extract commonly used fields from
426    the md structure (e.g. utf8, end_subject) into individual variables to improve
427  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
428  made performance worse.  made performance worse.
429    
# Line 384  Arguments: Line 441  Arguments:
441                   match_condassert - this is an assertion condition                   match_condassert - this is an assertion condition
442                   match_cbegroup - this is the start of an unlimited repeat                   match_cbegroup - this is the start of an unlimited repeat
443                     group that can match an empty string                     group that can match an empty string
                  match_tail_recursed - this is a tail_recursed group  
444     rdepth      the recursion depth     rdepth      the recursion depth
445    
446  Returns:       MATCH_MATCH if matched            )  these values are >= 0  Returns:       MATCH_MATCH if matched            )  these values are >= 0
# Line 394  Returns:       MATCH_MATCH if matched Line 450  Returns:       MATCH_MATCH if matched
450  */  */
451    
452  static int  static int
453  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, const uschar *mstart,  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, USPTR mstart,
454    int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,    int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,
455    int flags, unsigned int rdepth)    int flags, unsigned int rdepth)
456  {  {
# Line 408  register unsigned int c;   /* Character Line 464  register unsigned int c;   /* Character
464  register BOOL utf8;        /* Local copy of UTF-8 flag for speed */  register BOOL utf8;        /* Local copy of UTF-8 flag for speed */
465    
466  BOOL minimize, possessive; /* Quantifier options */  BOOL minimize, possessive; /* Quantifier options */
467    int condcode;
468    
469  /* When recursion is not being used, all "local" variables that have to be  /* When recursion is not being used, all "local" variables that have to be
470  preserved over calls to RMATCH() are part of a "frame" which is obtained from  preserved over calls to RMATCH() are part of a "frame" which is obtained from
# Line 450  HEAP_RECURSE: Line 507  HEAP_RECURSE:
507  #define charptr            frame->Xcharptr  #define charptr            frame->Xcharptr
508  #endif  #endif
509  #define callpat            frame->Xcallpat  #define callpat            frame->Xcallpat
510    #define codelink           frame->Xcodelink
511  #define data               frame->Xdata  #define data               frame->Xdata
512  #define next               frame->Xnext  #define next               frame->Xnext
513  #define pp                 frame->Xpp  #define pp                 frame->Xpp
# Line 530  int oclength; Line 588  int oclength;
588  uschar occhars[8];  uschar occhars[8];
589  #endif  #endif
590    
591    int codelink;
592  int ctype;  int ctype;
593  int length;  int length;
594  int max;  int max;
# Line 586  original_ims = ims;    /* Save for reset Line 645  original_ims = ims;    /* Save for reset
645  string, the match_cbegroup flag is set. When this is the case, add the current  string, the match_cbegroup flag is set. When this is the case, add the current
646  subject pointer to the chain of such remembered pointers, to be checked when we  subject pointer to the chain of such remembered pointers, to be checked when we
647  hit the closing ket, in order to break infinite loops that match no characters.  hit the closing ket, in order to break infinite loops that match no characters.
648  When match() is called in other circumstances, don't add to the chain. If this  When match() is called in other circumstances, don't add to the chain. The
649  is a tail recursion, use a block from the workspace, as the one on the stack is  match_cbegroup flag must NOT be used with tail recursion, because the memory
650  already used. */  block that is used is on the stack, so a new one may be required for each
651    match(). */
652    
653  if ((flags & match_cbegroup) != 0)  if ((flags & match_cbegroup) != 0)
654    {    {
655    eptrblock *p;    newptrb.epb_saved_eptr = eptr;
656    if ((flags & match_tail_recursed) != 0)    newptrb.epb_prev = eptrb;
657      {    eptrb = &newptrb;
     if (md->eptrn >= EPTR_WORK_SIZE) RRETURN(PCRE_ERROR_NULLWSLIMIT);  
     p = md->eptrchain + md->eptrn++;  
     }  
   else p = &newptrb;  
   p->epb_saved_eptr = eptr;  
   p->epb_prev = eptrb;  
   eptrb = p;  
658    }    }
659    
660  /* Now start processing the opcodes. */  /* Now start processing the opcodes. */
# Line 610  for (;;) Line 663  for (;;)
663    {    {
664    minimize = possessive = FALSE;    minimize = possessive = FALSE;
665    op = *ecode;    op = *ecode;
666    
   /* 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;  
   
667    switch(op)    switch(op)
668      {      {
669        case OP_FAIL:
670        RRETURN(MATCH_NOMATCH);
671    
672        case OP_PRUNE:
673        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
674          ims, eptrb, flags, RM51);
675        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
676        RRETURN(MATCH_PRUNE);
677    
678        case OP_COMMIT:
679        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
680          ims, eptrb, flags, RM52);
681        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
682        RRETURN(MATCH_COMMIT);
683    
684        case OP_SKIP:
685        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
686          ims, eptrb, flags, RM53);
687        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
688        md->start_match_ptr = eptr;   /* Pass back current position */
689        RRETURN(MATCH_SKIP);
690    
691        case OP_THEN:
692        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
693          ims, eptrb, flags, RM54);
694        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
695        RRETURN(MATCH_THEN);
696    
697      /* 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
698      the current subject position in the working slot at the top of the vector.      the current subject position in the working slot at the top of the vector.
699      We mustn't change the current values of the data slot, because they may be      We mustn't change the current values of the data slot, because they may be
# Line 662  for (;;) Line 735  for (;;)
735          {          {
736          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
737            ims, eptrb, flags, RM1);            ims, eptrb, flags, RM1);
738          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
739          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
740          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
741          }          }
# Line 677  for (;;) Line 750  for (;;)
750        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
751        }        }
752    
753      /* Insufficient room for saving captured contents. Treat as a non-capturing      /* FALL THROUGH ... Insufficient room for saving captured contents. Treat
754      bracket. */      as a non-capturing bracket. */
755    
756        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
757        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
758    
759      DPRINTF(("insufficient capture room: treat as non-capturing\n"));      DPRINTF(("insufficient capture room: treat as non-capturing\n"));
760    
761        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
762        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
763    
764      /* Non-capturing bracket. Loop for all the alternatives. When we get to the      /* Non-capturing bracket. Loop for all the alternatives. When we get to the
765      final alternative within the brackets, we would return the result of a      final alternative within the brackets, we would return the result of a
766      recursive call to match() whatever happened. We can reduce stack usage by      recursive call to match() whatever happened. We can reduce stack usage by
767      turning this into a tail recursion. */      turning this into a tail recursion, except in the case when match_cbegroup
768        is set.*/
769    
770      case OP_BRA:      case OP_BRA:
771      case OP_SBRA:      case OP_SBRA:
# Line 693  for (;;) Line 773  for (;;)
773      flags = (op >= OP_SBRA)? match_cbegroup : 0;      flags = (op >= OP_SBRA)? match_cbegroup : 0;
774      for (;;)      for (;;)
775        {        {
776        if (ecode[GET(ecode, 1)] != OP_ALT)        if (ecode[GET(ecode, 1)] != OP_ALT)   /* Final alternative */
777          {          {
778          ecode += _pcre_OP_lengths[*ecode];          if (flags == 0)    /* Not a possibly empty group */
779          flags |= match_tail_recursed;            {
780          DPRINTF(("bracket 0 tail recursion\n"));            ecode += _pcre_OP_lengths[*ecode];
781          goto TAIL_RECURSE;            DPRINTF(("bracket 0 tail recursion\n"));
782              goto TAIL_RECURSE;
783              }
784    
785            /* Possibly empty group; can't use tail recursion. */
786    
787            RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
788              eptrb, flags, RM48);
789            RRETURN(rrc);
790          }          }
791    
792        /* For non-final alternatives, continue the loop for a NOMATCH result;        /* For non-final alternatives, continue the loop for a NOMATCH result;
# Line 706  for (;;) Line 794  for (;;)
794    
795        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
796          eptrb, flags, RM2);          eptrb, flags, RM2);
797        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
798        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
799        }        }
800      /* Control never reaches here. */      /* Control never reaches here. */
# Line 719  for (;;) Line 807  for (;;)
807    
808      case OP_COND:      case OP_COND:
809      case OP_SCOND:      case OP_SCOND:
810      if (ecode[LINK_SIZE+1] == OP_RREF)         /* Recursion test */      codelink= GET(ecode, 1);
811    
812        /* Because of the way auto-callout works during compile, a callout item is
813        inserted between OP_COND and an assertion condition. */
814    
815        if (ecode[LINK_SIZE+1] == OP_CALLOUT)
816          {
817          if (pcre_callout != NULL)
818            {
819            pcre_callout_block cb;
820            cb.version          = 1;   /* Version 1 of the callout block */
821            cb.callout_number   = ecode[LINK_SIZE+2];
822            cb.offset_vector    = md->offset_vector;
823            cb.subject          = (PCRE_SPTR)md->start_subject;
824            cb.subject_length   = md->end_subject - md->start_subject;
825            cb.start_match      = mstart - md->start_subject;
826            cb.current_position = eptr - md->start_subject;
827            cb.pattern_position = GET(ecode, LINK_SIZE + 3);
828            cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);
829            cb.capture_top      = offset_top/2;
830            cb.capture_last     = md->capture_last;
831            cb.callout_data     = md->callout_data;
832            if ((rrc = (*pcre_callout)(&cb)) > 0) RRETURN(MATCH_NOMATCH);
833            if (rrc < 0) RRETURN(rrc);
834            }
835          ecode += _pcre_OP_lengths[OP_CALLOUT];
836          }
837    
838        condcode = ecode[LINK_SIZE+1];
839    
840        /* Now see what the actual condition is */
841    
842        if (condcode == OP_RREF)         /* Recursion test */
843        {        {
844        offset = GET2(ecode, LINK_SIZE + 2);     /* Recursion group number*/        offset = GET2(ecode, LINK_SIZE + 2);     /* Recursion group number*/
845        condition = md->recursive != NULL &&        condition = md->recursive != NULL &&
# Line 727  for (;;) Line 847  for (;;)
847        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 3 : GET(ecode, 1);
848        }        }
849    
850      else if (ecode[LINK_SIZE+1] == OP_CREF)    /* Group used test */      else if (condcode == OP_CREF)    /* Group used test */
851        {        {
852        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */
853        condition = offset < offset_top && md->offset_vector[offset] >= 0;        condition = offset < offset_top && md->offset_vector[offset] >= 0;
854        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 3 : GET(ecode, 1);
855        }        }
856    
857      else if (ecode[LINK_SIZE+1] == OP_DEF)     /* DEFINE - always false */      else if (condcode == OP_DEF)     /* DEFINE - always false */
858        {        {
859        condition = FALSE;        condition = FALSE;
860        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
# Line 754  for (;;) Line 874  for (;;)
874          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
875          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
876          }          }
877        else if (rrc != MATCH_NOMATCH)        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
878          {          {
879          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
880          }          }
881        else        else
882          {          {
883          condition = FALSE;          condition = FALSE;
884          ecode += GET(ecode, 1);          ecode += codelink;
885          }          }
886        }        }
887    
888      /* We are now at the branch that is to be obeyed. As there is only one,      /* We are now at the branch that is to be obeyed. As there is only one,
889      we can use tail recursion to avoid using another stack frame. If the second      we can use tail recursion to avoid using another stack frame, except when
890      alternative doesn't exist, we can just plough on. */      match_cbegroup is required for an unlimited repeat of a possibly empty
891        group. If the second alternative doesn't exist, we can just plough on. */
892    
893      if (condition || *ecode == OP_ALT)      if (condition || *ecode == OP_ALT)
894        {        {
895        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
896        flags = match_tail_recursed | ((op == OP_SCOND)? match_cbegroup : 0);        if (op == OP_SCOND)        /* Possibly empty group */
897        goto TAIL_RECURSE;          {
898            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, match_cbegroup, RM49);
899            RRETURN(rrc);
900            }
901          else                       /* Group must match something */
902            {
903            flags = 0;
904            goto TAIL_RECURSE;
905            }
906        }        }
907      else      else                         /* Condition false & no alternative */
908        {        {
909        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
910        }        }
911      break;      break;
912    
913    
914      /* End of the pattern. If we are in a top-level recursion, we should      /* End of the pattern, either real or forced. If we are in a top-level
915      restore the offsets appropriately and continue from after the call. */      recursion, we should restore the offsets appropriately and continue from
916        after the call. */
917    
918        case OP_ACCEPT:
919      case OP_END:      case OP_END:
920      if (md->recursive != NULL && md->recursive->group_num == 0)      if (md->recursive != NULL && md->recursive->group_num == 0)
921        {        {
# Line 805  for (;;) Line 936  for (;;)
936      if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);      if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);
937      md->end_match_ptr = eptr;           /* Record where we ended */      md->end_match_ptr = eptr;           /* Record where we ended */
938      md->end_offset_top = offset_top;    /* and how many extracts were taken */      md->end_offset_top = offset_top;    /* and how many extracts were taken */
939      md->start_match_ptr = mstart;  /* and the start (\K can modify) */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
940      RRETURN(MATCH_MATCH);      RRETURN(MATCH_MATCH);
941    
942      /* Change option settings */      /* Change option settings */
# Line 829  for (;;) Line 960  for (;;)
960        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
961          RM4);          RM4);
962        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
963        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
964        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
965        }        }
966      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 856  for (;;) Line 987  for (;;)
987        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
988          RM5);          RM5);
989        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);
990        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
991        ecode += GET(ecode,1);        ecode += GET(ecode,1);
992        }        }
993      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 880  for (;;) Line 1011  for (;;)
1011          {          {
1012          eptr--;          eptr--;
1013          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1014          BACKCHAR(eptr)          BACKCHAR(eptr);
1015          }          }
1016        }        }
1017      else      else
# Line 993  for (;;) Line 1124  for (;;)
1124              (pcre_free)(new_recursive.offset_save);              (pcre_free)(new_recursive.offset_save);
1125            RRETURN(MATCH_MATCH);            RRETURN(MATCH_MATCH);
1126            }            }
1127          else if (rrc != MATCH_NOMATCH)          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1128            {            {
1129            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1130              if (new_recursive.offset_save != stacksave)
1131                (pcre_free)(new_recursive.offset_save);
