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

Diff of /code/trunk/pcre_exec.c

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

revision 172 by ph10, Tue Jun 5 10:40:13 2007 UTC revision 426 by ph10, Wed Aug 26 15:38:32 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). The second one is used when we already know we are
407    past the end of the subject. */
408    
409    #define CHECK_PARTIAL()\
410      if (md->partial && eptr >= md->end_subject && eptr > mstart)\
411        md->hitend = TRUE
412    
413  Performance note: It might be tempting to extract commonly used fields from the  #define SCHECK_PARTIAL()\
414  md structure (e.g. utf8, end_subject) into individual variables to improve    if (md->partial && eptr > mstart) md->hitend = TRUE
415    
416    /* Performance note: It might be tempting to extract commonly used fields from
417    the md structure (e.g. utf8, end_subject) into individual variables to improve
418  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
419  made performance worse.  made performance worse.
420    
# Line 384  Arguments: Line 432  Arguments:
432                   match_condassert - this is an assertion condition                   match_condassert - this is an assertion condition
433                   match_cbegroup - this is the start of an unlimited repeat                   match_cbegroup - this is the start of an unlimited repeat
434                     group that can match an empty string                     group that can match an empty string
                  match_tail_recursed - this is a tail_recursed group  
435     rdepth      the recursion depth     rdepth      the recursion depth
436    
437  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 441  Returns:       MATCH_MATCH if matched
441  */  */
442    
443  static int  static int
444  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, const uschar *mstart,  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, USPTR mstart,
445    int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,    int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,
446    int flags, unsigned int rdepth)    int flags, unsigned int rdepth)
447  {  {
# Line 408  register unsigned int c;   /* Character Line 455  register unsigned int c;   /* Character
455  register BOOL utf8;        /* Local copy of UTF-8 flag for speed */  register BOOL utf8;        /* Local copy of UTF-8 flag for speed */
456    
457  BOOL minimize, possessive; /* Quantifier options */  BOOL minimize, possessive; /* Quantifier options */
458    int condcode;
459    
460  /* 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
461  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 498  HEAP_RECURSE:
498  #define charptr            frame->Xcharptr  #define charptr            frame->Xcharptr
499  #endif  #endif
500  #define callpat            frame->Xcallpat  #define callpat            frame->Xcallpat
501    #define codelink           frame->Xcodelink
502  #define data               frame->Xdata  #define data               frame->Xdata
503  #define next               frame->Xnext  #define next               frame->Xnext
504  #define pp                 frame->Xpp  #define pp                 frame->Xpp
# Line 530  int oclength; Line 579  int oclength;
579  uschar occhars[8];  uschar occhars[8];
580  #endif  #endif
581    
582    int codelink;
583  int ctype;  int ctype;
584  int length;  int length;
585  int max;  int max;
# Line 586  original_ims = ims;    /* Save for reset Line 636  original_ims = ims;    /* Save for reset
636  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
637  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
638  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.
639  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
640  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
641  already used. */  block that is used is on the stack, so a new one may be required for each
642    match(). */
643    
644  if ((flags & match_cbegroup) != 0)  if ((flags & match_cbegroup) != 0)
645    {    {
646    eptrblock *p;    newptrb.epb_saved_eptr = eptr;
647    if ((flags & match_tail_recursed) != 0)    newptrb.epb_prev = eptrb;
648      {    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;  
649    }    }
650    
651  /* Now start processing the opcodes. */  /* Now start processing the opcodes. */
# Line 612  for (;;) Line 656  for (;;)
656    op = *ecode;    op = *ecode;
657    
658    /* For partial matching, remember if we ever hit the end of the subject after    /* For partial matching, remember if we ever hit the end of the subject after
659    matching at least one subject character. */    matching at least one subject character. This code is now wrapped in a macro
660      because it appears several times below. */
661    
662    if (md->partial &&    CHECK_PARTIAL();
       eptr >= md->end_subject &&  
       eptr > mstart)  
     md->hitend = TRUE;  
663    
664    switch(op)    switch(op)
665      {      {
666        case OP_FAIL:
667        RRETURN(MATCH_NOMATCH);
668    
669        case OP_PRUNE:
670        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
671          ims, eptrb, flags, RM51);
672        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
673        RRETURN(MATCH_PRUNE);
674    
675        case OP_COMMIT:
676        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
677          ims, eptrb, flags, RM52);
678        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
679        RRETURN(MATCH_COMMIT);
680    
681        case OP_SKIP:
682        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
683          ims, eptrb, flags, RM53);
684        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
685        md->start_match_ptr = eptr;   /* Pass back current position */
686        RRETURN(MATCH_SKIP);
687    
688        case OP_THEN:
689        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
690          ims, eptrb, flags, RM54);
691        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
692        RRETURN(MATCH_THEN);
693    
694      /* 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
695      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.
696      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 732  for (;;)
732          {          {
733          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
734            ims, eptrb, flags, RM1);            ims, eptrb, flags, RM1);
735          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
736          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
737          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
738          }          }
# Line 677  for (;;) Line 747  for (;;)
747        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
748        }        }
749    
750      /* Insufficient room for saving captured contents. Treat as a non-capturing      /* FALL THROUGH ... Insufficient room for saving captured contents. Treat
751      bracket. */      as a non-capturing bracket. */
752    
753        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
754        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
755    
756      DPRINTF(("insufficient capture room: treat as non-capturing\n"));      DPRINTF(("insufficient capture room: treat as non-capturing\n"));
757    
758        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
759        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
760    
761      /* 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
762      final alternative within the brackets, we would return the result of a      final alternative within the brackets, we would return the result of a
763      recursive call to match() whatever happened. We can reduce stack usage by      recursive call to match() whatever happened. We can reduce stack usage by
764      turning this into a tail recursion. */      turning this into a tail recursion, except in the case when match_cbegroup
765        is set.*/
766    
767      case OP_BRA:      case OP_BRA:
768      case OP_SBRA:      case OP_SBRA:
# Line 693  for (;;) Line 770  for (;;)
770      flags = (op >= OP_SBRA)? match_cbegroup : 0;      flags = (op >= OP_SBRA)? match_cbegroup : 0;
771      for (;;)      for (;;)
772        {        {
773        if (ecode[GET(ecode, 1)] != OP_ALT)        if (ecode[GET(ecode, 1)] != OP_ALT)   /* Final alternative */
774          {          {
775          ecode += _pcre_OP_lengths[*ecode];          if (flags == 0)    /* Not a possibly empty group */
776          flags |= match_tail_recursed;            {
777          DPRINTF(("bracket 0 tail recursion\n"));            ecode += _pcre_OP_lengths[*ecode];
778          goto TAIL_RECURSE;            DPRINTF(("bracket 0 tail recursion\n"));
779              goto TAIL_RECURSE;
780              }
781    
782            /* Possibly empty group; can't use tail recursion. */
783    
784            RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
785              eptrb, flags, RM48);
786            RRETURN(rrc);
787          }          }
788    
789        /* 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 791  for (;;)
791    
792        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
793          eptrb, flags, RM2);          eptrb, flags, RM2);
794        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
795        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
796        }        }
797      /* Control never reaches here. */      /* Control never reaches here. */
# Line 719  for (;;) Line 804  for (;;)
804    
805      case OP_COND:      case OP_COND:
806      case OP_SCOND:      case OP_SCOND:
807      if (ecode[LINK_SIZE+1] == OP_RREF)         /* Recursion test */      codelink= GET(ecode, 1);
808    
809        /* Because of the way auto-callout works during compile, a callout item is
810        inserted between OP_COND and an assertion condition. */
811    
812        if (ecode[LINK_SIZE+1] == OP_CALLOUT)
813          {
814          if (pcre_callout != NULL)
815            {
816            pcre_callout_block cb;
817            cb.version          = 1;   /* Version 1 of the callout block */
818            cb.callout_number   = ecode[LINK_SIZE+2];
819            cb.offset_vector    = md->offset_vector;
820            cb.subject          = (PCRE_SPTR)md->start_subject;
821            cb.subject_length   = md->end_subject - md->start_subject;
822            cb.start_match      = mstart - md->start_subject;
823            cb.current_position = eptr - md->start_subject;
824            cb.pattern_position = GET(ecode, LINK_SIZE + 3);
825            cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);
826            cb.capture_top      = offset_top/2;
827            cb.capture_last     = md->capture_last;
828            cb.callout_data     = md->callout_data;
829            if ((rrc = (*pcre_callout)(&cb)) > 0) RRETURN(MATCH_NOMATCH);
830            if (rrc < 0) RRETURN(rrc);
831            }
832          ecode += _pcre_OP_lengths[OP_CALLOUT];
833          }
834    
835        condcode = ecode[LINK_SIZE+1];
836    
837        /* Now see what the actual condition is */
838    
839        if (condcode == OP_RREF)         /* Recursion test */
840        {        {
841        offset = GET2(ecode, LINK_SIZE + 2);     /* Recursion group number*/        offset = GET2(ecode, LINK_SIZE + 2);     /* Recursion group number*/
842        condition = md->recursive != NULL &&        condition = md->recursive != NULL &&
# Line 727  for (;;) Line 844  for (;;)
844        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 3 : GET(ecode, 1);
845        }        }
846    
847      else if (ecode[LINK_SIZE+1] == OP_CREF)    /* Group used test */      else if (condcode == OP_CREF)    /* Group used test */
848        {        {
849        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */
850        condition = offset < offset_top && md->offset_vector[offset] >= 0;        condition = offset < offset_top && md->offset_vector[offset] >= 0;
851        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 3 : GET(ecode, 1);
852        }        }
853    
854      else if (ecode[LINK_SIZE+1] == OP_DEF)     /* DEFINE - always false */      else if (condcode == OP_DEF)     /* DEFINE - always false */
855        {        {
856        condition = FALSE;        condition = FALSE;
857        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
# Line 754  for (;;) Line 871  for (;;)
871          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
872          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
873          }          }
874        else if (rrc != MATCH_NOMATCH)        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
875          {          {
876          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
877          }          }
878        else        else
879          {          {
880          condition = FALSE;          condition = FALSE;
881          ecode += GET(ecode, 1);          ecode += codelink;
882          }          }
883        }        }
884    
885      /* 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,
886      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
887      alternative doesn't exist, we can just plough on. */      match_cbegroup is required for an unlimited repeat of a possibly empty
888        group. If the second alternative doesn't exist, we can just plough on. */
889    
890      if (condition || *ecode == OP_ALT)      if (condition || *ecode == OP_ALT)
891        {        {
892        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
893        flags = match_tail_recursed | ((op == OP_SCOND)? match_cbegroup : 0);        if (op == OP_SCOND)        /* Possibly empty group */
894        goto TAIL_RECURSE;          {
895            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, match_cbegroup, RM49);
896            RRETURN(rrc);
897            }
898          else                       /* Group must match something */
899            {
900            flags = 0;
901            goto TAIL_RECURSE;
902            }
903        }        }
904      else      else                         /* Condition false & no alternative */
905        {        {
906        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
907        }        }
908      break;      break;
909    
910    
911      /* 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
912      restore the offsets appropriately and continue from after the call. */      recursion, we should restore the offsets appropriately and continue from
913        after the call. */
914    
915        case OP_ACCEPT:
916      case OP_END:      case OP_END:
917      if (md->recursive != NULL && md->recursive->group_num == 0)      if (md->recursive != NULL && md->recursive->group_num == 0)
918        {        {
# Line 805  for (;;) Line 933  for (;;)
933      if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);      if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);
934      md->end_match_ptr = eptr;           /* Record where we ended */      md->end_match_ptr = eptr;           /* Record where we ended */
935      md->end_offset_top = offset_top;    /* and how many extracts were taken */      md->end_offset_top = offset_top;    /* and how many extracts were taken */
936      md->start_match_ptr = mstart;  /* and the start (\K can modify) */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
937      RRETURN(MATCH_MATCH);      RRETURN(MATCH_MATCH);
938    
939      /* Change option settings */      /* Change option settings */
# Line 829  for (;;) Line 957  for (;;)
957        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
958          RM4);          RM4);
959        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
960        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
961        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
962        }        }
963      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 856  for (;;) Line 984  for (;;)
984        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
985          RM5);          RM5);
986        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);
987        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
988        ecode += GET(ecode,1);        ecode += GET(ecode,1);
989        }        }
990      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 880  for (;;) Line 1008  for (;;)
1008          {          {
1009          eptr--;          eptr--;
1010          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1011          BACKCHAR(eptr)          BACKCHAR(eptr);
1012          }          }
1013        }        }
1014      else      else
# Line 993  for (;;) Line 1121  for (;;)
1121              (pcre_free)(new_recursive.offset_save);              (pcre_free)(new_recursive.offset_save);
1122            RRETURN(MATCH_MATCH);            RRETURN(MATCH_MATCH);
1123            }            }
1124          else if (rrc != MATCH_NOMATCH)          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1125            {            {
1126            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1127              if (new_recursive.offset_save != stacksave)
1128                (pcre_free)(new_recursive.offset_save);
1129            RRETURN(rrc);            RRETURN(rrc);
1130            }            }
1131    
# Line 1027  for (;;) Line 1157  for (;;)
1157    
1158      do      do
1159        {        {
1160        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);  
1161        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
1162        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1163        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1164        }        }
1165      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 1073  for (;;) Line 1202  for (;;)
1202    
1203      if (*ecode == OP_KETRMIN)      if (*ecode == OP_KETRMIN)
1204        {        {
1205        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);  
1206        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1207        ecode = prev;        ecode = prev;
1208        flags = match_tail_recursed;        flags = 0;
1209        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1210        }        }
1211      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
# Line 1085  for (;;) Line 1213  for (;;)
1213        RMATCH(eptr, prev, offset_top, md, ims, eptrb, match_cbegroup, RM9);        RMATCH(eptr, prev, offset_top, md, ims, eptrb, match_cbegroup, RM9);
1214        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1215        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1216        flags = match_tail_recursed;        flags = 0;
1217        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1218        }        }
1219      /* Control never gets here */      /* Control never gets here */
# Line 1097  for (;;) Line 1225  for (;;)
1225      do ecode += GET(ecode,1); while (*ecode == OP_ALT);      do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1226      break;      break;
1227    
1228      /* BRAZERO and BRAMINZERO occur just before a bracket group, indicating      /* BRAZERO, BRAMINZERO and SKIPZERO occur just before a bracket group,
1229      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
1230      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
1231      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
1232      preceded by BRAZERO or BRAMINZERO. */      optional ones preceded by BRAZERO or BRAMINZERO. */
1233    
1234      case OP_BRAZERO:      case OP_BRAZERO:
1235        {        {
# Line 1123  for (;;) Line 1251  for (;;)
1251        }        }
1252      break;      break;
1253    
1254        case OP_SKIPZERO:
1255          {
1256          next = ecode+1;
1257          do next += GET(next,1); while (*next == OP_ALT);
1258          ecode = next + 1 + LINK_SIZE;
1259          }
1260        break;
1261    
1262      /* End of a group, repeated or non-repeating. */      /* End of a group, repeated or non-repeating. */
1263    
1264      case OP_KET:      case OP_KET:
# Line 1216  for (;;) Line 1352  for (;;)
1352    
1353      /* 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
1354      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
1355      tail recursion to avoid using another stack frame. */      tail recursion to avoid using another stack frame, unless we have an
1356        unlimited repeat of a group that can match an empty string. */
1357    
1358      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;
1359    
1360      if (*ecode == OP_KETRMIN)      if (*ecode == OP_KETRMIN)
1361        {        {
1362        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);  
1363        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1364          if (flags != 0)    /* Could match an empty string */
1365            {
1366            RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM50);
1367            RRETURN(rrc);
1368            }
1369        ecode = prev;        ecode = prev;
       flags |= match_tail_recursed;  
1370        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1371        }        }
1372      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
# Line 1234  for (;;) Line 1374  for (;;)
1374        RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);        RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);
1375        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1376        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1377        flags = match_tail_recursed;        flags = 0;
1378        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1379        }        }
1380      /* Control never gets here */      /* Control never gets here */
# Line 1332  for (;;) Line 1472  for (;;)
1472          {          {
1473          if (eptr == md->start_subject) prev_is_word = FALSE; else          if (eptr == md->start_subject) prev_is_word = FALSE; else
1474            {            {
1475            const uschar *lastptr = eptr - 1;            USPTR lastptr = eptr - 1;
1476            while((*lastptr & 0xc0) == 0x80) lastptr--;            while((*lastptr & 0xc0) == 0x80) lastptr--;
1477            GETCHAR(c, lastptr);            GETCHAR(c, lastptr);
1478            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
# Line 1366  for (;;) Line 1506  for (;;)
1506      /* Match a single character type; inline for speed */      /* Match a single character type; inline for speed */
1507    
1508      case OP_ANY:      case OP_ANY:
1509      if ((ims & PCRE_DOTALL) == 0)      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
1510        {      /* Fall through */
1511        if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);  
1512        }      case OP_ALLANY:
1513      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);
1514      if (utf8)      if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
       while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  
1515      ecode++;      ecode++;
1516      break;      break;
1517    
# Line 1471  for (;;) Line 1610  for (;;)
1610        case 0x000d:        case 0x000d:
1611        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
1612        break;        break;
1613    
1614        case 0x000a:        case 0x000a:
1615          break;
1616    
1617        case 0x000b:        case 0x000b:
1618        case 0x000c:        case 0x000c:
1619        case 0x0085:        case 0x0085:
1620        case 0x2028:        case 0x2028:
1621        case 0x2029:        case 0x2029:
1622          if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
1623          break;
1624          }
1625        ecode++;
1626        break;
1627    
1628        case OP_NOT_HSPACE:
1629        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1630        GETCHARINCTEST(c, eptr);
1631        switch(c)
1632          {
1633          default: break;
1634          case 0x09:      /* HT */
1635          case 0x20:      /* SPACE */
1636          case 0xa0:      /* NBSP */
1637          case 0x1680:    /* OGHAM SPACE MARK */
1638          case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
1639          case 0x2000:    /* EN QUAD */
1640          case 0x2001:    /* EM QUAD */
1641          case 0x2002:    /* EN SPACE */
1642          case 0x2003:    /* EM SPACE */
1643          case 0x2004:    /* THREE-PER-EM SPACE */
1644          case 0x2005:    /* FOUR-PER-EM SPACE */
1645          case 0x2006:    /* SIX-PER-EM SPACE */
1646          case 0x2007:    /* FIGURE SPACE */
1647          case 0x2008:    /* PUNCTUATION SPACE */
1648          case 0x2009:    /* THIN SPACE */
1649          case 0x200A:    /* HAIR SPACE */
1650          case 0x202f:    /* NARROW NO-BREAK SPACE */
1651          case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
1652          case 0x3000:    /* IDEOGRAPHIC SPACE */
1653          RRETURN(MATCH_NOMATCH);
1654          }
1655        ecode++;
1656        break;
1657    
1658        case OP_HSPACE:
1659        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1660        GETCHARINCTEST(c, eptr);
1661        switch(c)
1662          {
1663          default: RRETURN(MATCH_NOMATCH);
1664          case 0x09:      /* HT */
1665          case 0x20:      /* SPACE */
1666          case 0xa0:      /* NBSP */
1667          case 0x1680:    /* OGHAM SPACE MARK */
1668          case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
1669          case 0x2000:    /* EN QUAD */
1670          case 0x2001:    /* EM QUAD */
1671          case 0x2002:    /* EN SPACE */
1672          case 0x2003:    /* EM SPACE */
1673          case 0x2004:    /* THREE-PER-EM SPACE */
1674          case 0x2005:    /* FOUR-PER-EM SPACE */
1675          case 0x2006:    /* SIX-PER-EM SPACE */
1676          case 0x2007:    /* FIGURE SPACE */
1677          case 0x2008:    /* PUNCTUATION SPACE */
1678          case 0x2009:    /* THIN SPACE */
1679          case 0x200A:    /* HAIR SPACE */
1680          case 0x202f:    /* NARROW NO-BREAK SPACE */
1681          case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
1682          case 0x3000:    /* IDEOGRAPHIC SPACE */
1683          break;
1684          }
1685        ecode++;
1686        break;
1687    
1688        case OP_NOT_VSPACE:
1689        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1690        GETCHARINCTEST(c, eptr);
1691        switch(c)
1692          {
1693          default: break;
1694          case 0x0a:      /* LF */
1695          case 0x0b:      /* VT */
1696          case 0x0c:      /* FF */
1697          case 0x0d:      /* CR */
1698          case 0x85:      /* NEL */
1699          case 0x2028:    /* LINE SEPARATOR */
1700          case 0x2029:    /* PARAGRAPH SEPARATOR */
1701          RRETURN(MATCH_NOMATCH);
1702          }
1703        ecode++;
1704        break;
1705    
1706        case OP_VSPACE:
1707        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1708        GETCHARINCTEST(c, eptr);
1709        switch(c)
1710          {
1711          default: RRETURN(MATCH_NOMATCH);
1712          case 0x0a:      /* LF */
1713          case 0x0b:      /* VT */
1714          case 0x0c:      /* FF */
1715          case 0x0d:      /* CR */
1716          case 0x85:      /* NEL */
1717          case 0x2028:    /* LINE SEPARATOR */
1718          case 0x2029:    /* PARAGRAPH SEPARATOR */
1719        break;        break;
1720        }        }
1721      ecode++;      ecode++;
# Line 1491  for (;;) Line 1730  for (;;)
1730      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1731      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1732        {        {
1733        int chartype, script;        const ucd_record *prop = GET_UCD(c);
       int category = _pcre_ucp_findprop(c, &chartype, &script);  
1734    
1735        switch(ecode[1])        switch(ecode[1])
1736          {          {
# Line 1501  for (;;) Line 1739  for (;;)
1739          break;          break;
1740    
1741          case PT_LAMP:          case PT_LAMP:
1742          if ((chartype == ucp_Lu ||          if ((prop->chartype == ucp_Lu ||
1743               chartype == ucp_Ll ||               prop->chartype == ucp_Ll ||
1744               chartype == ucp_Lt) == (op == OP_NOTPROP))               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))
1745            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1746           break;           break;
1747    
1748          case PT_GC:          case PT_GC:
1749          if ((ecode[2] != category) == (op == OP_PROP))          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))
1750            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1751          break;          break;
1752    
1753          case PT_PC:          case PT_PC:
1754          if ((ecode[2] != chartype) == (op == OP_PROP))          if ((ecode[2] != prop->chartype) == (op == OP_PROP))
1755            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1756          break;          break;
1757    
1758          case PT_SC:          case PT_SC:
1759          if ((ecode[2] != script) == (op == OP_PROP))          if ((ecode[2] != prop->script) == (op == OP_PROP))
1760            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1761          break;          break;
1762    
# Line 1537  for (;;) Line 1775  for (;;)
1775      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1776      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1777        {        {
1778        int chartype, script;        int category = UCD_CATEGORY(c);
       int category = _pcre_ucp_findprop(c, &chartype, &script);  
1779        if (category == ucp_M) RRETURN(MATCH_NOMATCH);        if (category == ucp_M) RRETURN(MATCH_NOMATCH);
1780        while (eptr < md->end_subject)        while (eptr < md->end_subject)
1781          {          {
# Line 1547  for (;;) Line 1784  for (;;)
1784            {            {
1785            GETCHARLEN(c, eptr, len);            GETCHARLEN(c, eptr, len);
1786            }            }
1787          category = _pcre_ucp_findprop(c, &chartype, &script);          category = UCD_CATEGORY(c);
1788          if (category != ucp_M) break;          if (category != ucp_M) break;
1789          eptr += len;          eptr += len;
1790          }          }
# Line 1568  for (;;) Line 1805  for (;;)
1805      case OP_REF:      case OP_REF:
1806        {        {
1807        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
1808        ecode += 3;                                 /* Advance past item */        ecode += 3;
1809    
1810          /* If the reference is unset, there are two possibilities:
1811    
1812        /* 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
1813        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
1814        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
1815        minima. */        quantifiers with zero minima.
1816    
1817        length = (offset >= offset_top || md->offset_vector[offset] < 0)?        (b) If the JavaScript compatibility flag is set, set the length to zero
1818          md->end_subject - eptr + 1 :        so that the back reference matches an empty string.