1132            RRETURN(rrc);            RRETURN(rrc);
1133            }            }
1134    
# Line 1027  for (;;) Line 1160  for (;;)
1160    
1161      do      do
1162        {        {
1163        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM7);
         eptrb, 0, RM7);  
1164        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
1165        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1166        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1167        }        }
1168      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 1073  for (;;) Line 1205  for (;;)
1205    
1206      if (*ecode == OP_KETRMIN)      if (*ecode == OP_KETRMIN)
1207        {        {
1208        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM8);
         RM8);  
1209        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1210        ecode = prev;        ecode = prev;
1211        flags = match_tail_recursed;        flags = 0;
1212        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1213        }        }
1214      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
# Line 1085  for (;;) Line 1216  for (;;)
1216        RMATCH(eptr, prev, offset_top, md, ims, eptrb, match_cbegroup, RM9);        RMATCH(eptr, prev, offset_top, md, ims, eptrb, match_cbegroup, RM9);
1217        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1218        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1219        flags = match_tail_recursed;        flags = 0;
1220        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1221        }        }
1222      /* Control never gets here */      /* Control never gets here */
# Line 1097  for (;;) Line 1228  for (;;)
1228      do ecode += GET(ecode,1); while (*ecode == OP_ALT);      do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1229      break;      break;
1230    
1231      /* BRAZERO and BRAMINZERO occur just before a bracket group, indicating      /* BRAZERO, BRAMINZERO and SKIPZERO occur just before a bracket group,
1232      that it may occur zero times. It may repeat infinitely, or not at all -      indicating that it may occur zero times. It may repeat infinitely, or not
1233      i.e. it could be ()* or ()? in the pattern. Brackets with fixed upper      at all - i.e. it could be ()* or ()? or even (){0} in the pattern. Brackets
1234      repeat limits are compiled as a number of copies, with the optional ones      with fixed upper repeat limits are compiled as a number of copies, with the
1235      preceded by BRAZERO or BRAMINZERO. */      optional ones preceded by BRAZERO or BRAMINZERO. */
1236    
1237      case OP_BRAZERO:      case OP_BRAZERO:
1238        {        {
# Line 1123  for (;;) Line 1254  for (;;)
1254        }        }
1255      break;      break;
1256    
1257        case OP_SKIPZERO:
1258          {
1259          next = ecode+1;
1260          do next += GET(next,1); while (*next == OP_ALT);
1261          ecode = next + 1 + LINK_SIZE;
1262          }
1263        break;
1264    
1265      /* End of a group, repeated or non-repeating. */      /* End of a group, repeated or non-repeating. */
1266    
1267      case OP_KET:      case OP_KET:
# Line 1216  for (;;) Line 1355  for (;;)
1355    
1356      /* The repeating kets try the rest of the pattern or restart from the      /* The repeating kets try the rest of the pattern or restart from the
1357      preceding bracket, in the appropriate order. In the second case, we can use      preceding bracket, in the appropriate order. In the second case, we can use
1358      tail recursion to avoid using another stack frame. */      tail recursion to avoid using another stack frame, unless we have an
1359        unlimited repeat of a group that can match an empty string. */
1360    
1361      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;
1362    
1363      if (*ecode == OP_KETRMIN)      if (*ecode == OP_KETRMIN)
1364        {        {
1365        RMATCH(eptr, ecode + 1+LINK_SIZE, offset_top, md, ims, eptrb, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM12);
         RM12);  
1366        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1367          if (flags != 0)    /* Could match an empty string */
1368            {
1369            RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM50);
1370            RRETURN(rrc);
1371            }
1372        ecode = prev;        ecode = prev;
       flags |= match_tail_recursed;  
1373        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1374        }        }
1375      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
# Line 1234  for (;;) Line 1377  for (;;)
1377        RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);        RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);
1378        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1379        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1380        flags = match_tail_recursed;        flags = 0;
1381        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1382        }        }
1383      /* Control never gets here */      /* Control never gets here */
# Line 1332  for (;;) Line 1475  for (;;)
1475          {          {
1476          if (eptr == md->start_subject) prev_is_word = FALSE; else          if (eptr == md->start_subject) prev_is_word = FALSE; else
1477            {            {
1478            const uschar *lastptr = eptr - 1;            USPTR lastptr = eptr - 1;
1479            while((*lastptr & 0xc0) == 0x80) lastptr--;            while((*lastptr & 0xc0) == 0x80) lastptr--;
1480            GETCHAR(c, lastptr);            GETCHAR(c, lastptr);
1481            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
1482            }            }
1483          if (eptr >= md->end_subject) cur_is_word = FALSE; else          if (eptr >= md->end_subject)
1484              {
1485              SCHECK_PARTIAL();
1486              cur_is_word = FALSE;
1487              }
1488            else
1489            {            {
1490            GETCHAR(c, eptr);            GETCHAR(c, eptr);
1491            cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;            cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
# Line 1346  for (;;) Line 1494  for (;;)
1494        else        else
1495  #endif  #endif
1496    
1497        /* More streamlined when not in UTF-8 mode */        /* Not in UTF-8 mode */
1498    
1499          {          {
1500          prev_is_word = (eptr != md->start_subject) &&          prev_is_word = (eptr != md->start_subject) &&
1501            ((md->ctypes[eptr[-1]] & ctype_word) != 0);            ((md->ctypes[eptr[-1]] & ctype_word) != 0);
1502          cur_is_word = (eptr < md->end_subject) &&          if (eptr >= md->end_subject)
1503            ((md->ctypes[*eptr] & ctype_word) != 0);            {
1504              SCHECK_PARTIAL();
1505              cur_is_word = FALSE;
1506              }
1507            else cur_is_word = ((md->ctypes[*eptr] & ctype_word) != 0);
1508          }          }
1509    
1510        /* Now see if the situation is what we want */        /* Now see if the situation is what we want */
# Line 1366  for (;;) Line 1518  for (;;)
1518      /* Match a single character type; inline for speed */      /* Match a single character type; inline for speed */
1519    
1520      case OP_ANY:      case OP_ANY:
1521      if ((ims & PCRE_DOTALL) == 0)      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
1522        /* Fall through */
1523    
1524        case OP_ALLANY:
1525        if (eptr++ >= md->end_subject)
1526        {        {
1527        if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);        SCHECK_PARTIAL();
1528        }        RRETURN(MATCH_NOMATCH);
1529      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);        }
1530      if (utf8)      if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
       while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  
1531      ecode++;      ecode++;
1532      break;      break;
1533    
# Line 1380  for (;;) Line 1535  for (;;)
1535      any byte, even newline, independent of the setting of PCRE_DOTALL. */      any byte, even newline, independent of the setting of PCRE_DOTALL. */
1536    
1537      case OP_ANYBYTE:      case OP_ANYBYTE:
1538      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr++ >= md->end_subject)
1539          {
1540          SCHECK_PARTIAL();
1541          RRETURN(MATCH_NOMATCH);
1542          }
1543      ecode++;      ecode++;
1544      break;      break;
1545    
1546      case OP_NOT_DIGIT:      case OP_NOT_DIGIT:
1547      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1548          {
1549          SCHECK_PARTIAL();
1550          RRETURN(MATCH_NOMATCH);
1551          }
1552      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1553      if (      if (
1554  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1398  for (;;) Line 1561  for (;;)
1561      break;      break;
1562    
1563      case OP_DIGIT:      case OP_DIGIT:
1564      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1565          {
1566          SCHECK_PARTIAL();
1567          RRETURN(MATCH_NOMATCH);
1568          }
1569      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1570      if (      if (
1571  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1411  for (;;) Line 1578  for (;;)
1578      break;      break;
1579    
1580      case OP_NOT_WHITESPACE:      case OP_NOT_WHITESPACE:
1581      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1582          {
1583          SCHECK_PARTIAL();
1584          RRETURN(MATCH_NOMATCH);
1585          }
1586      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1587      if (      if (
1588  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1424  for (;;) Line 1595  for (;;)
1595      break;      break;
1596    
1597      case OP_WHITESPACE:      case OP_WHITESPACE:
1598      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1599          {
1600          SCHECK_PARTIAL();
1601          RRETURN(MATCH_NOMATCH);
1602          }
1603      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1604      if (      if (
1605  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1437  for (;;) Line 1612  for (;;)
1612      break;      break;
1613    
1614      case OP_NOT_WORDCHAR:      case OP_NOT_WORDCHAR:
1615      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1616          {
1617          SCHECK_PARTIAL();
1618          RRETURN(MATCH_NOMATCH);
1619          }
1620      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1621      if (      if (
1622  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1450  for (;;) Line 1629  for (;;)
1629      break;      break;
1630    
1631      case OP_WORDCHAR:      case OP_WORDCHAR:
1632      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1633          {
1634          SCHECK_PARTIAL();
1635          RRETURN(MATCH_NOMATCH);
1636          }
1637      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1638      if (      if (
1639  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1463  for (;;) Line 1646  for (;;)
1646      break;      break;
1647    
1648      case OP_ANYNL:      case OP_ANYNL:
1649      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1650          {
1651          SCHECK_PARTIAL();
1652          RRETURN(MATCH_NOMATCH);
1653          }
1654      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1655      switch(c)      switch(c)
1656        {        {
# Line 1471  for (;;) Line 1658  for (;;)
1658        case 0x000d:        case 0x000d:
1659        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
1660        break;        break;
1661    
1662        case 0x000a:        case 0x000a:
1663          break;
1664    
1665        case 0x000b:        case 0x000b:
1666        case 0x000c:        case 0x000c:
1667        case 0x0085:        case 0x0085:
1668        case 0x2028:        case 0x2028:
1669        case 0x2029:        case 0x2029:
1670          if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
1671        break;        break;
1672        }        }
1673      ecode++;      ecode++;
1674      break;      break;
1675    
1676      case OP_NOT_HSPACE:      case OP_NOT_HSPACE:
1677      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1678          {
1679          SCHECK_PARTIAL();
1680          RRETURN(MATCH_NOMATCH);
1681          }
1682      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1683      switch(c)      switch(c)
1684        {        {
# Line 1513  for (;;) Line 1708  for (;;)
1708      break;      break;
1709    
1710      case OP_HSPACE:      case OP_HSPACE:
1711      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1712          {
1713          SCHECK_PARTIAL();
1714          RRETURN(MATCH_NOMATCH);
1715          }
1716      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1717      switch(c)      switch(c)
1718        {        {
# Line 1543  for (;;) Line 1742  for (;;)
1742      break;      break;
1743    
1744      case OP_NOT_VSPACE:      case OP_NOT_VSPACE:
1745      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1746          {
1747          SCHECK_PARTIAL();
1748          RRETURN(MATCH_NOMATCH);
1749          }
1750      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1751      switch(c)      switch(c)
1752        {        {
# Line 1561  for (;;) Line 1764  for (;;)
1764      break;      break;
1765    
1766      case OP_VSPACE:      case OP_VSPACE:
1767      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1768          {
1769          SCHECK_PARTIAL();
1770          RRETURN(MATCH_NOMATCH);
1771          }
1772      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1773      switch(c)      switch(c)
1774        {        {
# Line 1584  for (;;) Line 1791  for (;;)
1791    
1792      case OP_PROP:      case OP_PROP:
1793      case OP_NOTPROP:      case OP_NOTPROP:
1794      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1795          {
1796          SCHECK_PARTIAL();
1797          RRETURN(MATCH_NOMATCH);
1798          }
1799      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1800        {        {
1801        int chartype, script;        const ucd_record *prop = GET_UCD(c);
       int category = _pcre_ucp_findprop(c, &chartype, &script);  
1802    
1803        switch(ecode[1])        switch(ecode[1])
1804          {          {
# Line 1597  for (;;) Line 1807  for (;;)
1807          break;          break;
1808    
1809          case PT_LAMP:          case PT_LAMP:
1810          if ((chartype == ucp_Lu ||          if ((prop->chartype == ucp_Lu ||
1811               chartype == ucp_Ll ||               prop->chartype == ucp_Ll ||
1812               chartype == ucp_Lt) == (op == OP_NOTPROP))               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))
1813            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1814           break;           break;
1815    
1816          case PT_GC:          case PT_GC:
1817          if ((ecode[2] != category) == (op == OP_PROP))          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))
1818            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1819          break;          break;
1820    
1821          case PT_PC:          case PT_PC:
1822          if ((ecode[2] != chartype) == (op == OP_PROP))          if ((ecode[2] != prop->chartype) == (op == OP_PROP))
1823            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1824          break;          break;
1825    
1826          case PT_SC:          case PT_SC:
1827          if ((ecode[2] != script) == (op == OP_PROP))          if ((ecode[2] != prop->script) == (op == OP_PROP))
1828            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1829          break;          break;
1830    
# Line 1630  for (;;) Line 1840  for (;;)
1840      is in the binary; otherwise a compile-time error occurs. */      is in the binary; otherwise a compile-time error occurs. */
1841    
1842      case OP_EXTUNI:      case OP_EXTUNI:
1843      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1844          {
1845          SCHECK_PARTIAL();
1846          RRETURN(MATCH_NOMATCH);
1847          }
1848      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1849        {        {
1850        int chartype, script;        int category = UCD_CATEGORY(c);
       int category = _pcre_ucp_findprop(c, &chartype, &script);  
1851        if (category == ucp_M) RRETURN(MATCH_NOMATCH);        if (category == ucp_M) RRETURN(MATCH_NOMATCH);
1852        while (eptr < md->end_subject)        while (eptr < md->end_subject)
1853          {          {
# Line 1643  for (;;) Line 1856  for (;;)
1856            {            {
1857            GETCHARLEN(c, eptr, len);            GETCHARLEN(c, eptr, len);
1858            }            }
1859          category = _pcre_ucp_findprop(c, &chartype, &script);          category = UCD_CATEGORY(c);
1860          if (category != ucp_M) break;          if (category != ucp_M) break;
1861          eptr += len;          eptr += len;
1862          }          }
# Line 1664  for (;;) Line 1877  for (;;)
1877      case OP_REF:      case OP_REF:
1878        {        {
1879        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
1880        ecode += 3;                                 /* Advance past item */        ecode += 3;
1881    
1882          /* If the reference is unset, there are two possibilities:
1883    
1884        /* If the reference is unset, set the length to be longer than the amount        (a) In the default, Perl-compatible state, set the length to be longer
1885        of subject left; this ensures that every attempt at a match fails. We        than the amount of subject left; this ensures that every attempt at a
1886        can't just fail here, because of the possibility of quantifiers with zero        match fails. We can't just fail here, because of the possibility of
1887        minima. */        quantifiers with zero minima.
1888    
1889        length = (offset >= offset_top || md->offset_vector[offset] < 0)?        (b) If the JavaScript compatibility flag is set, set the length to zero
1890          md->end_subject - eptr + 1 :        so that the back reference matches an empty string.
1891          md->offset_vector[offset+1] - md->offset_vector[offset];  
1892          Otherwise, set the length to the length of what was matched by the
1893          referenced subpattern. */
1894    
1895          if (offset >= offset_top || md->offset_vector[offset] < 0)
1896            length = (md->jscript_compat)? 0 : md->end_subject - eptr + 1;
1897          else
1898            length = md->offset_vector[offset+1] - md->offset_vector[offset];
1899    
1900        /* Set up for repetition, or handle the non-repeated case */        /* Set up for repetition, or handle the non-repeated case */
1901    
# Line 1702  for (;;) Line 1924  for (;;)
1924          break;          break;
1925    
1926          default:               /* No repeat follows */          default:               /* No repeat follows */
1927          if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH);          if (!match_ref(offset, eptr, length, md, ims))
1928              {
1929              CHECK_PARTIAL();
1930              RRETURN(MATCH_NOMATCH);
1931              }
1932          eptr += length;          eptr += length;
1933          continue;              /* With the main loop */          continue;              /* With the main loop */
1934          }          }
1935    
1936        /* If the length of the reference is zero, just continue with the        /* If the length of the reference is zero, just continue with the
1937        main loop. */        main loop. */
1938    
1939        if (length == 0) continue;        if (length == 0) continue;
1940    
1941        /* First, ensure the minimum number of matches are present. We get back        /* First, ensure the minimum number of matches are present. We get back
# Line 1718  for (;;) Line 1944  for (;;)
1944    
1945        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
1946          {          {
1947          if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH);          if (!match_ref(offset, eptr, length, md, ims))
1948              {
1949              CHECK_PARTIAL();
1950              RRETURN(MATCH_NOMATCH);
1951              }
1952          eptr += length;          eptr += length;
1953          }          }
1954    
# Line 1735  for (;;) Line 1965  for (;;)
1965            {            {
1966            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);
1967            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1968            if (fi >= max || !match_ref(offset, eptr, length, md, ims))            if (fi >= max) RRETURN(MATCH_NOMATCH);
1969              if (!match_ref(offset, eptr, length, md, ims))
1970                {
1971                CHECK_PARTIAL();
1972              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
1973                }
1974            eptr += length;            eptr += length;
1975            }            }
1976          /* Control never gets here */          /* Control never gets here */
# Line 1763  for (;;) Line 1997  for (;;)
1997        }        }
1998      /* Control never gets here */      /* Control never gets here */
1999    
   
   
2000      /* Match a bit-mapped character class, possibly repeatedly. This op code is      /* Match a bit-mapped character class, possibly repeatedly. This op code is
2001      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,
2002      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 1819  for (;;) Line 2051  for (;;)
2051          {          {
2052          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2053            {            {
2054            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
2055                {
2056                SCHECK_PARTIAL();
2057                RRETURN(MATCH_NOMATCH);
2058                }
2059            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
2060            if (c > 255)            if (c > 255)
2061              {              {
# Line 1837  for (;;) Line 2073  for (;;)
2073          {          {
2074          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2075            {            {
2076            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
2077                {
2078                SCHECK_PARTIAL();
2079                RRETURN(MATCH_NOMATCH);
2080                }
2081            c = *eptr++;            c = *eptr++;
2082            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2083            }            }
# Line 1861  for (;;) Line 2101  for (;;)
2101              {              {
2102              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);
2103              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2104              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2105                if (eptr >= md->end_subject)
2106                  {
2107                  SCHECK_PARTIAL();
2108                  RRETURN(MATCH_NOMATCH);
2109                  }
2110              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
2111              if (c > 255)              if (c > 255)
2112                {                {
# Line 1881  for (;;) Line 2126  for (;;)
2126              {              {
2127              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);
2128              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2129              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2130                if (eptr >= md->end_subject)
2131                  {
2132                  SCHECK_PARTIAL();
2133                  RRETURN(MATCH_NOMATCH);
2134                  }
2135              c = *eptr++;              c = *eptr++;
2136              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2137              }              }
# Line 1948  for (;;) Line 2198  for (;;)
2198    
2199    
2200      /* Match an extended character class. This opcode is encountered only      /* Match an extended character class. This opcode is encountered only
2201      in UTF-8 mode, because that's the only time it is compiled. */      when UTF-8 mode mode is supported. Nevertheless, we may not be in UTF-8
2202        mode, because Unicode properties are supported in non-UTF-8 mode. */
2203    
2204  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
2205      case OP_XCLASS:      case OP_XCLASS:
# Line 1989  for (;;) Line 2240  for (;;)
2240    
2241        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2242          {          {
2243          if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);          if (eptr >= md->end_subject)
2244          GETCHARINC(c, eptr);            {
2245              SCHECK_PARTIAL();
2246              RRETURN(MATCH_NOMATCH);
2247              }
2248            GETCHARINCTEST(c, eptr);
2249          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);
2250          }          }
2251    
# Line 2008  for (;;) Line 2263  for (;;)
2263            {            {
2264            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);
2265            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2266            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
2267            GETCHARINC(c, eptr);            if (eptr >= md->end_subject)
2268                {
2269                SCHECK_PARTIAL();
2270                RRETURN(MATCH_NOMATCH);
2271                }
2272              GETCHARINCTEST(c, eptr);
2273            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);
2274            }            }
2275          /* Control never gets here */          /* Control never gets here */
# Line 2024  for (;;) Line 2284  for (;;)
2284            {            {
2285            int len = 1;            int len = 1;
2286            if (eptr >= md->end_subject) break;            if (eptr >= md->end_subject) break;
2287            GETCHARLEN(c, eptr, len);            GETCHARLENTEST(c, eptr, len);
2288            if (!_pcre_xclass(c, data)) break;            if (!_pcre_xclass(c, data)) break;
2289            eptr += len;            eptr += len;
2290            }            }
# Line 2033  for (;;) Line 2293  for (;;)
2293            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);
2294            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2295            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
2296            BACKCHAR(eptr)            if (utf8) BACKCHAR(eptr);
2297            }            }
2298          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2299          }          }
# Line 2051  for (;;) Line 2311  for (;;)
2311        length = 1;        length = 1;
2312        ecode++;        ecode++;
2313        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
2314        if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);        if (length > md->end_subject - eptr)
2315            {
2316            CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
2317            RRETURN(MATCH_NOMATCH);
2318            }
2319        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);
2320        }        }
2321      else      else
# Line 2059  for (;;) Line 2323  for (;;)
2323    
2324      /* Non-UTF-8 mode */      /* Non-UTF-8 mode */
2325        {        {
2326        if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH);        if (md->end_subject - eptr < 1)
2327            {
2328            SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
2329            RRETURN(MATCH_NOMATCH);
2330            }
2331        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);
2332        ecode += 2;        ecode += 2;
2333        }        }
# Line 2075  for (;;) Line 2343  for (;;)
2343        ecode++;        ecode++;
2344        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
2345    
2346        if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);        if (length > md->end_subject - eptr)
2347            {
2348            CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
2349            RRETURN(MATCH_NOMATCH);
2350            }
2351    
2352        /* 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
2353        can use the fast lookup table. */        can use the fast lookup table. */
# Line 2099  for (;;) Line 2371  for (;;)
2371          if (fc != dc)          if (fc != dc)
2372            {            {
2373  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2374            if (dc != _pcre_ucp_othercase(fc))            if (dc != UCD_OTHERCASE(fc))
2375  #endif  #endif
2376              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2377            }            }
# Line 2110  for (;;) Line 2382  for (;;)
2382    
2383      /* Non-UTF-8 mode */      /* Non-UTF-8 mode */
2384        {        {
2385        if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH);        if (md->end_subject - eptr < 1)
2386            {
2387            SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
2388            RRETURN(MATCH_NOMATCH);
2389            }
2390        if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);        if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2391        ecode += 2;        ecode += 2;
2392        }        }
# Line 2164  for (;;) Line 2440  for (;;)
2440      case OP_MINQUERY:      case OP_MINQUERY:
2441      c = *ecode++ - OP_STAR;      c = *ecode++ - OP_STAR;
2442      minimize = (c & 1) != 0;      minimize = (c & 1) != 0;
2443    
2444      min = rep_min[c];                 /* Pick up values from tables; */      min = rep_min[c];                 /* Pick up values from tables; */
2445      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
2446      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
2447    
2448      /* 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. */  
2449    
2450      REPEATCHAR:      REPEATCHAR:
2451  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 2179  for (;;) Line 2454  for (;;)
2454        length = 1;        length = 1;
2455        charptr = ecode;        charptr = ecode;
2456        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
       if (min * length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
2457        ecode += length;        ecode += length;
2458    
2459        /* Handle multibyte character matching specially here. There is        /* Handle multibyte character matching specially here. There is
# Line 2190  for (;;) Line 2464  for (;;)
2464  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2465          unsigned int othercase;          unsigned int othercase;
2466          if ((ims & PCRE_CASELESS) != 0 &&          if ((ims & PCRE_CASELESS) != 0 &&
2467              (othercase = _pcre_ucp_othercase(fc)) != NOTACHAR)              (othercase = UCD_OTHERCASE(fc)) != fc)
2468            oclength = _pcre_ord2utf8(othercase, occhars);            oclength = _pcre_ord2utf8(othercase, occhars);
2469          else oclength = 0;          else oclength = 0;
2470  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2471    
2472          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2473            {            {
2474            if (memcmp(eptr, charptr, length) == 0) eptr += length;            if (eptr <= md->end_subject - length &&
2475                memcmp(eptr, charptr, length) == 0) eptr += length;
2476  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2477            /* Need braces because of following else */            else if (oclength > 0 &&
2478            else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                     eptr <= md->end_subject - oclength &&
2479                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2480    #endif  /* SUPPORT_UCP */
2481            else            else
2482              {              {
2483              if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);              CHECK_PARTIAL();
2484              eptr += oclength;              RRETURN(MATCH_NOMATCH);
2485              }              }
 #else   /* without SUPPORT_UCP */  
           else { RRETURN(MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2486            }            }
2487    
2488          if (min == max) continue;          if (min == max) continue;
# Line 2219  for (;;) Line 2493  for (;;)
2493              {              {
2494              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);
2495              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2496              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2497              if (memcmp(eptr, charptr, length) == 0) eptr += length;              if (eptr <= md->end_subject - length &&
2498                  memcmp(eptr, charptr, length) == 0) eptr += length;
2499  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2500              /* Need braces because of following else */              else if (oclength > 0 &&
2501              else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                       eptr <= md->end_subject - oclength &&
2502                         memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2503    #endif  /* SUPPORT_UCP */
2504              else              else
2505                {                {
2506                if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);                CHECK_PARTIAL();
2507                eptr += oclength;                RRETURN(MATCH_NOMATCH);
2508                }                }
 #else   /* without SUPPORT_UCP */  
             else { RRETURN (MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2509              }              }
2510            /* Control never gets here */            /* Control never gets here */
2511            }            }
# Line 2241  for (;;) Line 2515  for (;;)
2515            pp = eptr;            pp = eptr;
2516            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2517              {              {
2518              if (eptr > md->end_subject - length) break;              if (eptr <= md->end_subject - length &&
2519              if (memcmp(eptr, charptr, length) == 0) eptr += length;                  memcmp(eptr, charptr, length) == 0) eptr += length;
2520  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2521              else if (oclength == 0) break;              else if (oclength > 0 &&
2522              else                       eptr <= md->end_subject - oclength &&
2523                {                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
               if (memcmp(eptr, occhars, oclength) != 0) break;  
               eptr += oclength;  
               }  
 #else   /* without SUPPORT_UCP */  
             else break;  
2524  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2525                else break;
2526              }              }
2527    
2528            if (possessive) continue;            if (possessive) continue;
2529    
2530            for(;;)            for(;;)
2531             {              {
2532             RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);
2533             if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2534             if (eptr == pp) RRETURN(MATCH_NOMATCH);              if (eptr == pp) { RRETURN(MATCH_NOMATCH); }