1819          md->offset_vector[offset+1] - md->offset_vector[offset];  
1820          Otherwise, set the length to the length of what was matched by the
1821          referenced subpattern. */
1822    
1823          if (offset >= offset_top || md->offset_vector[offset] < 0)
1824            length = (md->jscript_compat)? 0 : md->end_subject - eptr + 1;
1825          else
1826            length = md->offset_vector[offset+1] - md->offset_vector[offset];
1827    
1828        /* Set up for repetition, or handle the non-repeated case */        /* Set up for repetition, or handle the non-repeated case */
1829    
# Line 1622  for (;;) Line 1868  for (;;)
1868    
1869        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
1870          {          {
1871          if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH);          if (!match_ref(offset, eptr, length, md, ims))
1872              {
1873              CHECK_PARTIAL();
1874              RRETURN(MATCH_NOMATCH);
1875              }
1876          eptr += length;          eptr += length;
1877          }          }
1878    
# Line 1640  for (;;) Line 1890  for (;;)
1890            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);
1891            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1892            if (fi >= max || !match_ref(offset, eptr, length, md, ims))            if (fi >= max || !match_ref(offset, eptr, length, md, ims))
1893                {
1894                CHECK_PARTIAL();
1895              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
1896                }
1897            eptr += length;            eptr += length;
1898            }            }
1899          /* Control never gets here */          /* Control never gets here */
# Line 1656  for (;;) Line 1909  for (;;)
1909            if (!match_ref(offset, eptr, length, md, ims)) break;            if (!match_ref(offset, eptr, length, md, ims)) break;
1910            eptr += length;            eptr += length;
1911            }            }
1912            CHECK_PARTIAL();
1913          while (eptr >= pp)          while (eptr >= pp)
1914            {            {
1915            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM15);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM15);
# Line 1723  for (;;) Line 1977  for (;;)
1977          {          {
1978          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
1979            {            {
1980            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
1981                {
1982                CHECK_PARTIAL();
1983                RRETURN(MATCH_NOMATCH);
1984                }
1985            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
1986            if (c > 255)            if (c > 255)
1987              {              {
# Line 1741  for (;;) Line 1999  for (;;)
1999          {          {
2000          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2001            {            {
2002            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
2003                {
2004                CHECK_PARTIAL();
2005                RRETURN(MATCH_NOMATCH);
2006                }
2007            c = *eptr++;            c = *eptr++;
2008            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2009            }            }
# Line 1765  for (;;) Line 2027  for (;;)
2027              {              {
2028              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);
2029              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2030              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max)
2031                  {
2032                  CHECK_PARTIAL();
2033                  RRETURN(MATCH_NOMATCH);
2034                  }
2035                if (eptr >= md->end_subject)
2036                  {
2037                  SCHECK_PARTIAL();
2038                  RRETURN(MATCH_NOMATCH);
2039                  }
2040              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
2041              if (c > 255)              if (c > 255)
2042                {                {
# Line 1785  for (;;) Line 2056  for (;;)
2056              {              {
2057              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);
2058              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2059              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max)
2060                  {
2061                  CHECK_PARTIAL();
2062                  RRETURN(MATCH_NOMATCH);
2063                  }
2064                if (eptr >= md->end_subject)
2065                  {
2066                  SCHECK_PARTIAL();
2067                  RRETURN(MATCH_NOMATCH);
2068                  }
2069              c = *eptr++;              c = *eptr++;
2070              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2071              }              }
# Line 1818  for (;;) Line 2098  for (;;)
2098                }                }
2099              eptr += len;              eptr += len;
2100              }              }
2101              CHECK_PARTIAL();
2102            for (;;)            for (;;)
2103              {              {
2104              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM18);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM18);
# Line 1837  for (;;) Line 2118  for (;;)
2118              if ((data[c/8] & (1 << (c&7))) == 0) break;              if ((data[c/8] & (1 << (c&7))) == 0) break;
2119              eptr++;              eptr++;
2120              }              }
2121              CHECK_PARTIAL();
2122            while (eptr >= pp)            while (eptr >= pp)
2123              {              {
2124              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM19);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM19);
# Line 1852  for (;;) Line 2134  for (;;)
2134    
2135    
2136      /* Match an extended character class. This opcode is encountered only      /* Match an extended character class. This opcode is encountered only
2137      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
2138        mode, because Unicode properties are supported in non-UTF-8 mode. */
2139    
2140  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
2141      case OP_XCLASS:      case OP_XCLASS:
# Line 1893  for (;;) Line 2176  for (;;)
2176    
2177        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2178          {          {
2179          if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);          if (eptr >= md->end_subject)
2180          GETCHARINC(c, eptr);            {
2181              SCHECK_PARTIAL();
2182              RRETURN(MATCH_NOMATCH);
2183              }
2184            GETCHARINCTEST(c, eptr);
2185          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);
2186          }          }
2187    
# Line 1912  for (;;) Line 2199  for (;;)
2199            {            {
2200            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);
2201            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2202            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max)
2203            GETCHARINC(c, eptr);              {
2204                CHECK_PARTIAL();
2205                RRETURN(MATCH_NOMATCH);
2206                }
2207              if (eptr >= md->end_subject)
2208                {
2209                SCHECK_PARTIAL();
2210                RRETURN(MATCH_NOMATCH);
2211                }
2212              GETCHARINCTEST(c, eptr);
2213            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);
2214            }            }
2215          /* Control never gets here */          /* Control never gets here */
# Line 1928  for (;;) Line 2224  for (;;)
2224            {            {
2225            int len = 1;            int len = 1;
2226            if (eptr >= md->end_subject) break;            if (eptr >= md->end_subject) break;
2227            GETCHARLEN(c, eptr, len);            GETCHARLENTEST(c, eptr, len);
2228            if (!_pcre_xclass(c, data)) break;            if (!_pcre_xclass(c, data)) break;
2229            eptr += len;            eptr += len;
2230            }            }
2231            CHECK_PARTIAL();
2232          for(;;)          for(;;)
2233            {            {
2234            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);
2235            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2236            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
2237            BACKCHAR(eptr)            if (utf8) BACKCHAR(eptr);
2238            }            }
2239          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2240          }          }
# Line 2003  for (;;) Line 2300  for (;;)
2300          if (fc != dc)          if (fc != dc)
2301            {            {
2302  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2303            if (dc != _pcre_ucp_othercase(fc))            if (dc != UCD_OTHERCASE(fc))
2304  #endif  #endif
2305              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2306            }            }
# Line 2072  for (;;) Line 2369  for (;;)
2369      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
2370      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
2371    
2372      /* 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. */  
2373    
2374      REPEATCHAR:      REPEATCHAR:
2375  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 2083  for (;;) Line 2378  for (;;)
2378        length = 1;        length = 1;
2379        charptr = ecode;        charptr = ecode;
2380        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
       if (min * length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
2381        ecode += length;        ecode += length;
2382    
2383        /* Handle multibyte character matching specially here. There is        /* Handle multibyte character matching specially here. There is
# Line 2094  for (;;) Line 2388  for (;;)
2388  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2389          unsigned int othercase;          unsigned int othercase;
2390          if ((ims & PCRE_CASELESS) != 0 &&          if ((ims & PCRE_CASELESS) != 0 &&
2391              (othercase = _pcre_ucp_othercase(fc)) != NOTACHAR)              (othercase = UCD_OTHERCASE(fc)) != fc)
2392            oclength = _pcre_ord2utf8(othercase, occhars);            oclength = _pcre_ord2utf8(othercase, occhars);
2393          else oclength = 0;          else oclength = 0;
2394  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2395    
2396          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2397            {            {
2398            if (memcmp(eptr, charptr, length) == 0) eptr += length;            if (eptr <= md->end_subject - length &&
2399                memcmp(eptr, charptr, length) == 0) eptr += length;
2400  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2401            /* Need braces because of following else */            else if (oclength > 0 &&
2402            else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                     eptr <= md->end_subject - oclength &&
2403                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2404    #endif  /* SUPPORT_UCP */
2405            else            else
2406              {              {
2407              if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);              CHECK_PARTIAL();
2408              eptr += oclength;              RRETURN(MATCH_NOMATCH);
2409              }              }
 #else   /* without SUPPORT_UCP */  
           else { RRETURN(MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2410            }            }
2411    
2412          if (min == max) continue;          if (min == max) continue;
# Line 2123  for (;;) Line 2417  for (;;)
2417              {              {
2418              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);
2419              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2420              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max)
2421              if (memcmp(eptr, charptr, length) == 0) eptr += length;                {
2422                  CHECK_PARTIAL();
2423                  RRETURN(MATCH_NOMATCH);
2424                  }
2425                if (eptr <= md->end_subject - length &&
2426                  memcmp(eptr, charptr, length) == 0) eptr += length;
2427  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2428              /* Need braces because of following else */              else if (oclength > 0 &&
2429              else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                       eptr <= md->end_subject - oclength &&
2430                         memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2431    #endif  /* SUPPORT_UCP */
2432              else              else
2433                {                {
2434                if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);                CHECK_PARTIAL();
2435                eptr += oclength;                RRETURN(MATCH_NOMATCH);
2436                }                }
 #else   /* without SUPPORT_UCP */  
             else { RRETURN (MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2437              }              }
2438            /* Control never gets here */            /* Control never gets here */
2439            }            }
# Line 2145  for (;;) Line 2443  for (;;)
2443            pp = eptr;            pp = eptr;
2444            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2445              {              {
2446              if (eptr > md->end_subject - length) break;              if (eptr <= md->end_subject - length &&
2447              if (memcmp(eptr, charptr, length) == 0) eptr += length;                  memcmp(eptr, charptr, length) == 0) eptr += length;
2448  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2449              else if (oclength == 0) break;              else if (oclength > 0 &&
2450              else                       eptr <= md->end_subject - oclength &&
2451                {                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
               if (memcmp(eptr, occhars, oclength) != 0) break;  
               eptr += oclength;  
               }  
 #else   /* without SUPPORT_UCP */  
             else break;  
2452  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2453                else break;
2454              }              }
2455    
2456              CHECK_PARTIAL();
2457            if (possessive) continue;            if (possessive) continue;
2458    
2459            for(;;)            for(;;)
2460             {              {
2461             RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);
2462             if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2463             if (eptr == pp) RRETURN(MATCH_NOMATCH);              if (eptr == pp) { RRETURN(MATCH_NOMATCH); }
2464  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2465             eptr--;              eptr--;
2466             BACKCHAR(eptr);              BACKCHAR(eptr);
2467  #else   /* without SUPPORT_UCP */  #else   /* without SUPPORT_UCP */
2468             eptr -= length;              eptr -= length;
2469  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2470             }              }
2471            }            }
2472          /* Control never gets here */          /* Control never gets here */
2473          }          }
# Line 2184  for (;;) Line 2480  for (;;)
2480  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF8 */
2481    
2482      /* When not in UTF-8 mode, load a single-byte character. */      /* When not in UTF-8 mode, load a single-byte character. */
2483        {  
2484        if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);      fc = *ecode++;
       fc = *ecode++;  
       }  
2485    
2486      /* 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
2487      may not be in UTF-8 mode. The code is duplicated for the caseless and      may not be in UTF-8 mode. The code is duplicated for the caseless and
# Line 2205  for (;;) Line 2499  for (;;)
2499        {        {
2500        fc = md->lcc[fc];        fc = md->lcc[fc];
2501        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2502            {
2503            if (eptr >= md->end_subject)
2504              {
2505              SCHECK_PARTIAL();
2506              RRETURN(MATCH_NOMATCH);
2507              }
2508          if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);          if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2509            }
2510        if (min == max) continue;        if (min == max) continue;
2511        if (minimize)        if (minimize)
2512          {          {
# Line 2213  for (;;) Line 2514  for (;;)
2514            {            {
2515            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);
2516            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2517            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max)
2518                fc != md->lcc[*eptr++])              {
2519                CHECK_PARTIAL();
2520                RRETURN(MATCH_NOMATCH);
2521                }
2522              if (eptr >= md->end_subject)
2523                {
2524                SCHECK_PARTIAL();
2525              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2526                }
2527              if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2528            }            }