2535  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2536             eptr--;              eptr--;
2537             BACKCHAR(eptr);              BACKCHAR(eptr);
2538  #else   /* without SUPPORT_UCP */  #else   /* without SUPPORT_UCP */
2539             eptr -= length;              eptr -= length;
2540  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2541             }              }
2542            }            }
2543          /* Control never gets here */          /* Control never gets here */
2544          }          }
# Line 2280  for (;;) Line 2551  for (;;)
2551  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF8 */
2552    
2553      /* When not in UTF-8 mode, load a single-byte character. */      /* When not in UTF-8 mode, load a single-byte character. */
       {  
       if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
       fc = *ecode++;  
       }  
2554    
2555        fc = *ecode++;
2556    
2557      /* 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
2558      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
2559      caseful cases, for speed, since matching characters is likely to be quite      caseful cases, for speed, since matching characters is likely to be quite
# Line 2301  for (;;) Line 2570  for (;;)
2570        {        {
2571        fc = md->lcc[fc];        fc = md->lcc[fc];
2572        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2573            {
2574            if (eptr >= md->end_subject)
2575              {
2576              SCHECK_PARTIAL();
2577              RRETURN(MATCH_NOMATCH);
2578              }
2579          if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);          if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2580            }
2581        if (min == max) continue;        if (min == max) continue;
2582        if (minimize)        if (minimize)
2583          {          {
# Line 2309  for (;;) Line 2585  for (;;)
2585            {            {
2586            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);
2587            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2588            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) RRETURN(MATCH_NOMATCH);
2589                fc != md->lcc[*eptr++])            if (eptr >= md->end_subject)
2590                {
2591                SCHECK_PARTIAL();
2592              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2593                }
2594              if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2595            }            }
2596          /* Control never gets here */          /* Control never gets here */
2597          }          }
# Line 2323  for (;;) Line 2603  for (;;)
2603            if (eptr >= md->end_subject || fc != md->lcc[*eptr]) break;            if (eptr >= md->end_subject || fc != md->lcc[*eptr]) break;
2604            eptr++;            eptr++;
2605            }            }
2606    
2607          if (possessive) continue;          if (possessive) continue;
2608    
2609          while (eptr >= pp)          while (eptr >= pp)
2610            {            {
2611            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);
# Line 2339  for (;;) Line 2621  for (;;)
2621    
2622      else      else
2623        {        {
2624        for (i = 1; i <= min; i++) if (fc != *eptr++) RRETURN(MATCH_NOMATCH);        for (i = 1; i <= min; i++)
2625            {
2626            if (eptr >= md->end_subject)
2627              {
2628              SCHECK_PARTIAL();
2629              RRETURN(MATCH_NOMATCH);
2630              }
2631            if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
2632            }
2633    
2634        if (min == max) continue;        if (min == max) continue;
2635    
2636        if (minimize)        if (minimize)
2637          {          {
2638          for (fi = min;; fi++)          for (fi = min;; fi++)
2639            {            {
2640            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);
2641            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2642            if (fi >= max || eptr >= md->end_subject || fc != *eptr++)            if (fi >= max) RRETURN(MATCH_NOMATCH);
2643              if (eptr >= md->end_subject)
2644                {
2645                SCHECK_PARTIAL();
2646              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2647                }
2648              if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
2649            }            }
2650          /* Control never gets here */          /* Control never gets here */
2651          }          }
# Line 2361  for (;;) Line 2658  for (;;)
2658            eptr++;            eptr++;
2659            }            }
2660          if (possessive) continue;          if (possessive) continue;
2661    
2662          while (eptr >= pp)          while (eptr >= pp)
2663            {            {
2664            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27);
# Line 2376  for (;;) Line 2674  for (;;)
2674      checking can be multibyte. */      checking can be multibyte. */
2675    
2676      case OP_NOT:      case OP_NOT:
2677      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2678          {
2679          SCHECK_PARTIAL();
2680          RRETURN(MATCH_NOMATCH);
2681          }
2682      ecode++;      ecode++;
2683      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2684      if ((ims & PCRE_CASELESS) != 0)      if ((ims & PCRE_CASELESS) != 0)
# Line 2453  for (;;) Line 2755  for (;;)
2755      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
2756      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
2757    
2758      /* 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. */  
2759    
2760      REPEATNOTCHAR:      REPEATNOTCHAR:
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
2761      fc = *ecode++;      fc = *ecode++;
2762    
2763      /* 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 2483  for (;;) Line 2782  for (;;)
2782          register unsigned int d;          register unsigned int d;
2783          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2784            {            {
2785              if (eptr >= md->end_subject)
2786                {
2787                SCHECK_PARTIAL();
2788                RRETURN(MATCH_NOMATCH);
2789                }
2790            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
2791            if (d < 256) d = md->lcc[d];            if (d < 256) d = md->lcc[d];
2792            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
# Line 2494  for (;;) Line 2798  for (;;)
2798        /* Not UTF-8 mode */        /* Not UTF-8 mode */
2799          {          {
2800          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2801              {
2802              if (eptr >= md->end_subject)
2803                {
2804                SCHECK_PARTIAL();
2805                RRETURN(MATCH_NOMATCH);
2806                }
2807            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2808              }
2809          }          }
2810    
2811        if (min == max) continue;        if (min == max) continue;
# Line 2510  for (;;) Line 2821  for (;;)
2821              {              {
2822              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);
2823              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2824                if (fi >= max) RRETURN(MATCH_NOMATCH);
2825                if (eptr >= md->end_subject)
2826                  {
2827                  SCHECK_PARTIAL();
2828                  RRETURN(MATCH_NOMATCH);
2829                  }
2830              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
2831              if (d < 256) d = md->lcc[d];              if (d < 256) d = md->lcc[d];
2832              if (fi >= max || eptr >= md->end_subject || fc == d)              if (fc == d) RRETURN(MATCH_NOMATCH);
               RRETURN(MATCH_NOMATCH);  
2833              }              }
2834            }            }
2835          else          else
# Line 2524  for (;;) Line 2840  for (;;)
2840              {              {
2841              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);
2842              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2843              if (fi >= max || eptr >= md->end_subject || fc == md->lcc[*eptr++])              if (fi >= max) RRETURN(MATCH_NOMATCH);
2844                if (eptr >= md->end_subject)
2845                  {
2846                  SCHECK_PARTIAL();
2847                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2848                  }
2849                if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2850              }              }
2851            }            }
2852          /* Control never gets here */          /* Control never gets here */
# Line 2594  for (;;) Line 2915  for (;;)
2915          register unsigned int d;          register unsigned int d;
2916          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2917            {            {
2918              if (eptr >= md->end_subject)
2919                {
2920                SCHECK_PARTIAL();
2921                RRETURN(MATCH_NOMATCH);
2922                }
2923            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
2924            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
2925            }            }
# Line 2603  for (;;) Line 2929  for (;;)
2929        /* Not UTF-8 mode */        /* Not UTF-8 mode */
2930          {          {
2931          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2932              {
2933              if (eptr >= md->end_subject)
2934                {
2935                SCHECK_PARTIAL();
2936                RRETURN(MATCH_NOMATCH);
2937                }
2938            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
2939              }
2940          }          }
2941    
2942        if (min == max) continue;        if (min == max) continue;
# Line 2619  for (;;) Line 2952  for (;;)
2952              {              {
2953              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);
2954              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2955              GETCHARINC(d, eptr);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2956              if (fi >= max || eptr >= md->end_subject || fc == d)              if (eptr >= md->end_subject)
2957                  {
2958                  SCHECK_PARTIAL();
2959                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2960                  }
2961                GETCHARINC(d, eptr);
2962                if (fc == d) RRETURN(MATCH_NOMATCH);
2963              }              }
2964            }            }
2965          else          else
# Line 2632  for (;;) Line 2970  for (;;)
2970              {              {
2971              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);
2972              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2973              if (fi >= max || eptr >= md->end_subject || fc == *eptr++)              if (fi >= max) RRETURN(MATCH_NOMATCH);
2974                if (eptr >= md->end_subject)
2975                  {
2976                  SCHECK_PARTIAL();
2977                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2978                  }
2979                if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
2980              }              }
2981            }            }
2982          /* Control never gets here */          /* Control never gets here */
# Line 2767  for (;;) Line 3110  for (;;)
3110    
3111      /* First, ensure the minimum number of matches are present. Use inline      /* First, ensure the minimum number of matches are present. Use inline
3112      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
3113      (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  
3114      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
3115      and single-bytes. */      and single-bytes. */
3116    
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
3117      if (min > 0)      if (min > 0)
3118        {        {
3119  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 2785  for (;;) Line 3125  for (;;)
3125            if (prop_fail_result) RRETURN(MATCH_NOMATCH);            if (prop_fail_result) RRETURN(MATCH_NOMATCH);
3126            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3127              {              {
3128              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3129              GETCHARINC(c, eptr);                {
3130                  SCHECK_PARTIAL();
3131                  RRETURN(MATCH_NOMATCH);
3132                  }
3133                GETCHARINCTEST(c, eptr);
3134              }              }
3135            break;            break;
3136    
3137            case PT_LAMP:            case PT_LAMP:
3138            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3139              {              {
3140              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3141              GETCHARINC(c, eptr);                {
3142              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3143                  RRETURN(MATCH_NOMATCH);
3144                  }
3145                GETCHARINCTEST(c, eptr);
3146                prop_chartype = UCD_CHARTYPE(c);
3147              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
3148                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
3149                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 2806  for (;;) Line 3154  for (;;)
3154            case PT_GC:            case PT_GC:
3155            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3156              {              {
3157              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3158              GETCHARINC(c, eptr);                {
3159              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3160                  RRETURN(MATCH_NOMATCH);
3161                  }
3162                GETCHARINCTEST(c, eptr);
3163                prop_category = UCD_CATEGORY(c);
3164              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
3165                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3166              }              }
# Line 2817  for (;;) Line 3169  for (;;)
3169            case PT_PC:            case PT_PC:
3170            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3171              {              {
3172              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3173              GETCHARINC(c, eptr);                {
3174              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3175                  RRETURN(MATCH_NOMATCH);
3176                  }
3177                GETCHARINCTEST(c, eptr);
3178                prop_chartype = UCD_CHARTYPE(c);
3179              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
3180                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3181              }              }
# Line 2828  for (;;) Line 3184  for (;;)
3184            case PT_SC:            case PT_SC:
3185            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3186              {              {
3187              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3188              GETCHARINC(c, eptr);                {
3189              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3190                  RRETURN(MATCH_NOMATCH);
3191                  }
3192                GETCHARINCTEST(c, eptr);
3193                prop_script = UCD_SCRIPT(c);
3194              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
3195                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3196              }              }
# Line 2848  for (;;) Line 3208  for (;;)
3208          {          {
3209          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3210            {            {
3211            GETCHARINCTEST(c, eptr);            if (eptr >= md->end_subject)
3212            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              {
3213            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
3214                RRETURN(MATCH_NOMATCH);
3215                }
3216              GETCHARINCTEST(c, eptr);
3217              prop_category = UCD_CATEGORY(c);
3218              if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);
3219            while (eptr < md->end_subject)            while (eptr < md->end_subject)
3220              {              {
3221              int len = 1;              int len = 1;
3222              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
3223                {                else { GETCHARLEN(c, eptr, len); }
3224                GETCHARLEN(c, eptr, len);              prop_category = UCD_CATEGORY(c);
               }  
             prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);  
3225              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
3226              eptr += len;              eptr += len;
3227              }              }
# Line 2876  for (;;) Line 3239  for (;;)
3239          case OP_ANY:          case OP_ANY:
3240          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3241            {            {
3242            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3243                 ((ims & PCRE_DOTALL) == 0 && IS_NEWLINE(eptr)))              {
3244                SCHECK_PARTIAL();
3245              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3246                }
3247              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
3248              eptr++;
3249              while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3250              }
3251            break;
3252    
3253            case OP_ALLANY:
3254            for (i = 1; i <= min; i++)
3255              {
3256              if (eptr >= md->end_subject)
3257                {
3258                SCHECK_PARTIAL();
3259                RRETURN(MATCH_NOMATCH);
3260                }
3261            eptr++;            eptr++;
3262            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3263            }            }
3264          break;          break;
3265    
3266          case OP_ANYBYTE:          case OP_ANYBYTE:
3267            if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
3268          eptr += min;          eptr += min;
3269          break;          break;
3270    
3271          case OP_ANYNL:          case OP_ANYNL:
3272          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3273            {            {
3274            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3275                {
3276                SCHECK_PARTIAL();
3277                RRETURN(MATCH_NOMATCH);
3278                }
3279            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3280            switch(c)            switch(c)
3281              {              {
# Line 2899  for (;;) Line 3283  for (;;)
3283              case 0x000d:              case 0x000d:
3284              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3285              break;              break;
3286    
3287              case 0x000a:              case 0x000a:
3288                break;
3289    
3290              case 0x000b:              case 0x000b:
3291              case 0x000c:              case 0x000c:
3292              case 0x0085:              case 0x0085:
3293              case 0x2028:              case 0x2028:
3294              case 0x2029:              case 0x2029:
3295                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3296              break;              break;
3297              }              }
3298            }            }
# Line 2913  for (;;) Line 3301  for (;;)
3301          case OP_NOT_HSPACE:          case OP_NOT_HSPACE:
3302          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3303            {            {
3304            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3305                {
3306                SCHECK_PARTIAL();
3307                RRETURN(MATCH_NOMATCH);
3308                }
3309            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3310            switch(c)            switch(c)
3311              {              {
# Line 2941  for (;;) Line 3333  for (;;)
3333              }              }
3334            }            }
3335          break;          break;
3336    
3337          case OP_HSPACE:          case OP_HSPACE:
3338          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3339            {            {
3340            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3341                {
3342                SCHECK_PARTIAL();
3343                RRETURN(MATCH_NOMATCH);
3344                }
3345            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3346            switch(c)            switch(c)
3347              {              {
# Line 2973  for (;;) Line 3369  for (;;)
3369              }              }
3370            }            }
3371          break;          break;
3372    
3373          case OP_NOT_VSPACE:          case OP_NOT_VSPACE:
3374          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3375            {            {
3376            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3377                {
3378                SCHECK_PARTIAL();
3379                RRETURN(MATCH_NOMATCH);
3380                }
3381            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3382            switch(c)            switch(c)
3383              {              {
# Line 2993  for (;;) Line 3393  for (;;)
3393              }              }
3394            }            }
3395          break;          break;
3396    
3397          case OP_VSPACE:          case OP_VSPACE:
3398          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3399            {            {
3400            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3401                {
3402                SCHECK_PARTIAL();
3403                RRETURN(MATCH_NOMATCH);
3404                }
3405            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3406            switch(c)            switch(c)
3407              {              {
# Line 3009  for (;;) Line 3413  for (;;)
3413              case 0x85:      /* NEL */              case 0x85:      /* NEL */
3414              case 0x2028:    /* LINE SEPARATOR */              case 0x2028:    /* LINE SEPARATOR */
3415              case 0x2029:    /* PARAGRAPH SEPARATOR */              case 0x2029:    /* PARAGRAPH SEPARATOR */
3416              break;              break;
3417              }              }
3418            }            }
3419          break;          break;
# Line 3017  for (;;) Line 3421  for (;;)
3421          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
3422          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3423            {            {
3424            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3425                {
3426                SCHECK_PARTIAL();
3427                RRETURN(MATCH_NOMATCH);
3428                }
3429            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3430            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)
3431              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
# Line 3027  for (;;) Line 3435  for (;;)
3435          case OP_DIGIT:          case OP_DIGIT:
3436          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3437            {            {
3438            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3439               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)              {
3440                SCHECK_PARTIAL();
3441                RRETURN(MATCH_NOMATCH);
3442                }
3443              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)
3444              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3445            /* 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 */
3446            }            }
# Line 3037  for (;;) Line 3449  for (;;)
3449          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
3450          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3451            {            {
3452            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3453               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_space) != 0))              {
3454                SCHECK_PARTIAL();
3455              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3456            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;              }
3457              if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)
3458                RRETURN(MATCH_NOMATCH);
3459              while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3460            }            }
3461          break;          break;
3462    
3463          case OP_WHITESPACE:          case OP_WHITESPACE:
3464          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3465            {            {
3466            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3467               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)              {
3468                SCHECK_PARTIAL();
3469                RRETURN(MATCH_NOMATCH);
3470                }
3471              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)
3472              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3473            /* 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 */
3474            }            }
# Line 3058  for (;;) Line 3478  for (;;)
3478          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3479            {            {
3480            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3481               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_word) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0))
3482              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3483            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3484            }            }
3485          break;          break;
3486    
3487          case OP_WORDCHAR:          case OP_WORDCHAR:
3488          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3489            {            {
3490            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3491               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)              {
3492                SCHECK_PARTIAL();
3493                RRETURN(MATCH_NOMATCH);
3494                }
3495              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)
3496              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3497            /* 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 */
3498            }            }
# Line 3082  for (;;) Line 3506  for (;;)
3506  #endif     /* SUPPORT_UTF8 */  #endif     /* SUPPORT_UTF8 */
3507    
3508        /* 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
3509        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. */  
3510    
3511        switch(ctype)        switch(ctype)
3512          {          {
3513          case OP_ANY:          case OP_ANY:
3514          if ((ims & PCRE_DOTALL) == 0)          for (i = 1; i <= min; i++)
3515            {            {
3516            for (i = 1; i <= min; i++)            if (eptr >= md->end_subject)
3517              {              {
3518              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
3519              eptr++;              RRETURN(MATCH_NOMATCH);
3520              }              }
3521              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
3522              eptr++;
3523            }            }
         else eptr += min;  
3524          break;          break;
3525    
3526          case OP_ANYBYTE:          case OP_ALLANY:
3527            if (eptr > md->end_subject - min)
3528              {
3529              SCHECK_PARTIAL();
3530              RRETURN(MATCH_NOMATCH);
3531              }
3532          eptr += min;          eptr += min;
3533          break;          break;
3534    
3535          /* Because of the CRLF case, we can't assume the minimum number of          case OP_ANYBYTE:
3536          bytes are present in this case. */          if (eptr > md->end_subject - min)
3537              {
3538              SCHECK_PARTIAL();
3539              RRETURN(MATCH_NOMATCH);
3540              }
3541            eptr += min;
3542            break;
3543    
3544          case OP_ANYNL:          case OP_ANYNL:
3545          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3546            {            {
3547            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3548                {
3549                SCHECK_PARTIAL();
3550                RRETURN(MATCH_NOMATCH);
3551                }
3552            switch(*eptr++)            switch(*eptr++)
3553              {              {
3554              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
# Line 3117  for (;;) Line 3556  for (;;)
3556              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3557              break;              break;
3558              case 0x000a:              case 0x000a:
3559                break;
3560    
3561              case 0x000b:              case 0x000b:
3562              case 0x000c:              case 0x000c:
3563              case 0x0085:              case 0x0085:
3564                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3565              break;              break;
3566              }              }
3567            }            }
# Line 3128  for (;;) Line 3570  for (;;)
3570          case OP_NOT_HSPACE:          case OP_NOT_HSPACE:
3571          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3572            {            {
3573            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3574                {
3575                SCHECK_PARTIAL();
3576                RRETURN(MATCH_NOMATCH);
3577                }
3578            switch(*eptr++)            switch(*eptr++)
3579              {              {
3580              default: break;              default: break;
# Line 3143  for (;;) Line 3589  for (;;)
3589          case OP_HSPACE:          case OP_HSPACE:
3590          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3591            {            {
3592            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3593                {
3594                SCHECK_PARTIAL();
3595                RRETURN(MATCH_NOMATCH);
3596                }
3597            switch(*eptr++)            switch(*eptr++)
3598              {              {
3599              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
3600              case 0x09:      /* HT */              case 0x09:      /* HT */
3601              case 0x20:      /* SPACE */              case 0x20:      /* SPACE */
3602              case 0xa0:      /* NBSP */              case 0xa0:      /* NBSP */
3603              break;              break;
3604              }              }
3605            }            }
3606          break;          break;
# Line 3158  for (;;) Line 3608  for (;;)
3608          case OP_NOT_VSPACE:          case OP_NOT_VSPACE:
3609          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3610            {            {
3611            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3612                {
3613                SCHECK_PARTIAL();
3614                RRETURN(MATCH_NOMATCH);
3615                }
3616            switch(*eptr++)            switch(*eptr++)
3617              {              {
3618              default: break;              default: break;
# Line 3175  for (;;) Line 3629  for (;;)
3629          case OP_VSPACE:          case OP_VSPACE:
3630          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3631            {            {
3632            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3633                {
3634                SCHECK_PARTIAL();
3635                RRETURN(MATCH_NOMATCH);
3636                }
3637            switch(*eptr++)            switch(*eptr++)
3638              {              {
3639              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
# Line 3184  for (;;) Line 3642  for (;;)
3642              case 0x0c:      /* FF */              case 0x0c:      /* FF */
3643              case 0x0d:      /* CR */              case 0x0d:      /* CR */
3644              case 0x85:      /* NEL */              case 0x85:      /* NEL */
3645              break;              break;
3646              }              }
3647            }            }
3648          break;          break;
3649    
3650          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
3651          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3652              {
3653              if (eptr >= md->end_subject)
3654                {
3655                SCHECK_PARTIAL();
3656                RRETURN(MATCH_NOMATCH);
3657                }
3658            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);
3659              }
3660          break;          break;
3661    
3662          case OP_DIGIT:          case OP_DIGIT:
3663          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3664              {
3665              if (eptr >= md->end_subject)
3666                {
3667                SCHECK_PARTIAL();
3668                RRETURN(MATCH_NOMATCH);
3669                }
3670            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);
3671              }
3672          break;          break;
3673    
3674          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
3675          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3676              {
3677              if (eptr >= md->end_subject)
3678                {
3679                SCHECK_PARTIAL();
3680                RRETURN(MATCH_NOMATCH);
3681                }
3682            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);
3683              }
3684          break;          break;
3685    
3686          case OP_WHITESPACE:          case OP_WHITESPACE:
3687          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3688              {
3689              if (eptr >= md->end_subject)
3690                {
3691                SCHECK_PARTIAL();
3692                RRETURN(MATCH_NOMATCH);
3693                }
3694            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);
3695              }
3696          break;          break;
3697    
3698          case OP_NOT_WORDCHAR:          case OP_NOT_WORDCHAR:
3699          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3700              {
3701              if (eptr >= md->end_subject)
3702                {