2529          /* Control never gets here */          /* Control never gets here */
2530          }          }
# Line 2227  for (;;) Line 2536  for (;;)
2536            if (eptr >= md->end_subject || fc != md->lcc[*eptr]) break;            if (eptr >= md->end_subject || fc != md->lcc[*eptr]) break;
2537            eptr++;            eptr++;
2538            }            }
2539    
2540            CHECK_PARTIAL();
2541          if (possessive) continue;          if (possessive) continue;
2542    
2543          while (eptr >= pp)          while (eptr >= pp)
2544            {            {
2545            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);
# Line 2243  for (;;) Line 2555  for (;;)
2555    
2556      else      else
2557        {        {
2558        for (i = 1; i <= min; i++) if (fc != *eptr++) RRETURN(MATCH_NOMATCH);        for (i = 1; i <= min; i++)
2559            {
2560            if (eptr >= md->end_subject)
2561              {
2562              SCHECK_PARTIAL();
2563              RRETURN(MATCH_NOMATCH);
2564              }
2565            if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
2566            }
2567        if (min == max) continue;        if (min == max) continue;
2568        if (minimize)        if (minimize)
2569          {          {
# Line 2251  for (;;) Line 2571  for (;;)
2571            {            {
2572            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);
2573            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2574            if (fi >= max || eptr >= md->end_subject || fc != *eptr++)            if (fi >= max)
2575                {
2576                CHECK_PARTIAL();
2577                RRETURN(MATCH_NOMATCH);
2578                }
2579              if (eptr >= md->end_subject)
2580                {
2581                SCHECK_PARTIAL();
2582              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2583                }
2584              if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
2585            }            }
2586          /* Control never gets here */          /* Control never gets here */
2587          }          }
# Line 2264  for (;;) Line 2593  for (;;)
2593            if (eptr >= md->end_subject || fc != *eptr) break;            if (eptr >= md->end_subject || fc != *eptr) break;
2594            eptr++;            eptr++;
2595            }            }
2596            CHECK_PARTIAL();
2597          if (possessive) continue;          if (possessive) continue;
2598          while (eptr >= pp)          while (eptr >= pp)
2599            {            {
# Line 2357  for (;;) Line 2687  for (;;)
2687      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
2688      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
2689    
2690      /* 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. */  
2691    
2692      REPEATNOTCHAR:      REPEATNOTCHAR:
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
2693      fc = *ecode++;      fc = *ecode++;
2694    
2695      /* 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 2387  for (;;) Line 2714  for (;;)
2714          register unsigned int d;          register unsigned int d;
2715          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2716            {            {
2717              if (eptr >= md->end_subject)
2718                {
2719                SCHECK_PARTIAL();
2720                RRETURN(MATCH_NOMATCH);
2721                }
2722            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
2723            if (d < 256) d = md->lcc[d];            if (d < 256) d = md->lcc[d];
2724            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
# Line 2398  for (;;) Line 2730  for (;;)
2730        /* Not UTF-8 mode */        /* Not UTF-8 mode */
2731          {          {
2732          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2733              {
2734              if (eptr >= md->end_subject)
2735                {
2736                SCHECK_PARTIAL();
2737                RRETURN(MATCH_NOMATCH);
2738                }
2739            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2740              }
2741          }          }
2742    
2743        if (min == max) continue;        if (min == max) continue;
# Line 2414  for (;;) Line 2753  for (;;)
2753              {              {
2754              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);
2755              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2756                if (fi >= max)
2757                  {
2758                  CHECK_PARTIAL();
2759                  RRETURN(MATCH_NOMATCH);
2760                  }
2761                if (eptr >= md->end_subject)
2762                  {
2763                  SCHECK_PARTIAL();
2764                  RRETURN(MATCH_NOMATCH);
2765                  }
2766              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
2767              if (d < 256) d = md->lcc[d];              if (d < 256) d = md->lcc[d];
2768              if (fi >= max || eptr >= md->end_subject || fc == d)              if (fc == d) RRETURN(MATCH_NOMATCH);
               RRETURN(MATCH_NOMATCH);  
2769              }              }
2770            }            }
2771          else          else
# Line 2428  for (;;) Line 2776  for (;;)
2776              {              {
2777              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);
2778              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2779              if (fi >= max || eptr >= md->end_subject || fc == md->lcc[*eptr++])              if (fi >= max)
2780                  {
2781                  CHECK_PARTIAL();
2782                  RRETURN(MATCH_NOMATCH);
2783                  }
2784                if (eptr >= md->end_subject)
2785                  {
2786                  SCHECK_PARTIAL();
2787                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2788                  }
2789                if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2790              }              }
2791            }            }
2792          /* Control never gets here */          /* Control never gets here */
# Line 2455  for (;;) Line 2812  for (;;)
2812              if (fc == d) break;              if (fc == d) break;
2813              eptr += len;              eptr += len;
2814              }              }
2815            CHECK_PARTIAL();
2816          if (possessive) continue;          if (possessive) continue;
2817          for(;;)          for(;;)
2818              {              {
# Line 2473  for (;;) Line 2831  for (;;)
2831              if (eptr >= md->end_subject || fc == md->lcc[*eptr]) break;              if (eptr >= md->end_subject || fc == md->lcc[*eptr]) break;
2832              eptr++;              eptr++;
2833              }              }
2834              CHECK_PARTIAL();
2835            if (possessive) continue;            if (possessive) continue;
2836            while (eptr >= pp)            while (eptr >= pp)
2837              {              {
# Line 2498  for (;;) Line 2857  for (;;)
2857          register unsigned int d;          register unsigned int d;
2858          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2859            {            {
2860              if (eptr >= md->end_subject)
2861                {
2862                SCHECK_PARTIAL();
2863                RRETURN(MATCH_NOMATCH);
2864                }
2865            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
2866            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
2867            }            }
# Line 2507  for (;;) Line 2871  for (;;)
2871        /* Not UTF-8 mode */        /* Not UTF-8 mode */
2872          {          {
2873          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2874              {
2875              if (eptr >= md->end_subject)
2876                {
2877                SCHECK_PARTIAL();
2878                RRETURN(MATCH_NOMATCH);
2879                }
2880            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
2881              }
2882          }          }
2883    
2884        if (min == max) continue;        if (min == max) continue;
# Line 2523  for (;;) Line 2894  for (;;)
2894              {              {
2895              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);
2896              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2897              GETCHARINC(d, eptr);              if (fi >= max)
2898              if (fi >= max || eptr >= md->end_subject || fc == d)                {
2899                  CHECK_PARTIAL();
2900                  RRETURN(MATCH_NOMATCH);
2901                  }
2902                if (eptr >= md->end_subject)
2903                  {
2904                  SCHECK_PARTIAL();
2905                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2906                  }
2907                GETCHARINC(d, eptr);
2908                if (fc == d) RRETURN(MATCH_NOMATCH);
2909              }              }
2910            }            }
2911          else          else
# Line 2536  for (;;) Line 2916  for (;;)
2916              {              {
2917              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);
2918              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2919              if (fi >= max || eptr >= md->end_subject || fc == *eptr++)              if (fi >= max)
2920                  {
2921                  CHECK_PARTIAL();
2922                  RRETURN(MATCH_NOMATCH);
2923                  }
2924                if (eptr >= md->end_subject)
2925                  {
2926                  SCHECK_PARTIAL();
2927                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2928                  }
2929                if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
2930              }              }
2931            }            }
2932          /* Control never gets here */          /* Control never gets here */
# Line 2562  for (;;) Line 2951  for (;;)
2951              if (fc == d) break;              if (fc == d) break;
2952              eptr += len;              eptr += len;
2953              }              }
2954              CHECK_PARTIAL();
2955            if (possessive) continue;            if (possessive) continue;
2956            for(;;)            for(;;)
2957              {              {
# Line 2580  for (;;) Line 2970  for (;;)
2970              if (eptr >= md->end_subject || fc == *eptr) break;              if (eptr >= md->end_subject || fc == *eptr) break;
2971              eptr++;              eptr++;
2972              }              }
2973              CHECK_PARTIAL();
2974            if (possessive) continue;            if (possessive) continue;
2975            while (eptr >= pp)            while (eptr >= pp)
2976              {              {
# Line 2671  for (;;) Line 3062  for (;;)
3062    
3063      /* First, ensure the minimum number of matches are present. Use inline      /* First, ensure the minimum number of matches are present. Use inline
3064      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
3065      (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  
3066      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
3067      and single-bytes. */      and single-bytes. */
3068    
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
3069      if (min > 0)      if (min > 0)
3070        {        {
3071  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 2689  for (;;) Line 3077  for (;;)
3077            if (prop_fail_result) RRETURN(MATCH_NOMATCH);            if (prop_fail_result) RRETURN(MATCH_NOMATCH);
3078            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3079              {              {
3080              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3081              GETCHARINC(c, eptr);                {
3082                  SCHECK_PARTIAL();
3083                  RRETURN(MATCH_NOMATCH);
3084                  }
3085                GETCHARINCTEST(c, eptr);
3086              }              }
3087            break;            break;
3088    
3089            case PT_LAMP:            case PT_LAMP:
3090            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3091              {              {
3092              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3093              GETCHARINC(c, eptr);                {
3094              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3095                  RRETURN(MATCH_NOMATCH);
3096                  }
3097                GETCHARINCTEST(c, eptr);
3098                prop_chartype = UCD_CHARTYPE(c);
3099              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
3100                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
3101                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 2710  for (;;) Line 3106  for (;;)
3106            case PT_GC:            case PT_GC:
3107            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3108              {              {
3109              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3110              GETCHARINC(c, eptr);                {
3111              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3112                  RRETURN(MATCH_NOMATCH);
3113                  }
3114                GETCHARINCTEST(c, eptr);
3115                prop_category = UCD_CATEGORY(c);
3116              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
3117                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3118              }              }
# Line 2721  for (;;) Line 3121  for (;;)
3121            case PT_PC:            case PT_PC:
3122            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3123              {              {
3124              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3125              GETCHARINC(c, eptr);                {
3126              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3127                  RRETURN(MATCH_NOMATCH);
3128                  }
3129                GETCHARINCTEST(c, eptr);
3130                prop_chartype = UCD_CHARTYPE(c);
3131              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
3132                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3133              }              }
# Line 2732  for (;;) Line 3136  for (;;)
3136            case PT_SC:            case PT_SC:
3137            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3138              {              {
3139              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3140              GETCHARINC(c, eptr);                {
3141              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3142                  RRETURN(MATCH_NOMATCH);
3143                  }
3144                GETCHARINCTEST(c, eptr);
3145                prop_script = UCD_SCRIPT(c);
3146              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
3147                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3148              }              }
# Line 2752  for (;;) Line 3160  for (;;)
3160          {          {
3161          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3162            {            {
3163              if (eptr >= md->end_subject)
3164                {
3165                SCHECK_PARTIAL();
3166                RRETURN(MATCH_NOMATCH);
3167                }
3168            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3169            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
3170            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);
3171            while (eptr < md->end_subject)            while (eptr < md->end_subject)
3172              {              {
3173              int len = 1;              int len = 1;
3174              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
3175                {                else { GETCHARLEN(c, eptr, len); }
3176                GETCHARLEN(c, eptr, len);              prop_category = UCD_CATEGORY(c);
               }  
             prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);  
3177              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
3178              eptr += len;              eptr += len;
3179              }              }
# Line 2780  for (;;) Line 3191  for (;;)
3191          case OP_ANY:          case OP_ANY:
3192          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3193            {            {
3194            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3195                 ((ims & PCRE_DOTALL) == 0 && IS_NEWLINE(eptr)))              {
3196                SCHECK_PARTIAL();
3197                RRETURN(MATCH_NOMATCH);
3198                }
3199              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
3200              eptr++;
3201              while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3202              }
3203            break;
3204    
3205            case OP_ALLANY:
3206            for (i = 1; i <= min; i++)
3207              {
3208              if (eptr >= md->end_subject)
3209                {
3210                SCHECK_PARTIAL();
3211              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3212                }
3213            eptr++;            eptr++;
3214            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3215            }            }
3216          break;          break;
3217    
3218          case OP_ANYBYTE:          case OP_ANYBYTE:
3219            if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
3220          eptr += min;          eptr += min;
3221          break;          break;
3222    
3223          case OP_ANYNL:          case OP_ANYNL:
3224          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3225            {            {
3226            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3227                {
3228                SCHECK_PARTIAL();
3229                RRETURN(MATCH_NOMATCH);
3230                }
3231            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3232            switch(c)            switch(c)
3233              {              {
# Line 2803  for (;;) Line 3235  for (;;)
3235              case 0x000d:              case 0x000d:
3236              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3237              break;              break;
3238    
3239              case 0x000a:              case 0x000a:
3240                break;
3241    
3242              case 0x000b:              case 0x000b:
3243              case 0x000c:              case 0x000c:
3244              case 0x0085:              case 0x0085:
3245              case 0x2028:              case 0x2028:
3246              case 0x2029:              case 0x2029:
3247                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3248              break;              break;
3249              }              }
3250            }            }
3251          break;          break;
3252    
3253          case OP_NOT_DIGIT:          case OP_NOT_HSPACE:
3254          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3255            {            {
3256            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3257                {
3258                SCHECK_PARTIAL();
3259                RRETURN(MATCH_NOMATCH);
3260                }
3261            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3262            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)            switch(c)
3263                {
3264                default: break;
3265                case 0x09:      /* HT */
3266                case 0x20:      /* SPACE */
3267                case 0xa0:      /* NBSP */
3268                case 0x1680:    /* OGHAM SPACE MARK */
3269                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3270                case 0x2000:    /* EN QUAD */
3271                case 0x2001:    /* EM QUAD */
3272                case 0x2002:    /* EN SPACE */
3273                case 0x2003:    /* EM SPACE */
3274                case 0x2004:    /* THREE-PER-EM SPACE */
3275                case 0x2005:    /* FOUR-PER-EM SPACE */
3276                case 0x2006:    /* SIX-PER-EM SPACE */
3277                case 0x2007:    /* FIGURE SPACE */
3278                case 0x2008:    /* PUNCTUATION SPACE */
3279                case 0x2009:    /* THIN SPACE */
3280                case 0x200A:    /* HAIR SPACE */
3281                case 0x202f:    /* NARROW NO-BREAK SPACE */
3282                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3283                case 0x3000:    /* IDEOGRAPHIC SPACE */
3284                RRETURN(MATCH_NOMATCH);
3285                }
3286              }
3287            break;
3288    
3289            case OP_HSPACE:
3290            for (i = 1; i <= min; i++)
3291              {
3292              if (eptr >= md->end_subject)
3293                {
3294                SCHECK_PARTIAL();
3295                RRETURN(MATCH_NOMATCH);
3296                }
3297              GETCHARINC(c, eptr);
3298              switch(c)
3299                {
3300                default: RRETURN(MATCH_NOMATCH);
3301                case 0x09:      /* HT */
3302                case 0x20:      /* SPACE */
3303                case 0xa0:      /* NBSP */
3304                case 0x1680:    /* OGHAM SPACE MARK */
3305                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3306                case 0x2000:    /* EN QUAD */
3307                case 0x2001:    /* EM QUAD */
3308                case 0x2002:    /* EN SPACE */
3309                case 0x2003:    /* EM SPACE */
3310                case 0x2004:    /* THREE-PER-EM SPACE */
3311                case 0x2005:    /* FOUR-PER-EM SPACE */
3312                case 0x2006:    /* SIX-PER-EM SPACE */
3313                case 0x2007:    /* FIGURE SPACE */
3314                case 0x2008:    /* PUNCTUATION SPACE */
3315                case 0x2009:    /* THIN SPACE */
3316                case 0x200A:    /* HAIR SPACE */
3317                case 0x202f:    /* NARROW NO-BREAK SPACE */
3318                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3319                case 0x3000:    /* IDEOGRAPHIC SPACE */
3320                break;
3321                }
3322              }
3323            break;
3324    
3325            case OP_NOT_VSPACE:
3326            for (i = 1; i <= min; i++)
3327              {
3328              if (eptr >= md->end_subject)
3329                {
3330                SCHECK_PARTIAL();
3331                RRETURN(MATCH_NOMATCH);
3332                }
3333              GETCHARINC(c, eptr);
3334              switch(c)
3335                {
3336                default: break;
3337                case 0x0a:      /* LF */
3338                case 0x0b:      /* VT */
3339                case 0x0c:      /* FF */
3340                case 0x0d:      /* CR */
3341                case 0x85:      /* NEL */
3342                case 0x2028:    /* LINE SEPARATOR */
3343                case 0x2029:    /* PARAGRAPH SEPARATOR */
3344                RRETURN(MATCH_NOMATCH);
3345                }
3346              }
3347            break;
3348    
3349            case OP_VSPACE:
3350            for (i = 1; i <= min; i++)
3351              {
3352              if (eptr >= md->end_subject)
3353                {
3354                SCHECK_PARTIAL();
3355                RRETURN(MATCH_NOMATCH);
3356                }
3357              GETCHARINC(c, eptr);
3358              switch(c)
3359                {
3360                default: RRETURN(MATCH_NOMATCH);
3361                case 0x0a:      /* LF */
3362                case 0x0b:      /* VT */
3363                case 0x0c:      /* FF */
3364                case 0x0d:      /* CR */
3365                case 0x85:      /* NEL */
3366                case 0x2028:    /* LINE SEPARATOR */
3367                case 0x2029:    /* PARAGRAPH SEPARATOR */
3368                break;
3369                }
3370              }
3371            break;
3372    
3373            case OP_NOT_DIGIT:
3374            for (i = 1; i <= min; i++)
3375              {
3376              if (eptr >= md->end_subject)
3377                {
3378                SCHECK_PARTIAL();
3379                RRETURN(MATCH_NOMATCH);
3380                }
3381              GETCHARINC(c, eptr);
3382              if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)
3383              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3384            }            }
3385          break;          break;
# Line 2827  for (;;) Line 3387  for (;;)
3387          case OP_DIGIT:          case OP_DIGIT:
3388          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3389            {            {
3390            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3391               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)              {
3392                SCHECK_PARTIAL();
3393                RRETURN(MATCH_NOMATCH);
3394                }
3395              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)
3396              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3397            /* 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 */
3398            }            }
# Line 2837  for (;;) Line 3401  for (;;)
3401          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
3402          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3403            {            {
3404            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3405               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_space) != 0))              {
3406                SCHECK_PARTIAL();
3407              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3408            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;              }
3409              if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)
3410                RRETURN(MATCH_NOMATCH);
3411              while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3412            }            }
3413          break;          break;
3414    
3415          case OP_WHITESPACE:          case OP_WHITESPACE:
3416          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3417            {            {
3418            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3419               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)              {
3420                SCHECK_PARTIAL();
3421                RRETURN(MATCH_NOMATCH);
3422                }
3423              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)
3424              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3425            /* 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 */
3426            }            }
# Line 2858  for (;;) Line 3430  for (;;)
3430          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3431            {            {
3432            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3433               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_word) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0))
3434              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3435            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3436            }            }
3437          break;          break;
3438    
3439          case OP_WORDCHAR:          case OP_WORDCHAR:
3440          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3441            {            {
3442            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3443               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)              {
3444                SCHECK_PARTIAL();
3445                RRETURN(MATCH_NOMATCH);
3446                }
3447              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)
3448              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3449            /* 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 */
3450            }            }
# Line 2882  for (;;) Line 3458  for (;;)
3458  #endif     /* SUPPORT_UTF8 */  #endif     /* SUPPORT_UTF8 */
3459    
3460        /* 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
3461        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. */  
3462    
3463        switch(ctype)        switch(ctype)
3464          {          {
3465          case OP_ANY:          case OP_ANY:
3466          if ((ims & PCRE_DOTALL) == 0)          for (i = 1; i <= min; i++)
3467            {            {
3468            for (i = 1; i <= min; i++)            if (eptr >= md->end_subject)
3469              {              {
3470              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
3471              eptr++;              RRETURN(MATCH_NOMATCH);
3472              }              }
3473              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
3474              eptr++;
3475            }            }
         else eptr += min;  
3476          break;          break;
3477    
3478          case OP_ANYBYTE:          case OP_ALLANY:
3479            if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
3480          eptr += min;          eptr += min;
3481          break;          break;
3482    
3483          /* Because of the CRLF case, we can't assume the minimum number of          case OP_ANYBYTE:
3484          bytes are present in this case. */          if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
3485            eptr += min;
3486            break;
3487    
3488          case OP_ANYNL:          case OP_ANYNL:
3489          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3490            {            {
3491            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3492                {
3493                SCHECK_PARTIAL();
3494                RRETURN(MATCH_NOMATCH);
3495                }
3496            switch(*eptr++)            switch(*eptr++)
3497              {              {
3498              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
# Line 2917  for (;;) Line 3500  for (;;)
3500              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3501              break;              break;
3502              case 0x000a:              case 0x000a:
3503                break;
3504    
3505              case 0x000b:              case 0x000b:
3506              case 0x000c:              case 0x000c:
3507              case 0x0085:              case 0x0085:
3508                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3509                break;
3510                }
3511              }
3512            break;
3513    
3514            case OP_NOT_HSPACE:
3515            for (i = 1; i <= min; i++)
3516              {
3517              if (eptr >= md->end_subject)
3518                {
3519                SCHECK_PARTIAL();
3520                RRETURN(MATCH_NOMATCH);
3521                }
3522              switch(*eptr++)
3523                {
3524                default: break;
3525                case 0x09:      /* HT */
3526                case 0x20:      /* SPACE */
3527                case 0xa0:      /* NBSP */
3528                RRETURN(MATCH_NOMATCH);
3529                }
3530              }
3531            break;
3532    
3533            case OP_HSPACE:
3534            for (i = 1; i <= min; i++)
3535              {
3536              if (eptr >= md->end_subject)
3537                {
3538                SCHECK_PARTIAL();
3539                RRETURN(MATCH_NOMATCH);
3540                }
3541              switch(*eptr++)
3542                {
3543                default: RRETURN(MATCH_NOMATCH);
3544                case 0x09:      /* HT */
3545                case 0x20:      /* SPACE */
3546                case 0xa0:      /* NBSP */
3547                break;
3548                }
3549              }
3550            break;
3551    
3552            case OP_NOT_VSPACE:
3553            for (i = 1; i <= min; i++)
3554              {
3555              if (eptr >= md->end_subject)
3556                {
3557                SCHECK_PARTIAL();
3558                RRETURN(MATCH_NOMATCH);
3559                }
3560              switch(*eptr++)
3561                {
3562                default: break;
3563                case 0x0a:      /* LF */
3564                case 0x0b:      /* VT */
3565                case 0x0c:      /* FF */
3566                case 0x0d:      /* CR */
3567                case 0x85:      /* NEL */
3568                RRETURN(MATCH_NOMATCH);
3569                }
3570              }
3571            break;
3572    
3573            case OP_VSPACE:
3574            for (i = 1; i <= min; i++)
3575              {
3576              if (eptr >= md->end_subject)
3577                {
3578                SCHECK_PARTIAL();
3579                RRETURN(MATCH_NOMATCH);
3580                }
3581              switch(*eptr++)
3582                {
3583                default: RRETURN(MATCH_NOMATCH);
3584                case 0x0a:      /* LF */
3585                case 0x0b:      /* VT */
3586                case 0x0c:      /* FF */
3587                case 0x0d:      /* CR */
3588                case 0x85:      /* NEL */
3589              break;              break;
3590              }              }
3591            }            }
# Line 2927  for (;;) Line 3593  for (;;)
3593    
3594          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
3595          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3596              {
3597              if (eptr >= md->end_subject)
3598                {
3599                SCHECK_PARTIAL();
3600                RRETURN(MATCH_NOMATCH);
3601                }
3602            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);
3603              }
3604          break;          break;
3605    
3606          case OP_DIGIT:          case OP_DIGIT:
3607          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3608              {
3609              if (eptr >= md->end_subject)
3610                {
3611                SCHECK_PARTIAL();
3612                RRETURN(MATCH_NOMATCH);
3613                }
3614            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);
3615              }
3616          break;          break;
3617    
3618          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
3619          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3620              {
3621              if (eptr >= md->end_subject)
3622                {
3623                SCHECK_PARTIAL();
3624                RRETURN(MATCH_NOMATCH);
3625                }
3626            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);
3627              }
3628          break;          break;
3629    
3630          case OP_WHITESPACE:          case OP_WHITESPACE:
3631          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3632              {