3703                SCHECK_PARTIAL();
3704                RRETURN(MATCH_NOMATCH);
3705                }
3706            if ((md->ctypes[*eptr++] & ctype_word) != 0)            if ((md->ctypes[*eptr++] & ctype_word) != 0)
3707              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3708              }
3709          break;          break;
3710    
3711          case OP_WORDCHAR:          case OP_WORDCHAR:
3712          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3713              {
3714              if (eptr >= md->end_subject)
3715                {
3716                SCHECK_PARTIAL();
3717                RRETURN(MATCH_NOMATCH);
3718                }
3719            if ((md->ctypes[*eptr++] & ctype_word) == 0)            if ((md->ctypes[*eptr++] & ctype_word) == 0)
3720              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3721              }
3722          break;          break;
3723    
3724          default:          default:
# Line 3246  for (;;) Line 3746  for (;;)
3746              {              {
3747              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);
3748              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3749              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3750                if (eptr >= md->end_subject)
3751                  {
3752                  SCHECK_PARTIAL();
3753                  RRETURN(MATCH_NOMATCH);
3754                  }
3755              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3756              if (prop_fail_result) RRETURN(MATCH_NOMATCH);              if (prop_fail_result) RRETURN(MATCH_NOMATCH);
3757              }              }
# Line 3257  for (;;) Line 3762  for (;;)
3762              {              {
3763              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);
3764              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3765              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3766                if (eptr >= md->end_subject)
3767                  {
3768                  SCHECK_PARTIAL();
3769                  RRETURN(MATCH_NOMATCH);
3770                  }
3771              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3772              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
3773              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
3774                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
3775                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 3272  for (;;) Line 3782  for (;;)
3782              {              {
3783              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);
3784              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3785              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3786                if (eptr >= md->end_subject)
3787                  {
3788                  SCHECK_PARTIAL();
3789                  RRETURN(MATCH_NOMATCH);
3790                  }
3791              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3792              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
3793              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
3794                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3795              }              }
# Line 3285  for (;;) Line 3800  for (;;)
3800              {              {
3801              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);
3802              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3803              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3804                if (eptr >= md->end_subject)
3805                  {
3806                  SCHECK_PARTIAL();
3807                  RRETURN(MATCH_NOMATCH);
3808                  }
3809              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3810              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
3811              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
3812                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3813              }              }
# Line 3298  for (;;) Line 3818  for (;;)
3818              {              {
3819              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);
3820              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3821              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3822                if (eptr >= md->end_subject)
3823                  {
3824                  SCHECK_PARTIAL();
3825                  RRETURN(MATCH_NOMATCH);
3826                  }
3827              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3828              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_script = UCD_SCRIPT(c);
3829              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
3830                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3831              }              }
# Line 3320  for (;;) Line 3845  for (;;)
3845            {            {
3846            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);
3847            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3848            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3849              if (eptr >= md->end_subject)
3850                {
3851                SCHECK_PARTIAL();
3852                RRETURN(MATCH_NOMATCH);
3853                }
3854            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3855            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
3856            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);
3857            while (eptr < md->end_subject)            while (eptr < md->end_subject)
3858              {              {
3859              int len = 1;              int len = 1;
3860              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
3861                {                else { GETCHARLEN(c, eptr, len); }
3862                GETCHARLEN(c, eptr, len);              prop_category = UCD_CATEGORY(c);
               }  
             prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);  
3863              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
3864              eptr += len;              eptr += len;
3865              }              }
# Line 3349  for (;;) Line 3877  for (;;)
3877            {            {
3878            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);
3879            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3880            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) RRETURN(MATCH_NOMATCH);
3881                 (ctype == OP_ANY && (ims & PCRE_DOTALL) == 0 &&            if (eptr >= md->end_subject)
3882                  IS_NEWLINE(eptr)))              {
3883                SCHECK_PARTIAL();
3884                RRETURN(MATCH_NOMATCH);
3885                }
3886              if (ctype == OP_ANY && IS_NEWLINE(eptr))
3887              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
   
3888            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3889            switch(ctype)            switch(ctype)
3890              {              {
3891              case OP_ANY:        /* This is the DOTALL case */              case OP_ANY:        /* This is the non-NL case */
3892              break;              case OP_ALLANY:
   
3893              case OP_ANYBYTE:              case OP_ANYBYTE:
3894              break;              break;
3895    
# Line 3371  for (;;) Line 3901  for (;;)
3901                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3902                break;                break;
3903                case 0x000a:                case 0x000a:
3904                  break;
3905    
3906                case 0x000b:                case 0x000b:
3907                case 0x000c:                case 0x000c:
3908                case 0x0085:                case 0x0085:
3909                case 0x2028:                case 0x2028:
3910                case 0x2029:                case 0x2029:
3911                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3912                break;                break;
3913                }                }
3914              break;              break;
# Line 3507  for (;;) Line 4040  for (;;)
4040            {            {
4041            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);
4042            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4043            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) RRETURN(MATCH_NOMATCH);
4044                 ((ims & PCRE_DOTALL) == 0 && IS_NEWLINE(eptr)))            if (eptr >= md->end_subject)
4045                {
4046                SCHECK_PARTIAL();
4047                RRETURN(MATCH_NOMATCH);
4048                }
4049              if (ctype == OP_ANY && IS_NEWLINE(eptr))
4050              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
   
4051            c = *eptr++;            c = *eptr++;
4052            switch(ctype)            switch(ctype)
4053              {              {
4054              case OP_ANY:   /* This is the DOTALL case */              case OP_ANY:     /* This is the non-NL case */
4055              break;              case OP_ALLANY:
   
4056              case OP_ANYBYTE:              case OP_ANYBYTE:
4057              break;              break;
4058    
# Line 3527  for (;;) Line 4063  for (;;)
4063                case 0x000d:                case 0x000d:
4064                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
4065                break;                break;
4066    
4067                case 0x000a:                case 0x000a:
4068                  break;
4069    
4070                case 0x000b:                case 0x000b:
4071                case 0x000c:                case 0x000c:
4072                case 0x0085:                case 0x0085:
4073                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4074                break;                break;
4075                }                }
4076              break;              break;
# Line 3645  for (;;) Line 4185  for (;;)
4185              int len = 1;              int len = 1;
4186              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4187              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4188              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
4189              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
4190                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
4191                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 3660  for (;;) Line 4200  for (;;)
4200              int len = 1;              int len = 1;
4201              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4202              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4203              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4204              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
4205                break;                break;
4206              eptr+= len;              eptr+= len;
# Line 3673  for (;;) Line 4213  for (;;)
4213              int len = 1;              int len = 1;
4214              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4215              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4216              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
4217              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
4218                break;                break;
4219              eptr+= len;              eptr+= len;
# Line 3686  for (;;) Line 4226  for (;;)
4226              int len = 1;              int len = 1;
4227              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4228              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4229              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_script = UCD_SCRIPT(c);
4230              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
4231                break;                break;
4232              eptr+= len;              eptr+= len;
# Line 3702  for (;;) Line 4242  for (;;)
4242            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);
4243            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4244            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
4245            BACKCHAR(eptr);            if (utf8) BACKCHAR(eptr);
4246            }            }
4247          }          }
4248    
# Line 3715  for (;;) Line 4255  for (;;)
4255            {            {
4256            if (eptr >= md->end_subject) break;            if (eptr >= md->end_subject) break;
4257            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
4258            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
4259            if (prop_category == ucp_M) break;            if (prop_category == ucp_M) break;
4260            while (eptr < md->end_subject)            while (eptr < md->end_subject)
4261              {              {
# Line 3724  for (;;) Line 4264  for (;;)
4264                {                {
4265                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
4266                }                }
4267              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4268              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
4269              eptr += len;              eptr += len;
4270              }              }
# Line 3741  for (;;) Line 4281  for (;;)
4281            for (;;)                        /* Move back over one extended */            for (;;)                        /* Move back over one extended */
4282              {              {
4283              int len = 1;              int len = 1;
             BACKCHAR(eptr);  
4284              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr; else
4285                {                {
4286                  BACKCHAR(eptr);
4287                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
4288                }                }
4289              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4290              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
4291              eptr--;              eptr--;
4292              }              }
# Line 3764  for (;;) Line 4304  for (;;)
4304          switch(ctype)          switch(ctype)
4305            {            {
4306            case OP_ANY:            case OP_ANY:
   
           /* Special code is required for UTF8, but when the maximum is  
           unlimited we don't need it, so we repeat the non-UTF8 code. This is  
           probably worth it, because .* is quite a common idiom. */  
   
4307            if (max < INT_MAX)            if (max < INT_MAX)
4308              {              {
4309              if ((ims & PCRE_DOTALL) == 0)              for (i = min; i < max; i++)
               {  
               for (i = min; i < max; i++)  
                 {  
                 if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;  
                 eptr++;  
                 while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  
                 }  
               }  
             else  
4310                {                {
4311                for (i = min; i < max; i++)                if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
4312                  {                eptr++;
4313                  if (eptr >= md->end_subject) break;                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
                 eptr++;  
                 while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  
                 }  
4314                }                }
4315              }              }
4316    
# Line 3795  for (;;) Line 4318  for (;;)
4318    
4319            else            else
4320              {              {
4321              if ((ims & PCRE_DOTALL) == 0)              for (i = min; i < max; i++)
4322                {                {
4323                for (i = min; i < max; i++)                if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
4324                  {                eptr++;
4325                  if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
                 eptr++;  
                 }  
               break;  
4326                }                }
4327              else              }
4328              break;
4329    
4330              case OP_ALLANY:
4331              if (max < INT_MAX)
4332                {
4333                for (i = min; i < max; i++)
4334                {                {
4335                c = max - min;                if (eptr >= md->end_subject) break;
4336                if (c > (unsigned int)(md->end_subject - eptr))                eptr++;
4337                  c = md->end_subject - eptr;                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
               eptr += c;  
4338                }                }
4339              }              }
4340              else eptr = md->end_subject;   /* Unlimited UTF-8 repeat */
4341            break;            break;
4342    
4343            /* The byte case is the same as non-UTF8 */            /* The byte case is the same as non-UTF8 */
# Line 3836  for (;;) Line 4362  for (;;)
4362                }                }
4363              else              else
4364                {                {