3633              if (eptr >= md->end_subject)
3634                {
3635                SCHECK_PARTIAL();
3636                RRETURN(MATCH_NOMATCH);
3637                }
3638            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);
3639              }
3640          break;          break;
3641    
3642          case OP_NOT_WORDCHAR:          case OP_NOT_WORDCHAR:
3643          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3644              {
3645              if (eptr >= md->end_subject)
3646                {
3647                SCHECK_PARTIAL();
3648                RRETURN(MATCH_NOMATCH);
3649                }
3650            if ((md->ctypes[*eptr++] & ctype_word) != 0)            if ((md->ctypes[*eptr++] & ctype_word) != 0)
3651              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3652              }
3653          break;          break;
3654    
3655          case OP_WORDCHAR:          case OP_WORDCHAR:
3656          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3657              {
3658              if (eptr >= md->end_subject)
3659                {
3660                SCHECK_PARTIAL();
3661                RRETURN(MATCH_NOMATCH);
3662                }
3663            if ((md->ctypes[*eptr++] & ctype_word) == 0)            if ((md->ctypes[*eptr++] & ctype_word) == 0)
3664              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3665              }
3666          break;          break;
3667    
3668          default:          default:
# Line 2982  for (;;) Line 3690  for (;;)
3690              {              {
3691              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);
3692              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3693              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max)
3694                  {
3695                  CHECK_PARTIAL();
3696                  RRETURN(MATCH_NOMATCH);
3697                  }
3698                if (eptr >= md->end_subject)
3699                  {
3700                  SCHECK_PARTIAL();
3701                  RRETURN(MATCH_NOMATCH);
3702                  }
3703              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3704              if (prop_fail_result) RRETURN(MATCH_NOMATCH);              if (prop_fail_result) RRETURN(MATCH_NOMATCH);
3705              }              }
# Line 2993  for (;;) Line 3710  for (;;)
3710              {              {
3711              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);
3712              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3713              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max)
3714                  {
3715                  CHECK_PARTIAL();
3716                  RRETURN(MATCH_NOMATCH);
3717                  }
3718                if (eptr >= md->end_subject)
3719                  {
3720                  SCHECK_PARTIAL();
3721                  RRETURN(MATCH_NOMATCH);
3722                  }
3723              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3724              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
3725              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
3726                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
3727                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 3008  for (;;) Line 3734  for (;;)
3734              {              {
3735              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);
3736              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3737              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max)
3738                  {
3739                  CHECK_PARTIAL();
3740                  RRETURN(MATCH_NOMATCH);
3741                  }
3742                if (eptr >= md->end_subject)
3743                  {
3744                  SCHECK_PARTIAL();
3745                  RRETURN(MATCH_NOMATCH);
3746                  }
3747              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3748              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
3749              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
3750                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3751              }              }
# Line 3021  for (;;) Line 3756  for (;;)
3756              {              {
3757              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);
3758              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3759              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max)
3760                  {
3761                  CHECK_PARTIAL();
3762                  RRETURN(MATCH_NOMATCH);
3763                  }
3764                if (eptr >= md->end_subject)
3765                  {
3766                  SCHECK_PARTIAL();
3767                  RRETURN(MATCH_NOMATCH);
3768                  }
3769              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3770              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
3771              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
3772                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3773              }              }
# Line 3034  for (;;) Line 3778  for (;;)
3778              {              {
3779              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);
3780              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3781              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max)
3782                  {
3783                  CHECK_PARTIAL();
3784                  RRETURN(MATCH_NOMATCH);
3785                  }
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_script = UCD_SCRIPT(c);
3793              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
3794                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3795              }              }
# Line 3056  for (;;) Line 3809  for (;;)
3809            {            {
3810            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);
3811            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3812            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max)
3813                {
3814                CHECK_PARTIAL();
3815                RRETURN(MATCH_NOMATCH);
3816                }
3817              if (eptr >= md->end_subject)
3818                {
3819                SCHECK_PARTIAL();
3820                RRETURN(MATCH_NOMATCH);
3821                }
3822            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3823            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
3824            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);
3825            while (eptr < md->end_subject)            while (eptr < md->end_subject)
3826              {              {
3827              int len = 1;              int len = 1;
3828              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
3829                {                else { GETCHARLEN(c, eptr, len); }
3830                GETCHARLEN(c, eptr, len);              prop_category = UCD_CATEGORY(c);
               }  
             prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);  
3831              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
3832              eptr += len;              eptr += len;
3833              }              }
# Line 3085  for (;;) Line 3845  for (;;)
3845            {            {
3846            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);
3847            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3848            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max)
3849                 (ctype == OP_ANY && (ims & PCRE_DOTALL) == 0 &&              {
3850                  IS_NEWLINE(eptr)))              CHECK_PARTIAL();
3851                RRETURN(MATCH_NOMATCH);
3852                }
3853              if (eptr >= md->end_subject)
3854                {
3855                SCHECK_PARTIAL();
3856                RRETURN(MATCH_NOMATCH);
3857                }
3858              if (ctype == OP_ANY && IS_NEWLINE(eptr))
3859              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
   
3860            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3861            switch(ctype)            switch(ctype)
3862              {              {
3863              case OP_ANY:        /* This is the DOTALL case */              case OP_ANY:        /* This is the non-NL case */
3864              break;              case OP_ALLANY:
   
3865              case OP_ANYBYTE:              case OP_ANYBYTE:
3866              break;              break;
3867    
# Line 3107  for (;;) Line 3873  for (;;)
3873                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3874                break;                break;
3875                case 0x000a:                case 0x000a:
3876                  break;
3877    
3878                case 0x000b:                case 0x000b:
3879                case 0x000c:                case 0x000c:
3880                case 0x0085:                case 0x0085:
3881                case 0x2028:                case 0x2028:
3882                case 0x2029:                case 0x2029:
3883                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3884                  break;
3885                  }
3886                break;
3887    
3888                case OP_NOT_HSPACE:
3889                switch(c)
3890                  {
3891                  default: break;
3892                  case 0x09:      /* HT */
3893                  case 0x20:      /* SPACE */
3894                  case 0xa0:      /* NBSP */
3895                  case 0x1680:    /* OGHAM SPACE MARK */
3896                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3897                  case 0x2000:    /* EN QUAD */
3898                  case 0x2001:    /* EM QUAD */
3899                  case 0x2002:    /* EN SPACE */
3900                  case 0x2003:    /* EM SPACE */
3901                  case 0x2004:    /* THREE-PER-EM SPACE */
3902                  case 0x2005:    /* FOUR-PER-EM SPACE */
3903                  case 0x2006:    /* SIX-PER-EM SPACE */
3904                  case 0x2007:    /* FIGURE SPACE */
3905                  case 0x2008:    /* PUNCTUATION SPACE */
3906                  case 0x2009:    /* THIN SPACE */
3907                  case 0x200A:    /* HAIR SPACE */
3908                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3909                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3910                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3911                  RRETURN(MATCH_NOMATCH);
3912                  }
3913                break;
3914    
3915                case OP_HSPACE:
3916                switch(c)
3917                  {
3918                  default: RRETURN(MATCH_NOMATCH);
3919                  case 0x09:      /* HT */
3920                  case 0x20:      /* SPACE */
3921                  case 0xa0:      /* NBSP */
3922                  case 0x1680:    /* OGHAM SPACE MARK */
3923                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3924                  case 0x2000:    /* EN QUAD */
3925                  case 0x2001:    /* EM QUAD */
3926                  case 0x2002:    /* EN SPACE */
3927                  case 0x2003:    /* EM SPACE */
3928                  case 0x2004:    /* THREE-PER-EM SPACE */
3929                  case 0x2005:    /* FOUR-PER-EM SPACE */
3930                  case 0x2006:    /* SIX-PER-EM SPACE */
3931                  case 0x2007:    /* FIGURE SPACE */
3932                  case 0x2008:    /* PUNCTUATION SPACE */
3933                  case 0x2009:    /* THIN SPACE */
3934                  case 0x200A:    /* HAIR SPACE */
3935                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3936                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3937                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3938                  break;
3939                  }
3940                break;
3941    
3942                case OP_NOT_VSPACE:
3943                switch(c)
3944                  {
3945                  default: break;
3946                  case 0x0a:      /* LF */
3947                  case 0x0b:      /* VT */
3948                  case 0x0c:      /* FF */
3949                  case 0x0d:      /* CR */
3950                  case 0x85:      /* NEL */
3951                  case 0x2028:    /* LINE SEPARATOR */
3952                  case 0x2029:    /* PARAGRAPH SEPARATOR */
3953                  RRETURN(MATCH_NOMATCH);
3954                  }
3955                break;
3956    
3957                case OP_VSPACE:
3958                switch(c)
3959                  {
3960                  default: RRETURN(MATCH_NOMATCH);
3961                  case 0x0a:      /* LF */
3962                  case 0x0b:      /* VT */
3963                  case 0x0c:      /* FF */
3964                  case 0x0d:      /* CR */
3965                  case 0x85:      /* NEL */
3966                  case 0x2028:    /* LINE SEPARATOR */
3967                  case 0x2029:    /* PARAGRAPH SEPARATOR */
3968                break;                break;
3969                }                }
3970              break;              break;
# Line 3159  for (;;) Line 4012  for (;;)
4012            {            {
4013            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);
4014            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4015            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max)
4016                 ((ims & PCRE_DOTALL) == 0 && IS_NEWLINE(eptr)))              {
4017                CHECK_PARTIAL();
4018                RRETURN(MATCH_NOMATCH);
4019                }
4020              if (eptr >= md->end_subject)
4021                {
4022                SCHECK_PARTIAL();
4023                RRETURN(MATCH_NOMATCH);
4024                }
4025              if (ctype == OP_ANY && IS_NEWLINE(eptr))
4026              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
   
4027            c = *eptr++;            c = *eptr++;
4028            switch(ctype)            switch(ctype)
4029              {              {
4030              case OP_ANY:   /* This is the DOTALL case */              case OP_ANY:     /* This is the non-NL case */
4031              break;              case OP_ALLANY:
   
4032              case OP_ANYBYTE:              case OP_ANYBYTE:
4033              break;              break;
4034    
# Line 3179  for (;;) Line 4039  for (;;)
4039                case 0x000d:                case 0x000d:
4040                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
4041                break;                break;
4042    
4043                case 0x000a:                case 0x000a:
4044                  break;
4045    
4046                case 0x000b:                case 0x000b:
4047                case 0x000c:                case 0x000c:
4048                case 0x0085:                case 0x0085:
4049                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4050                  break;
4051                  }
4052                break;
4053    
4054                case OP_NOT_HSPACE:
4055                switch(c)
4056                  {
4057                  default: break;
4058                  case 0x09:      /* HT */
4059                  case 0x20:      /* SPACE */
4060                  case 0xa0:      /* NBSP */
4061                  RRETURN(MATCH_NOMATCH);
4062                  }
4063                break;
4064    
4065                case OP_HSPACE:
4066                switch(c)
4067                  {
4068                  default: RRETURN(MATCH_NOMATCH);
4069                  case 0x09:      /* HT */
4070                  case 0x20:      /* SPACE */
4071                  case 0xa0:      /* NBSP */
4072                  break;
4073                  }
4074                break;
4075    
4076                case OP_NOT_VSPACE:
4077                switch(c)
4078                  {
4079                  default: break;
4080                  case 0x0a:      /* LF */
4081                  case 0x0b:      /* VT */
4082                  case 0x0c:      /* FF */
4083                  case 0x0d:      /* CR */
4084                  case 0x85:      /* NEL */
4085                  RRETURN(MATCH_NOMATCH);
4086                  }
4087                break;
4088    
4089                case OP_VSPACE:
4090                switch(c)
4091                  {
4092                  default: RRETURN(MATCH_NOMATCH);
4093                  case 0x0a:      /* LF */
4094                  case 0x0b:      /* VT */
4095                  case 0x0c:      /* FF */
4096                  case 0x0d:      /* CR */
4097                  case 0x85:      /* NEL */
4098                break;                break;
4099                }                }
4100              break;              break;
# Line 3249  for (;;) Line 4161  for (;;)
4161              int len = 1;              int len = 1;
4162              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4163              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4164              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
4165              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
4166                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