4365                if (c != 0x000a && c != 0x000b && c != 0x000c &&                if (c != 0x000a &&
4366                    c != 0x0085 && c != 0x2028 && c != 0x2029)                    (md->bsr_anycrlf ||
4367                       (c != 0x000b && c != 0x000c &&
4368                        c != 0x0085 && c != 0x2028 && c != 0x2029)))
4369                  break;                  break;
4370                eptr += len;                eptr += len;
4371                }                }
# Line 3845  for (;;) Line 4373  for (;;)
4373            break;            break;
4374    
4375            case OP_NOT_HSPACE:            case OP_NOT_HSPACE:
4376            case OP_HSPACE:            case OP_HSPACE:
4377            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4378              {              {
4379              BOOL gotspace;              BOOL gotspace;
4380              int len = 1;              int len = 1;
4381              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4382              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4383              switch(c)              switch(c)
4384                {                {
4385                default: gotspace = FALSE; break;                default: gotspace = FALSE; break;
4386                case 0x09:      /* HT */                case 0x09:      /* HT */
4387                case 0x20:      /* SPACE */                case 0x20:      /* SPACE */
4388                case 0xa0:      /* NBSP */                case 0xa0:      /* NBSP */
# Line 3875  for (;;) Line 4403  for (;;)
4403                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4404                case 0x3000:    /* IDEOGRAPHIC SPACE */                case 0x3000:    /* IDEOGRAPHIC SPACE */
4405                gotspace = TRUE;                gotspace = TRUE;
4406                break;                break;
4407                }                }
4408              if (gotspace == (ctype == OP_NOT_HSPACE)) break;              if (gotspace == (ctype == OP_NOT_HSPACE)) break;
4409              eptr += len;              eptr += len;
# Line 3883  for (;;) Line 4411  for (;;)
4411            break;            break;
4412    
4413            case OP_NOT_VSPACE:            case OP_NOT_VSPACE:
4414            case OP_VSPACE:            case OP_VSPACE:
4415            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4416              {              {
4417              BOOL gotspace;              BOOL gotspace;
4418              int len = 1;              int len = 1;
4419              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4420              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4421              switch(c)              switch(c)
4422                {                {
4423                default: gotspace = FALSE; break;                default: gotspace = FALSE; break;
4424                case 0x0a:      /* LF */                case 0x0a:      /* LF */
4425                case 0x0b:      /* VT */                case 0x0b:      /* VT */
4426                case 0x0c:      /* FF */                case 0x0c:      /* FF */
# Line 3903  for (;;) Line 4431  for (;;)
4431                gotspace = TRUE;                gotspace = TRUE;
4432                break;                break;
4433                }                }
4434              if (gotspace == (ctype == OP_NOT_VSPACE)) break;              if (gotspace == (ctype == OP_NOT_VSPACE)) break;
4435              eptr += len;              eptr += len;
4436              }              }
4437            break;            break;
# Line 3990  for (;;) Line 4518  for (;;)
4518            }            }
4519          }          }
4520        else        else
4521  #endif  #endif  /* SUPPORT_UTF8 */
4522    
4523        /* Not UTF-8 mode */        /* Not UTF-8 mode */
4524          {          {
4525          switch(ctype)          switch(ctype)
4526            {            {
4527            case OP_ANY:            case OP_ANY:
4528            if ((ims & PCRE_DOTALL) == 0)            for (i = min; i < max; i++)
4529              {              {
4530              for (i = min; i < max; i++)              if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
4531                {              eptr++;
               if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;  
               eptr++;  
               }  
             break;  
4532              }              }
4533            /* For DOTALL case, fall through and treat as \C */            break;
4534    
4535              case OP_ALLANY:
4536            case OP_ANYBYTE:            case OP_ANYBYTE:
4537            c = max - min;            c = max - min;
4538            if (c > (unsigned int)(md->end_subject - eptr))            if (c > (unsigned int)(md->end_subject - eptr))
# Line 4027  for (;;) Line 4552  for (;;)
4552                }                }
4553              else              else
4554                {                {
4555                if (c != 0x000a && c != 0x000b && c != 0x000c && c != 0x0085)                if (c != 0x000a &&
4556                      (md->bsr_anycrlf ||
4557                        (c != 0x000b && c != 0x000c && c != 0x0085)))
4558                  break;                  break;
4559                eptr++;                eptr++;
4560                }                }
# Line 4040  for (;;) Line 4567  for (;;)
4567              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4568              c = *eptr;              c = *eptr;
4569              if (c == 0x09 || c == 0x20 || c == 0xa0) break;              if (c == 0x09 || c == 0x20 || c == 0xa0) break;
4570              eptr++;              eptr++;
4571              }              }
4572            break;            break;
4573    
# Line 4050  for (;;) Line 4577  for (;;)
4577              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4578              c = *eptr;              c = *eptr;
4579              if (c != 0x09 && c != 0x20 && c != 0xa0) break;              if (c != 0x09 && c != 0x20 && c != 0xa0) break;
4580              eptr++;              eptr++;
4581              }              }
4582            break;            break;
4583    
# Line 4061  for (;;) Line 4588  for (;;)
4588              c = *eptr;              c = *eptr;
4589              if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)              if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)
4590                break;                break;
4591              eptr++;              eptr++;
4592              }              }
4593            break;            break;
4594    
# Line 4177  HEAP_RETURN: Line 4704  HEAP_RETURN:
4704  switch (frame->Xwhere)  switch (frame->Xwhere)
4705    {    {
4706    LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8)    LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8)
4707    LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(16)    LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(17)
4708    LBL(17) LBL(18) LBL(19) LBL(20) LBL(21) LBL(22) LBL(23) LBL(24)    LBL(19) LBL(24) LBL(25) LBL(26) LBL(27) LBL(29) LBL(31) LBL(33)
4709    LBL(25) LBL(26) LBL(27) LBL(28) LBL(29) LBL(30) LBL(31) LBL(32)    LBL(35) LBL(43) LBL(47) LBL(48) LBL(49) LBL(50) LBL(51) LBL(52)
4710    LBL(33) LBL(34) LBL(35) LBL(36) LBL(37) LBL(38) LBL(39) LBL(40)    LBL(53) LBL(54)
4711    LBL(41) LBL(42) LBL(43) LBL(44) LBL(45) LBL(46) LBL(47)  #ifdef SUPPORT_UTF8
4712      LBL(16) LBL(18) LBL(20) LBL(21) LBL(22) LBL(23) LBL(28) LBL(30)
4713      LBL(32) LBL(34) LBL(42) LBL(46)
4714    #ifdef SUPPORT_UCP
4715      LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
4716    #endif  /* SUPPORT_UCP */
4717    #endif  /* SUPPORT_UTF8 */
4718    default:    default:
4719    DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));    DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));
4720    return PCRE_ERROR_INTERNAL;    return PCRE_ERROR_INTERNAL;
# Line 4273  Returns:          > 0 => success; value Line 4806  Returns:          > 0 => success; value
4806                   < -1 => some kind of unexpected problem                   < -1 => some kind of unexpected problem
4807  */  */
4808    
4809  PCRE_EXP_DEFN int  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
4810  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,
4811    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,
4812    int offsetcount)    int offsetcount)
# Line 4297  const uschar *tables; Line 4830  const uschar *tables;
4830  const uschar *start_bits = NULL;  const uschar *start_bits = NULL;
4831  USPTR start_match = (USPTR)subject + start_offset;  USPTR start_match = (USPTR)subject + start_offset;
4832  USPTR end_subject;  USPTR end_subject;
4833    USPTR start_partial = NULL;
4834  USPTR req_byte_ptr = start_match - 1;  USPTR req_byte_ptr = start_match - 1;
 eptrblock eptrchain[EPTR_WORK_SIZE];  
4835    
4836  pcre_study_data internal_study;  pcre_study_data internal_study;
4837  const pcre_study_data *study;  const pcre_study_data *study;
# Line 4361  if (re->magic_number != MAGIC_NUMBER) Line 4894  if (re->magic_number != MAGIC_NUMBER)
4894  /* Set up other data */  /* Set up other data */
4895    
4896  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
4897  startline = (re->options & PCRE_STARTLINE) != 0;  startline = (re->flags & PCRE_STARTLINE) != 0;
4898  firstline = (re->options & PCRE_FIRSTLINE) != 0;  firstline = (re->options & PCRE_FIRSTLINE) != 0;
4899    
4900  /* The code starts after the real_pcre block and the capture name table. */  /* The code starts after the real_pcre block and the capture name table. */
# Line 4376  end_subject = md->end_subject; Line 4909  end_subject = md->end_subject;
4909    
4910  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;
4911  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;
4912    md->jscript_compat = (re->options & PCRE_JAVASCRIPT_COMPAT) != 0;
4913    
4914  md->notbol = (options & PCRE_NOTBOL) != 0;  md->notbol = (options & PCRE_NOTBOL) != 0;
4915  md->noteol = (options & PCRE_NOTEOL) != 0;  md->noteol = (options & PCRE_NOTEOL) != 0;
4916  md->notempty = (options & PCRE_NOTEMPTY) != 0;  md->notempty = (options & PCRE_NOTEMPTY) != 0;
4917  md->partial = (options & PCRE_PARTIAL) != 0;  md->partial = ((options & PCRE_PARTIAL_HARD) != 0)? 2 :
4918                  ((options & PCRE_PARTIAL_SOFT) != 0)? 1 : 0;
4919  md->hitend = FALSE;  md->hitend = FALSE;
4920    
4921  md->recursive = NULL;                   /* No recursion at top level */  md->recursive = NULL;                   /* No recursion at top level */
 md->eptrchain = eptrchain;              /* Make workspace generally available */  
4922    
4923  md->lcc = tables + lcc_offset;  md->lcc = tables + lcc_offset;
4924  md->ctypes = tables + ctypes_offset;  md->ctypes = tables + ctypes_offset;
4925    
4926    /* Handle different \R options. */
4927    
4928    switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
4929      {
4930      case 0:
4931      if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
4932        md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0;
4933      else
4934    #ifdef BSR_ANYCRLF
4935      md->bsr_anycrlf = TRUE;
4936    #else
4937      md->bsr_anycrlf = FALSE;
4938    #endif
4939      break;
4940    
4941      case PCRE_BSR_ANYCRLF:
4942      md->bsr_anycrlf = TRUE;
4943      break;
4944    
4945      case PCRE_BSR_UNICODE:
4946      md->bsr_anycrlf = FALSE;
4947      break;
4948    
4949      default: return PCRE_ERROR_BADNEWLINE;
4950      }
4951    
4952  /* Handle different types of newline. The three bits give eight cases. If  /* Handle different types of newline. The three bits give eight cases. If
4953  nothing is set at run time, whatever was used at compile time applies. */  nothing is set at run time, whatever was used at compile time applies. */
4954    
4955  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
4956         PCRE_NEWLINE_BITS)          (pcre_uint32)options) & PCRE_NEWLINE_BITS)
4957    {    {
4958    case 0: newline = NEWLINE; break;   /* Compile-time default */    case 0: newline = NEWLINE; break;   /* Compile-time default */
4959    case PCRE_NEWLINE_CR: newline = '\r'; break;    case PCRE_NEWLINE_CR: newline = CHAR_CR; break;
4960    case PCRE_NEWLINE_LF: newline = '\n'; break;    case PCRE_NEWLINE_LF: newline = CHAR_NL; break;
4961    case PCRE_NEWLINE_CR+    case PCRE_NEWLINE_CR+
4962         PCRE_NEWLINE_LF: newline = ('\r' << 8) | '\n'; break;         PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break;
4963    case PCRE_NEWLINE_ANY: newline = -1; break;    case PCRE_NEWLINE_ANY: newline = -1; break;
4964    case PCRE_NEWLINE_ANYCRLF: newline = -2; break;    case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
4965    default: return PCRE_ERROR_BADNEWLINE;    default: return PCRE_ERROR_BADNEWLINE;
# Line 4429  else Line 4989  else
4989      }      }
4990    }    }
4991    
4992  /* Partial matching is supported only for a restricted set of regexes at the  /* Partial matching was originally supported only for a restricted set of
4993  moment. */  regexes; from release 8.00 there are no restrictions, but the bits are still
4994    defined (though never set). So there's no harm in leaving this code. */
4995    
4996  if (md->partial && (re->options & PCRE_NOPARTIAL) != 0)  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
4997    return PCRE_ERROR_BADPARTIAL;    return PCRE_ERROR_BADPARTIAL;
4998    
4999  /* Check a UTF-8 string if required. Unfortunately there's no way of passing  /* Check a UTF-8 string if required. Unfortunately there's no way of passing
# Line 4441  back the character offset. */ Line 5002  back the character offset. */
5002  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
5003  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)
5004    {    {
5005    if (_pcre_valid_utf8((uschar *)subject, length) >= 0)    if (_pcre_valid_utf8((USPTR)subject, length) >= 0)
5006      return PCRE_ERROR_BADUTF8;      return PCRE_ERROR_BADUTF8;
5007    if (start_offset > 0 && start_offset < length)    if (start_offset > 0 && start_offset < length)
5008      {      {
5009      int tb = ((uschar *)subject)[start_offset];      int tb = ((USPTR)subject)[start_offset];
5010      if (tb > 127)      if (tb > 127)
5011        {        {
5012        tb &= 0xc0;        tb &= 0xc0;
# Line 4509  studied, there may be a bitmap of possib Line 5070  studied, there may be a bitmap of possib
5070    
5071  if (!anchored)  if (!anchored)
5072    {    {
5073    if ((re->options & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
5074      {      {
5075      first_byte = re->first_byte & 255;      first_byte = re->first_byte & 255;
5076      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)
# Line 4524  if (!anchored) Line 5085  if (!anchored)
5085  /* For anchored or unanchored matches, there may be a "last known required  /* For anchored or unanchored matches, there may be a "last known required
5086  character" set. */  character" set. */
5087    
5088  if ((re->options & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
5089    {    {
5090    req_byte = re->req_byte & 255;    req_byte = re->req_byte & 255;
5091    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;
# Line 4540  the loop runs just once. */ Line 5101  the loop runs just once. */
5101  for(;;)  for(;;)
5102    {    {
5103    USPTR save_end_subject = end_subject;    USPTR save_end_subject = end_subject;
5104      USPTR new_start_match;
5105    
5106    /* Reset the maximum number of extractions we might see. */    /* Reset the maximum number of extractions we might see. */
5107    
# Line 4550  for(;;) Line 5112  for(;;)
5112      while (iptr < iend) *iptr++ = -1;      while (iptr < iend) *iptr++ = -1;
5113      }      }
5114    
5115    /* Advance to a unique first char if possible. If firstline is TRUE, the    /* If firstline is TRUE, the start of the match is constrained to the first
5116    start of the match is constrained to the first line of a multiline string.    line of a multiline string. That is, the match must be before or at the first
5117    That is, the match must be before or at the first newline. Implement this by    newline. Implement this by temporarily adjusting end_subject so that we stop
5118    temporarily adjusting end_subject so that we stop scanning at a newline. If    scanning at a newline. If the match fails at the newline, later code breaks
5119    the match fails at the newline, later code breaks this loop. */    this loop. */
5120    
5121    if (firstline)    if (firstline)
5122      {      {
5123      USPTR t = start_match;      USPTR t = start_match;
5124    #ifdef SUPPORT_UTF8
5125        if (utf8)
5126          {
5127          while (t < md->end_subject && !IS_NEWLINE(t))
5128            {
5129            t++;
5130            while (t < end_subject && (*t & 0xc0) == 0x80) t++;
5131            }
5132          }
5133        else
5134    #endif
5135      while (t < md->end_subject && !IS_NEWLINE(t)) t++;      while (t < md->end_subject && !IS_NEWLINE(t)) t++;
5136      end_subject = t;      end_subject = t;
5137      }      }
5138    
5139    /* Now test for a unique first byte */    /* There are some optimizations that avoid running the match if a known
5140      starting point is not found, or if a known later character is not present.