4167                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 3264  for (;;) Line 4176  for (;;)
4176              int len = 1;              int len = 1;
4177              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4178              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4179              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4180              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
4181                break;                break;
4182              eptr+= len;              eptr+= len;
# Line 3277  for (;;) Line 4189  for (;;)
4189              int len = 1;              int len = 1;
4190              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4191              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4192              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
4193              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
4194                break;                break;
4195              eptr+= len;              eptr+= len;
# Line 3290  for (;;) Line 4202  for (;;)
4202              int len = 1;              int len = 1;
4203              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4204              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4205              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_script = UCD_SCRIPT(c);
4206              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
4207                break;                break;
4208              eptr+= len;              eptr+= len;
# Line 3300  for (;;) Line 4212  for (;;)
4212    
4213          /* eptr is now past the end of the maximum run */          /* eptr is now past the end of the maximum run */
4214    
4215            CHECK_PARTIAL();
4216          if (possessive) continue;          if (possessive) continue;
4217          for(;;)          for(;;)
4218            {            {
4219            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);
4220            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4221            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
4222            BACKCHAR(eptr);            if (utf8) BACKCHAR(eptr);
4223            }            }
4224          }          }
4225    
# Line 3319  for (;;) Line 4232  for (;;)
4232            {            {
4233            if (eptr >= md->end_subject) break;            if (eptr >= md->end_subject) break;
4234            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
4235            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
4236            if (prop_category == ucp_M) break;            if (prop_category == ucp_M) break;
4237            while (eptr < md->end_subject)            while (eptr < md->end_subject)
4238              {              {
# Line 3328  for (;;) Line 4241  for (;;)
4241                {                {
4242                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
4243                }                }
4244              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4245              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
4246              eptr += len;              eptr += len;
4247              }              }
# Line 3336  for (;;) Line 4249  for (;;)
4249    
4250          /* eptr is now past the end of the maximum run */          /* eptr is now past the end of the maximum run */
4251    
4252            CHECK_PARTIAL();
4253          if (possessive) continue;          if (possessive) continue;
4254          for(;;)          for(;;)
4255            {            {
# Line 3345  for (;;) Line 4259  for (;;)
4259            for (;;)                        /* Move back over one extended */            for (;;)                        /* Move back over one extended */
4260              {              {
4261              int len = 1;              int len = 1;
             BACKCHAR(eptr);  
4262              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr; else
4263                {                {
4264                  BACKCHAR(eptr);
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--;              eptr--;
4270              }              }
# Line 3368  for (;;) Line 4282  for (;;)
4282          switch(ctype)          switch(ctype)
4283            {            {
4284            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. */  
   
4285            if (max < INT_MAX)            if (max < INT_MAX)
4286              {              {
4287              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  
4288                {                {
4289                for (i = min; i < max; i++)                if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
4290                  {                eptr++;
4291                  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++;  
                 }  
4292                }                }
4293              }              }
4294    
# Line 3399  for (;;) Line 4296  for (;;)
4296    
4297            else            else
4298              {              {
4299              if ((ims & PCRE_DOTALL) == 0)              for (i = min; i < max; i++)
4300                {                {
4301                for (i = min; i < max; i++)                if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
4302                  {                eptr++;
4303                  if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
                 eptr++;  
                 }  
               break;  
4304                }                }
4305              else              }
4306              break;
4307    
4308              case OP_ALLANY:
4309              if (max < INT_MAX)
4310                {
4311                for (i = min; i < max; i++)
4312                {                {
4313                c = max - min;                if (eptr >= md->end_subject) break;
4314                if (c > (unsigned int)(md->end_subject - eptr))                eptr++;
4315                  c = md->end_subject - eptr;                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
               eptr += c;  
4316                }                }
4317              }              }
4318              else eptr = md->end_subject;   /* Unlimited UTF-8 repeat */
4319            break;            break;
4320    
4321            /* The byte case is the same as non-UTF8 */            /* The byte case is the same as non-UTF8 */
# Line 3440  for (;;) Line 4340  for (;;)
4340                }                }
4341              else              else
4342                {                {
4343                if (c != 0x000a && c != 0x000b && c != 0x000c &&                if (c != 0x000a &&
4344                    c != 0x0085 && c != 0x2028 && c != 0x2029)                    (md->bsr_anycrlf ||
4345                       (c != 0x000b && c != 0x000c &&
4346                        c != 0x0085 && c != 0x2028 && c != 0x2029)))
4347                  break;                  break;
4348                eptr += len;                eptr += len;
4349                }                }
4350              }              }
4351            break;            break;
4352    
4353              case OP_NOT_HSPACE:
4354              case OP_HSPACE:
4355              for (i = min; i < max; i++)
4356                {
4357                BOOL gotspace;
4358                int len = 1;
4359                if (eptr >= md->end_subject) break;
4360                GETCHARLEN(c, eptr, len);
4361                switch(c)
4362                  {
4363                  default: gotspace = FALSE; break;
4364                  case 0x09:      /* HT */
4365                  case 0x20:      /* SPACE */
4366                  case 0xa0:      /* NBSP */
4367                  case 0x1680:    /* OGHAM SPACE MARK */
4368                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
4369                  case 0x2000:    /* EN QUAD */
4370                  case 0x2001:    /* EM QUAD */
4371                  case 0x2002:    /* EN SPACE */
4372                  case 0x2003:    /* EM SPACE */
4373                  case 0x2004:    /* THREE-PER-EM SPACE */
4374                  case 0x2005:    /* FOUR-PER-EM SPACE */
4375                  case 0x2006:    /* SIX-PER-EM SPACE */
4376                  case 0x2007:    /* FIGURE SPACE */
4377                  case 0x2008:    /* PUNCTUATION SPACE */
4378                  case 0x2009:    /* THIN SPACE */
4379                  case 0x200A:    /* HAIR SPACE */
4380                  case 0x202f:    /* NARROW NO-BREAK SPACE */
4381                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4382                  case 0x3000:    /* IDEOGRAPHIC SPACE */
4383                  gotspace = TRUE;
4384                  break;
4385                  }
4386                if (gotspace == (ctype == OP_NOT_HSPACE)) break;
4387                eptr += len;
4388                }
4389              break;
4390    
4391              case OP_NOT_VSPACE:
4392              case OP_VSPACE:
4393              for (i = min; i < max; i++)
4394                {
4395                BOOL gotspace;
4396                int len = 1;
4397                if (eptr >= md->end_subject) break;
4398                GETCHARLEN(c, eptr, len);
4399                switch(c)
4400                  {
4401                  default: gotspace = FALSE; break;
4402                  case 0x0a:      /* LF */
4403                  case 0x0b:      /* VT */
4404                  case 0x0c:      /* FF */
4405                  case 0x0d:      /* CR */
4406                  case 0x85:      /* NEL */
4407                  case 0x2028:    /* LINE SEPARATOR */
4408                  case 0x2029:    /* PARAGRAPH SEPARATOR */
4409                  gotspace = TRUE;
4410                  break;
4411                  }
4412                if (gotspace == (ctype == OP_NOT_VSPACE)) break;
4413                eptr += len;
4414                }
4415              break;
4416    
4417            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
4418            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4419              {              {
# Line 3520  for (;;) Line 4486  for (;;)
4486    
4487          /* eptr is now past the end of the maximum run */          /* eptr is now past the end of the maximum run */
4488    
4489            CHECK_PARTIAL();
4490          if (possessive) continue;          if (possessive) continue;
4491          for(;;)          for(;;)
4492            {            {
# Line 3530  for (;;) Line 4497  for (;;)
4497            }            }
4498          }          }
4499        else        else
4500  #endif  #endif  /* SUPPORT_UTF8 */
4501    
4502        /* Not UTF-8 mode */        /* Not UTF-8 mode */
4503          {          {
4504          switch(ctype)          switch(ctype)
4505            {            {
4506            case OP_ANY:            case OP_ANY:
4507            if ((ims & PCRE_DOTALL) == 0)            for (i = min; i < max; i++)
4508              {              {
4509              for (i = min; i < max; i++)              if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
4510                {              eptr++;
               if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;  
               eptr++;  
               }  
             break;  
4511              }              }
4512            /* For DOTALL case, fall through and treat as \C */            break;
4513    
4514              case OP_ALLANY:
4515            case OP_ANYBYTE:            case OP_ANYBYTE:
4516            c = max - min;            c = max - min;
4517            if (c > (unsigned int)(md->end_subject - eptr))            if (c > (unsigned int)(md->end_subject - eptr))
# Line 3567  for (;;) Line 4531  for (;;)
4531                }                }
4532              else              else
4533                {                {
4534                if (c != 0x000a && c != 0x000b && c != 0x000c && c != 0x0085)                if (c != 0x000a &&
4535                      (md->bsr_anycrlf ||
4536                        (c != 0x000b && c != 0x000c && c != 0x0085)))
4537                  break;                  break;
4538                eptr++;                eptr++;
4539                }                }
4540              }              }
4541            break;            break;
4542    
4543              case OP_NOT_HSPACE:
4544              for (i = min; i < max; i++)
4545                {
4546                if (eptr >= md->end_subject) break;
4547                c = *eptr;
4548                if (c == 0x09 || c == 0x20 || c == 0xa0) break;
4549                eptr++;
4550                }
4551              break;
4552    
4553              case OP_HSPACE:
4554              for (i = min; i < max; i++)
4555                {
4556                if (eptr >= md->end_subject) break;
4557                c = *eptr;
4558                if (c != 0x09 && c != 0x20 && c != 0xa0) break;
4559                eptr++;
4560                }
4561              break;
4562    
4563              case OP_NOT_VSPACE:
4564              for (i = min; i < max; i++)
4565                {
4566                if (eptr >= md->end_subject) break;
4567                c = *eptr;
4568                if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)
4569                  break;
4570                eptr++;
4571                }
4572              break;
4573    
4574              case OP_VSPACE:
4575              for (i = min; i < max; i++)
4576                {
4577                if (eptr >= md->end_subject) break;
4578                c = *eptr;
4579                if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85)
4580                  break;
4581                eptr++;
4582                }
4583              break;
4584    
4585            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
4586            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4587              {              {
# Line 3634  for (;;) Line 4642  for (;;)
4642    
4643          /* eptr is now past the end of the maximum run */          /* eptr is now past the end of the maximum run */
4644    
4645            CHECK_PARTIAL();
4646          if (possessive) continue;          if (possessive) continue;
4647          while (eptr >= pp)          while (eptr >= pp)
4648            {            {
# Line 3675  HEAP_RETURN: Line 4684  HEAP_RETURN:
4684  switch (frame->Xwhere)  switch (frame->Xwhere)
4685    {    {
4686    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)
4687    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)
4688    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)
4689    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)
4690    LBL(33) LBL(34) LBL(35) LBL(36) LBL(37) LBL(38) LBL(39) LBL(40)    LBL(53) LBL(54)
4691    LBL(41) LBL(42) LBL(43) LBL(44) LBL(45) LBL(46) LBL(47)  #ifdef SUPPORT_UTF8
4692      LBL(16) LBL(18) LBL(20) LBL(21) LBL(22) LBL(23) LBL(28) LBL(30)
4693      LBL(32) LBL(34) LBL(42) LBL(46)
4694    #ifdef SUPPORT_UCP
4695      LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
4696    #endif  /* SUPPORT_UCP */
4697    #endif  /* SUPPORT_UTF8 */
4698    default:    default:
4699    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));
4700    return PCRE_ERROR_INTERNAL;    return PCRE_ERROR_INTERNAL;
# Line 3771  Returns:          > 0 => success; value Line 4786  Returns:          > 0 => success; value
4786                   < -1 => some kind of unexpected problem                   < -1 => some kind of unexpected problem
4787  */  */
4788    
4789  PCRE_EXP_DEFN int  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
4790  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,
4791    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,
4792    int offsetcount)    int offsetcount)
# Line 3795  const uschar *tables; Line 4810  const uschar *tables;
4810  const uschar *start_bits = NULL;  const uschar *start_bits = NULL;
4811  USPTR start_match = (USPTR)subject + start_offset;  USPTR start_match = (USPTR)subject + start_offset;
4812  USPTR end_subject;  USPTR end_subject;
4813    USPTR start_partial = NULL;
4814  USPTR req_byte_ptr = start_match - 1;  USPTR req_byte_ptr = start_match - 1;
 eptrblock eptrchain[EPTR_WORK_SIZE];  
4815    
4816  pcre_study_data internal_study;  pcre_study_data internal_study;
4817  const pcre_study_data *study;  const pcre_study_data *study;
# Line 3859  if (re->magic_number != MAGIC_NUMBER) Line 4874  if (re->magic_number != MAGIC_NUMBER)
4874  /* Set up other data */  /* Set up other data */
4875    
4876  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
4877  startline = (re->options & PCRE_STARTLINE) != 0;  startline = (re->flags & PCRE_STARTLINE) != 0;
4878  firstline = (re->options & PCRE_FIRSTLINE) != 0;  firstline = (re->options & PCRE_FIRSTLINE) != 0;
4879    
4880  /* 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 3874  end_subject = md->end_subject; Line 4889  end_subject = md->end_subject;
4889    
4890  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;
4891  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;
4892    md->jscript_compat = (re->options & PCRE_JAVASCRIPT_COMPAT) != 0;
4893    
4894  md->notbol = (options & PCRE_NOTBOL) != 0;  md->notbol = (options & PCRE_NOTBOL) != 0;
4895  md->noteol = (options & PCRE_NOTEOL) != 0;  md->noteol = (options & PCRE_NOTEOL) != 0;