5141      However, there is an option that disables these, for testing and for ensuring
5142      that all callouts do actually occur. */
5143    
5144    if (first_byte >= 0)    if ((options & PCRE_NO_START_OPTIMIZE) == 0)
5145      {      {
5146      if (first_byte_caseless)      /* Advance to a unique first byte if there is one. */
       while (start_match < end_subject &&  
              md->lcc[*start_match] != first_byte)  
         start_match++;  
     else  
       while (start_match < end_subject && *start_match != first_byte)  
         start_match++;  
     }  
5147    
5148    /* Or to just after a linebreak for a multiline match if possible */      if (first_byte >= 0)
5149          {
5150          if (first_byte_caseless)
5151            while (start_match < end_subject && md->lcc[*start_match] != first_byte)
5152              start_match++;
5153          else
5154            while (start_match < end_subject && *start_match != first_byte)
5155              start_match++;
5156          }
5157    
5158    else if (startline)      /* Or to just after a linebreak for a multiline match */
5159      {  
5160      if (start_match > md->start_subject + start_offset)      else if (startline)
5161        {        {
5162        while (start_match <= end_subject && !WAS_NEWLINE(start_match))        if (start_match > md->start_subject + start_offset)
5163          start_match++;          {
5164    #ifdef SUPPORT_UTF8
5165            if (utf8)
5166              {
5167              while (start_match < end_subject && !WAS_NEWLINE(start_match))
5168                {
5169                start_match++;
5170                while(start_match < end_subject && (*start_match & 0xc0) == 0x80)
5171                  start_match++;
5172                }
5173              }
5174            else
5175    #endif
5176            while (start_match < end_subject && !WAS_NEWLINE(start_match))
5177              start_match++;
5178    
5179            /* If we have just passed a CR and the newline option is ANY or ANYCRLF,
5180            and we are now at a LF, advance the match position by one more character.
5181            */
5182    
5183        /* If we have just passed a CR and the newline option is ANY or ANYCRLF,          if (start_match[-1] == CHAR_CR &&
5184        and we are now at a LF, advance the match position by one more character.               (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
5185        */               start_match < end_subject &&
5186                 *start_match == CHAR_NL)
5187        if (start_match[-1] == '\r' &&            start_match++;
5188             (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&          }
            start_match < end_subject &&  
            *start_match == '\n')  
         start_match++;  
5189        }        }
     }  
5190    
5191    /* Or to a non-unique first char after study */      /* Or to a non-unique first byte after study */
5192    
5193    else if (start_bits != NULL)      else if (start_bits != NULL)
     {  
     while (start_match < end_subject)  
5194        {        {
5195        register unsigned int c = *start_match;        while (start_match < end_subject)
5196        if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; else break;          {
5197            register unsigned int c = *start_match;
5198            if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++;
5199              else break;
5200            }
5201        }        }
5202      }      }   /* Starting optimizations */
5203    
5204    /* Restore fudged end_subject */    /* Restore fudged end_subject */
5205    
# Line 4618  for(;;) Line 5211  for(;;)
5211    printf("\n");    printf("\n");
5212  #endif  #endif
5213    
5214    /* If req_byte is set, we know that that character must appear in the subject    /* If req_byte is set, we know that that character must appear in the
5215    for the match to succeed. If the first character is set, req_byte must be    subject for the match to succeed. If the first character is set, req_byte
5216    later in the subject; otherwise the test starts at the match point. This    must be later in the subject; otherwise the test starts at the match point.
5217    optimization can save a huge amount of backtracking in patterns with nested    This optimization can save a huge amount of backtracking in patterns with
5218    unlimited repeats that aren't going to match. Writing separate code for    nested unlimited repeats that aren't going to match. Writing separate code
5219    cased/caseless versions makes it go faster, as does using an autoincrement    for cased/caseless versions makes it go faster, as does using an
5220    and backing off on a match.    autoincrement and backing off on a match.
5221    
5222    HOWEVER: when the subject string is very, very long, searching to its end can    HOWEVER: when the subject string is very, very long, searching to its end
5223    take a long time, and give bad performance on quite ordinary patterns. This    can take a long time, and give bad performance on quite ordinary patterns.
5224    showed up when somebody was matching something like /^\d+C/ on a 32-megabyte    This showed up when somebody was matching something like /^\d+C/ on a
5225    string... so we don't do this when the string is sufficiently long.    32-megabyte string... so we don't do this when the string is sufficiently
5226      long.
5227    
5228    ALSO: this processing is disabled when partial matching is requested.    ALSO: this processing is disabled when partial matching is requested, or if
5229    */    disabling is explicitly requested. */
5230    
5231    if (req_byte >= 0 &&    if ((options & PCRE_NO_START_OPTIMIZE) == 0 &&
5232          req_byte >= 0 &&
5233        end_subject - start_match < REQ_BYTE_MAX &&        end_subject - start_match < REQ_BYTE_MAX &&
5234        !md->partial)        !md->partial)
5235      {      {
# Line 4678  for(;;) Line 5273  for(;;)
5273        }        }
5274      }      }
5275    
5276    /* OK, we can now run the match. */    /* OK, we can now run the match. If "hitend" is set afterwards, remember the
5277      first starting point for which a partial match was found. */
5278    
5279    md->start_match_ptr = start_match;      /* Insurance */    md->start_match_ptr = start_match;
5280    md->match_call_count = 0;    md->match_call_count = 0;
5281    md->eptrn = 0;                          /* Next free eptrchain slot */    rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0);
5282    rc = match(start_match, md->start_code, start_match, 2, md,    if (md->hitend && start_partial == NULL) start_partial = start_match;
5283      ims, NULL, 0, 0);  
5284      switch(rc)
5285        {
5286        /* NOMATCH and PRUNE advance by one character. THEN at this level acts
5287        exactly like PRUNE. */
5288    
5289        case MATCH_NOMATCH:
5290        case MATCH_PRUNE:
5291        case MATCH_THEN:
5292        new_start_match = start_match + 1;
5293    #ifdef SUPPORT_UTF8
5294        if (utf8)
5295          while(new_start_match < end_subject && (*new_start_match & 0xc0) == 0x80)
5296            new_start_match++;
5297    #endif
5298        break;
5299    
5300    /* Any return other than MATCH_NOMATCH breaks the loop. */      /* SKIP passes back the next starting point explicitly. */
5301    
5302        case MATCH_SKIP:
5303        new_start_match = md->start_match_ptr;
5304        break;
5305    
5306    if (rc != MATCH_NOMATCH) break;      /* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */
5307    
5308        case MATCH_COMMIT:
5309        rc = MATCH_NOMATCH;
5310        goto ENDLOOP;
5311    
5312        /* Any other return is some kind of error. */
5313    
5314        default:
5315        goto ENDLOOP;
5316        }
5317    
5318      /* Control reaches here for the various types of "no match at this point"
5319      result. Reset the code to MATCH_NOMATCH for subsequent checking. */
5320    
5321      rc = MATCH_NOMATCH;
5322    
5323    /* If PCRE_FIRSTLINE is set, the match must happen before or at the first    /* If PCRE_FIRSTLINE is set, the match must happen before or at the first
5324    newline in the subject (though it may continue over the newline). Therefore,    newline in the subject (though it may continue over the newline). Therefore,
# Line 4696  for(;;) Line 5326  for(;;)
5326    
5327    if (firstline && IS_NEWLINE(start_match)) break;    if (firstline && IS_NEWLINE(start_match)) break;
5328    
5329    /* Advance the match position by one character. */    /* Advance to new matching position */
5330    
5331    start_match++;    start_match = new_start_match;
 #ifdef SUPPORT_UTF8  
   if (utf8)  
     while(start_match < end_subject && (*start_match & 0xc0) == 0x80)  
       start_match++;  
 #endif  
5332    
5333    /* Break the loop if the pattern is anchored or if we have passed the end of    /* Break the loop if the pattern is anchored or if we have passed the end of
5334    the subject. */    the subject. */
5335    
5336    if (anchored || start_match > end_subject) break;    if (anchored || start_match > end_subject) break;
5337    
5338    /* If we have just passed a CR and the newline option is CRLF or ANY or    /* If we have just passed a CR and we are now at a LF, and the pattern does
5339    ANYCRLF, and we are now at a LF, advance the match position by one more    not contain any explicit matches for \r or \n, and the newline option is CRLF
5340    character. */    or ANY or ANYCRLF, advance the match position by one more character. */
5341    
5342    if (start_match[-1] == '\r' &&    if (start_match[-1] == CHAR_CR &&
5343         (md->nltype == NLTYPE_ANY ||        start_match < end_subject &&
5344          md->nltype == NLTYPE_ANYCRLF ||        *start_match == CHAR_NL &&
5345          md->nllen == 2) &&        (re->flags & PCRE_HASCRORLF) == 0 &&
5346         start_match < end_subject &&          (md->nltype == NLTYPE_ANY ||
5347         *start_match == '\n')           md->nltype == NLTYPE_ANYCRLF ||
5348             md->nllen == 2))
5349      start_match++;      start_match++;
5350    
5351    }   /* End of for(;;) "bumpalong" loop */    }   /* End of for(;;) "bumpalong" loop */
# Line 4729  for(;;) Line 5355  for(;;)
5355  /* We reach here when rc is not MATCH_NOMATCH, or if one of the stopping  /* We reach here when rc is not MATCH_NOMATCH, or if one of the stopping
5356  conditions is true:  conditions is true:
5357    
5358  (1) The pattern is anchored;  (1) The pattern is anchored or the match was failed by (*COMMIT);
5359    
5360  (2) We are past the end of the subject;  (2) We are past the end of the subject;
5361    
# Line 4744  processing, copy those that we can. In t Line 5370  processing, copy those that we can. In t
5370  certain parts of the pattern were not used, even though there are more  certain parts of the pattern were not used, even though there are more
5371  capturing parentheses than vector slots. */  capturing parentheses than vector slots. */
5372    
5373    ENDLOOP:
5374    
5375  if (rc == MATCH_MATCH)  if (rc == MATCH_MATCH)
5376    {    {
5377    if (using_temporary_offsets)    if (using_temporary_offsets)
# Line 4787  if (using_temporary_offsets) Line 5415  if (using_temporary_offsets)
5415    (pcre_free)(md->offset_vector);    (pcre_free)(md->offset_vector);
5416    }    }
5417    
5418  if (rc != MATCH_NOMATCH)  if (rc != MATCH_NOMATCH && rc != PCRE_ERROR_PARTIAL)
5419    {    {
5420    DPRINTF((">>>> error: returning %d\n", rc));    DPRINTF((">>>> error: returning %d\n", rc));
5421    return rc;    return rc;
5422    }    }
5423  else if (md->partial && md->hitend)  else if (start_partial != NULL)
5424    {    {
5425    DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n"));    DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n"));
5426      if (offsetcount > 1)
5427        {
5428        offsets[0] = start_partial - (USPTR)subject;
5429        offsets[1] = end_subject - (USPTR)subject;
5430        }
5431    return PCRE_ERROR_PARTIAL;    return PCRE_ERROR_PARTIAL;
5432    }    }
5433  else  else

Legend:
Removed from v.178  
changed lines
  Added in v.428

  ViewVC Help
Powered by ViewVC 1.1.5