# Line 3882  md->partial = (options & PCRE_PARTIAL) ! Line 4898  md->partial = (options & PCRE_PARTIAL) !
4898  md->hitend = FALSE;  md->hitend = FALSE;
4899    
4900  md->recursive = NULL;                   /* No recursion at top level */  md->recursive = NULL;                   /* No recursion at top level */
 md->eptrchain = eptrchain;              /* Make workspace generally available */  
4901    
4902  md->lcc = tables + lcc_offset;  md->lcc = tables + lcc_offset;
4903  md->ctypes = tables + ctypes_offset;  md->ctypes = tables + ctypes_offset;
4904    
4905    /* Handle different \R options. */
4906    
4907    switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
4908      {
4909      case 0:
4910      if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
4911        md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0;
4912      else
4913    #ifdef BSR_ANYCRLF
4914      md->bsr_anycrlf = TRUE;
4915    #else
4916      md->bsr_anycrlf = FALSE;
4917    #endif
4918      break;
4919    
4920      case PCRE_BSR_ANYCRLF:
4921      md->bsr_anycrlf = TRUE;
4922      break;
4923    
4924      case PCRE_BSR_UNICODE:
4925      md->bsr_anycrlf = FALSE;
4926      break;
4927    
4928      default: return PCRE_ERROR_BADNEWLINE;
4929      }
4930    
4931  /* Handle different types of newline. The three bits give eight cases. If  /* Handle different types of newline. The three bits give eight cases. If
4932  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. */
4933    
4934  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
4935         PCRE_NEWLINE_BITS)          (pcre_uint32)options) & PCRE_NEWLINE_BITS)
4936    {    {
4937    case 0: newline = NEWLINE; break;   /* Compile-time default */    case 0: newline = NEWLINE; break;   /* Compile-time default */
4938    case PCRE_NEWLINE_CR: newline = '\r'; break;    case PCRE_NEWLINE_CR: newline = CHAR_CR; break;
4939    case PCRE_NEWLINE_LF: newline = '\n'; break;    case PCRE_NEWLINE_LF: newline = CHAR_NL; break;
4940    case PCRE_NEWLINE_CR+    case PCRE_NEWLINE_CR+
4941         PCRE_NEWLINE_LF: newline = ('\r' << 8) | '\n'; break;         PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break;
4942    case PCRE_NEWLINE_ANY: newline = -1; break;    case PCRE_NEWLINE_ANY: newline = -1; break;
4943    case PCRE_NEWLINE_ANYCRLF: newline = -2; break;    case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
4944    default: return PCRE_ERROR_BADNEWLINE;    default: return PCRE_ERROR_BADNEWLINE;
# Line 3927  else Line 4968  else
4968      }      }
4969    }    }
4970    
4971  /* Partial matching is supported only for a restricted set of regexes at the  /* Partial matching was originally supported only for a restricted set of
4972  moment. */  regexes; from release 8.00 there are no restrictions, but the bits are still
4973    defined (though never set). So there's no harm in leaving this code. */
4974    
4975  if (md->partial && (re->options & PCRE_NOPARTIAL) != 0)  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
4976    return PCRE_ERROR_BADPARTIAL;    return PCRE_ERROR_BADPARTIAL;
4977    
4978  /* 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 3939  back the character offset. */ Line 4981  back the character offset. */
4981  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
4982  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)
4983    {    {
4984    if (_pcre_valid_utf8((uschar *)subject, length) >= 0)    if (_pcre_valid_utf8((USPTR)subject, length) >= 0)
4985      return PCRE_ERROR_BADUTF8;      return PCRE_ERROR_BADUTF8;
4986    if (start_offset > 0 && start_offset < length)    if (start_offset > 0 && start_offset < length)
4987      {      {
4988      int tb = ((uschar *)subject)[start_offset];      int tb = ((USPTR)subject)[start_offset];
4989      if (tb > 127)      if (tb > 127)
4990        {        {
4991        tb &= 0xc0;        tb &= 0xc0;
# Line 4007  studied, there may be a bitmap of possib Line 5049  studied, there may be a bitmap of possib
5049    
5050  if (!anchored)  if (!anchored)
5051    {    {
5052    if ((re->options & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
5053      {      {
5054      first_byte = re->first_byte & 255;      first_byte = re->first_byte & 255;
5055      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)
# Line 4022  if (!anchored) Line 5064  if (!anchored)
5064  /* For anchored or unanchored matches, there may be a "last known required  /* For anchored or unanchored matches, there may be a "last known required
5065  character" set. */  character" set. */
5066    
5067  if ((re->options & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
5068    {    {
5069    req_byte = re->req_byte & 255;    req_byte = re->req_byte & 255;
5070    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;
# Line 4038  the loop runs just once. */ Line 5080  the loop runs just once. */
5080  for(;;)  for(;;)
5081    {    {
5082    USPTR save_end_subject = end_subject;    USPTR save_end_subject = end_subject;
5083      USPTR new_start_match;
5084    
5085    /* Reset the maximum number of extractions we might see. */    /* Reset the maximum number of extractions we might see. */
5086    
# Line 4048  for(;;) Line 5091  for(;;)
5091      while (iptr < iend) *iptr++ = -1;      while (iptr < iend) *iptr++ = -1;
5092      }      }
5093    
5094    /* 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
5095    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
5096    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
5097    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
5098    the match fails at the newline, later code breaks this loop. */    this loop. */
5099    
5100    if (firstline)    if (firstline)
5101      {      {
5102      USPTR t = start_match;      USPTR t = start_match;
5103    #ifdef SUPPORT_UTF8
5104        if (utf8)
5105          {
5106          while (t < md->end_subject && !IS_NEWLINE(t))
5107            {
5108            t++;
5109            while (t < end_subject && (*t & 0xc0) == 0x80) t++;
5110            }
5111          }
5112        else
5113    #endif
5114      while (t < md->end_subject && !IS_NEWLINE(t)) t++;      while (t < md->end_subject && !IS_NEWLINE(t)) t++;
5115      end_subject = t;      end_subject = t;
5116      }      }
5117    
5118    /* Now test for a unique first byte */    /* There are some optimizations that avoid running the match if a known
5119      starting point is not found, or if a known later character is not present.
5120      However, there is an option that disables these, for testing and for ensuring
5121      that all callouts do actually occur. */
5122    
5123    if (first_byte >= 0)    if ((options & PCRE_NO_START_OPTIMIZE) == 0)
5124      {      {
5125      if (first_byte_caseless)      /* Advance to a unique first byte if there is one. */
5126        while (start_match < end_subject &&  
5127               md->lcc[*start_match] != first_byte)      if (first_byte >= 0)
5128          start_match++;        {
5129      else        if (first_byte_caseless)
5130        while (start_match < end_subject && *start_match != first_byte)          while (start_match < end_subject && md->lcc[*start_match] != first_byte)
5131          start_match++;            start_match++;
5132      }        else
5133            while (start_match < end_subject && *start_match != first_byte)
5134              start_match++;
5135          }
5136    
5137    /* Or to just after a linebreak for a multiline match if possible */      /* Or to just after a linebreak for a multiline match */
5138    
5139    else if (startline)      else if (startline)
     {  
     if (start_match > md->start_subject + start_offset)  
5140        {        {
5141        while (start_match <= end_subject && !WAS_NEWLINE(start_match))        if (start_match > md->start_subject + start_offset)
5142          start_match++;          {
5143    #ifdef SUPPORT_UTF8
5144            if (utf8)
5145              {
5146              while (start_match < end_subject && !WAS_NEWLINE(start_match))
5147                {
5148                start_match++;
5149                while(start_match < end_subject && (*start_match & 0xc0) == 0x80)
5150                  start_match++;
5151                }
5152              }
5153            else
5154    #endif
5155            while (start_match < end_subject && !WAS_NEWLINE(start_match))
5156              start_match++;
5157    
5158        /* If we have just passed a CR and the newline option is ANY or ANYCRLF,          /* If we have just passed a CR and the newline option is ANY or ANYCRLF,
5159        and we are now at a LF, advance the match position by one more character.          and we are now at a LF, advance the match position by one more character.
5160        */          */
5161    
5162        if (start_match[-1] == '\r' &&          if (start_match[-1] == CHAR_CR &&
5163             (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&               (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
5164             start_match < end_subject &&               start_match < end_subject &&
5165             *start_match == '\n')               *start_match == CHAR_NL)
5166          start_match++;            start_match++;
5167            }
5168        }        }
     }  
5169    
5170    /* Or to a non-unique first char after study */      /* Or to a non-unique first byte after study */
5171    
5172    else if (start_bits != NULL)      else if (start_bits != NULL)
     {  
     while (start_match < end_subject)  
5173        {        {
5174        register unsigned int c = *start_match;        while (start_match < end_subject)
5175        if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; else break;          {
5176            register unsigned int c = *start_match;
5177            if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++;
5178              else break;
5179            }
5180        }        }
5181      }      }   /* Starting optimizations */
5182    
5183    /* Restore fudged end_subject */    /* Restore fudged end_subject */
5184    
# Line 4116  for(;;) Line 5190  for(;;)
5190    printf("\n");    printf("\n");
5191  #endif  #endif
5192    
5193    /* 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
5194    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
5195    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.
5196    optimization can save a huge amount of backtracking in patterns with nested    This optimization can save a huge amount of backtracking in patterns with
5197    unlimited repeats that aren't going to match. Writing separate code for    nested unlimited repeats that aren't going to match. Writing separate code
5198    cased/caseless versions makes it go faster, as does using an autoincrement    for cased/caseless versions makes it go faster, as does using an
5199    and backing off on a match.    autoincrement and backing off on a match.
5200    
5201    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
5202    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.
5203    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
5204    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
5205      long.
5206    
5207    ALSO: this processing is disabled when partial matching is requested.    ALSO: this processing is disabled when partial matching is requested, or if
5208    */    disabling is explicitly requested. */
5209    
5210    if (req_byte >= 0 &&    if ((options & PCRE_NO_START_OPTIMIZE) == 0 &&
5211          req_byte >= 0 &&
5212        end_subject - start_match < REQ_BYTE_MAX &&        end_subject - start_match < REQ_BYTE_MAX &&
5213        !md->partial)        !md->partial)
5214      {      {
# Line 4176  for(;;) Line 5252  for(;;)
5252        }        }
5253      }      }
5254    
5255    /* OK, we can now run the match. */    /* OK, we can now run the match. If "hitend" is set afterwards, remember the
5256      first starting point for which a partial match was found. */
5257    
5258    md->start_match_ptr = start_match;      /* Insurance */    md->start_match_ptr = start_match;
5259    md->match_call_count = 0;    md->match_call_count = 0;
5260    md->eptrn = 0;                          /* Next free eptrchain slot */    rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0);
5261    rc = match(start_match, md->start_code, start_match, 2, md,    if (md->hitend && start_partial == NULL) start_partial = start_match;
5262      ims, NULL, 0, 0);  
5263      switch(rc)
5264        {
5265        /* NOMATCH and PRUNE advance by one character. THEN at this level acts
5266        exactly like PRUNE. */
5267    
5268        case MATCH_NOMATCH:
5269        case MATCH_PRUNE:
5270        case MATCH_THEN:
5271        new_start_match = start_match + 1;
5272    #ifdef SUPPORT_UTF8
5273        if (utf8)
5274          while(new_start_match < end_subject && (*new_start_match & 0xc0) == 0x80)
5275            new_start_match++;
5276    #endif
5277        break;
5278    
5279        /* SKIP passes back the next starting point explicitly. */
5280    
5281    /* Any return other than MATCH_NOMATCH breaks the loop. */      case MATCH_SKIP:
5282        new_start_match = md->start_match_ptr;
5283        break;
5284    
5285        /* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */
5286    
5287        case MATCH_COMMIT:
5288        rc = MATCH_NOMATCH;
5289        goto ENDLOOP;
5290    
5291        /* Any other return is some kind of error. */
5292    
5293        default:
5294        goto ENDLOOP;
5295        }
5296    
5297      /* Control reaches here for the various types of "no match at this point"
5298      result. Reset the code to MATCH_NOMATCH for subsequent checking. */
5299    
5300    if (rc != MATCH_NOMATCH) break;    rc = MATCH_NOMATCH;
5301    
5302    /* 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
5303    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 4194  for(;;) Line 5305  for(;;)
5305    
5306    if (firstline && IS_NEWLINE(start_match)) break;    if (firstline && IS_NEWLINE(start_match)) break;
5307    
5308    /* Advance the match position by one character. */    /* Advance to new matching position */
5309    
5310    start_match++;    start_match = new_start_match;
 #ifdef SUPPORT_UTF8  
   if (utf8)  
     while(start_match < end_subject && (*start_match & 0xc0) == 0x80)  
       start_match++;  
 #endif  
5311    
5312    /* 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
5313    the subject. */    the subject. */
5314    
5315    if (anchored || start_match > end_subject) break;    if (anchored || start_match > end_subject) break;
5316    
5317    /* 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
5318    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
5319    character. */    or ANY or ANYCRLF, advance the match position by one more character. */
5320    
5321    if (start_match[-1] == '\r' &&    if (start_match[-1] == CHAR_CR &&
5322         (md->nltype == NLTYPE_ANY ||        start_match < end_subject &&
5323          md->nltype == NLTYPE_ANYCRLF ||        *start_match == CHAR_NL &&
5324          md->nllen == 2) &&        (re->flags & PCRE_HASCRORLF) == 0 &&
5325         start_match < end_subject &&          (md->nltype == NLTYPE_ANY ||
5326         *start_match == '\n')           md->nltype == NLTYPE_ANYCRLF ||
5327             md->nllen == 2))
5328      start_match++;      start_match++;
5329    
5330    }   /* End of for(;;) "bumpalong" loop */    }   /* End of for(;;) "bumpalong" loop */
# Line 4227  for(;;) Line 5334  for(;;)
5334  /* 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
5335  conditions is true:  conditions is true:
5336    
5337  (1) The pattern is anchored;  (1) The pattern is anchored or the match was failed by (*COMMIT);
5338    
5339  (2) We are past the end of the subject;  (2) We are past the end of the subject;
5340    
# Line 4242  processing, copy those that we can. In t Line 5349  processing, copy those that we can. In t
5349  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
5350  capturing parentheses than vector slots. */  capturing parentheses than vector slots. */
5351    
5352    ENDLOOP:
5353    
5354  if (rc == MATCH_MATCH)  if (rc == MATCH_MATCH)
5355    {    {
5356    if (using_temporary_offsets)    if (using_temporary_offsets)
# Line 4290  if (rc != MATCH_NOMATCH) Line 5399  if (rc != MATCH_NOMATCH)
5399    DPRINTF((">>>> error: returning %d\n", rc));    DPRINTF((">>>> error: returning %d\n", rc));
5400    return rc;    return rc;
5401    }    }
5402  else if (md->partial && md->hitend)  else if (md->partial && start_partial != NULL)
5403    {    {
5404    DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n"));    DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n"));
5405      if (offsetcount > 1)
5406        {
5407        offsets[0] = start_partial - (USPTR)subject;
5408        offsets[1] = end_subject - (USPTR)subject;
5409        }
5410    return PCRE_ERROR_PARTIAL;    return PCRE_ERROR_PARTIAL;
5411    }    }
5412  else  else

Legend:
Removed from v.172  
changed lines
  Added in v.426

  ViewVC Help
Powered by ViewVC 1.1.5