/[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 168 by ph10, Tue May 29 15:18:18 2007 UTC revision 335 by ph10, Sat Apr 12 14:36:14 2008 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) 1997-2008 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 205  variable instead of being passed in the Line 211  variable instead of being passed in the
211  ****************************************************************************  ****************************************************************************
212  ***************************************************************************/  ***************************************************************************/
213    
214    /* Numbers for RMATCH calls. When this list is changed, the code at HEAP_RETURN
215  /* Numbers for RMATCH calls */  below must be updated in sync.  */
216    
217  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM6,  RM7,  RM8,  RM9,  RM10,  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM6,  RM7,  RM8,  RM9,  RM10,
218         RM11,  RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,         RM11,  RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,
219         RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,         RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,
220         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
221         RM41,  RM42, RM43, RM44, RM45, RM46, RM47 };         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
222           RM51,  RM52, RM53, RM54 };
223    
224  /* 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
225  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 292  typedef struct heapframe { Line 298  typedef struct heapframe {
298    
299    const uschar *Xeptr;    const uschar *Xeptr;
300    const uschar *Xecode;    const uschar *Xecode;
301    const uschar *Xmstart;    const uschar *Xmstart;
302    int Xoffset_top;    int Xoffset_top;
303    long int Xims;    long int Xims;
304    eptrblock *Xeptrb;    eptrblock *Xeptrb;
# Line 374  Arguments: Line 380  Arguments:
380     eptr        pointer to current character in subject     eptr        pointer to current character in subject
381     ecode       pointer to current position in compiled code     ecode       pointer to current position in compiled code
382     mstart      pointer to the current match start position (can be modified     mstart      pointer to the current match start position (can be modified
383                   by encountering \K)                   by encountering \K)
384     offset_top  current top pointer     offset_top  current top pointer
385     md          pointer to "static" info for the match     md          pointer to "static" info for the match
386     ims         current /i, /m, and /s options     ims         current /i, /m, and /s options
# Line 384  Arguments: Line 390  Arguments:
390                   match_condassert - this is an assertion condition                   match_condassert - this is an assertion condition
391                   match_cbegroup - this is the start of an unlimited repeat                   match_cbegroup - this is the start of an unlimited repeat
392                     group that can match an empty string                     group that can match an empty string
                  match_tail_recursed - this is a tail_recursed group  
393     rdepth      the recursion depth     rdepth      the recursion depth
394    
395  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 399  Returns: MATCH_MATCH if matched
399  */  */
400    
401  static int  static int
402  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, const uschar *mstart,  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, const uschar *mstart,
403    int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,    int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,
404    int flags, unsigned int rdepth)    int flags, unsigned int rdepth)
405  {  {
# Line 586  original_ims = ims; /* Save for reset Line 591  original_ims = ims; /* Save for reset
591  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
592  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
593  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.
594  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
595  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
596  already used. */  block that is used is on the stack, so a new one may be required for each
597    match(). */
598    
599  if ((flags & match_cbegroup) != 0)  if ((flags & match_cbegroup) != 0)
600    {    {
601    eptrblock *p;    newptrb.epb_saved_eptr = eptr;
602    if ((flags & match_tail_recursed) != 0)    newptrb.epb_prev = eptrb;
603      {    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;  
604    }    }
605    
606  /* Now start processing the opcodes. */  /* Now start processing the opcodes. */
# Line 621  for (;;) Line 620  for (;;)
620    
621    switch(op)    switch(op)
622      {      {
623        case OP_FAIL:
624        RRETURN(MATCH_NOMATCH);
625    
626        case OP_PRUNE:
627        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
628          ims, eptrb, flags, RM51);
629        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
630        RRETURN(MATCH_PRUNE);
631    
632        case OP_COMMIT:
633        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
634          ims, eptrb, flags, RM52);
635        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
636        RRETURN(MATCH_COMMIT);
637    
638        case OP_SKIP:
639        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
640          ims, eptrb, flags, RM53);
641        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
642        md->start_match_ptr = eptr;   /* Pass back current position */
643        RRETURN(MATCH_SKIP);
644    
645        case OP_THEN:
646        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
647          ims, eptrb, flags, RM54);
648        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
649        RRETURN(MATCH_THEN);
650    
651      /* 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
652      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.
653      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 689  for (;;)
689          {          {
690          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
691            ims, eptrb, flags, RM1);            ims, eptrb, flags, RM1);
692          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
693          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
694          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
695          }          }
# Line 677  for (;;) Line 704  for (;;)
704        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
705        }        }
706    
707      /* Insufficient room for saving captured contents. Treat as a non-capturing      /* FALL THROUGH ... Insufficient room for saving captured contents. Treat
708      bracket. */      as a non-capturing bracket. */
709    
710        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
711        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
712    
713      DPRINTF(("insufficient capture room: treat as non-capturing\n"));      DPRINTF(("insufficient capture room: treat as non-capturing\n"));
714    
715        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
716        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
717    
718      /* 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
719      final alternative within the brackets, we would return the result of a      final alternative within the brackets, we would return the result of a
720      recursive call to match() whatever happened. We can reduce stack usage by      recursive call to match() whatever happened. We can reduce stack usage by
721      turning this into a tail recursion. */      turning this into a tail recursion, except in the case when match_cbegroup
722        is set.*/
723    
724      case OP_BRA:      case OP_BRA:
725      case OP_SBRA:      case OP_SBRA:
# Line 693  for (;;) Line 727  for (;;)
727      flags = (op >= OP_SBRA)? match_cbegroup : 0;      flags = (op >= OP_SBRA)? match_cbegroup : 0;
728      for (;;)      for (;;)
729        {        {
730        if (ecode[GET(ecode, 1)] != OP_ALT)        if (ecode[GET(ecode, 1)] != OP_ALT)   /* Final alternative */
731          {          {
732          ecode += _pcre_OP_lengths[*ecode];          if (flags == 0)    /* Not a possibly empty group */
733          flags |= match_tail_recursed;            {
734          DPRINTF(("bracket 0 tail recursion\n"));            ecode += _pcre_OP_lengths[*ecode];
735          goto TAIL_RECURSE;            DPRINTF(("bracket 0 tail recursion\n"));
736              goto TAIL_RECURSE;
737              }
738    
739            /* Possibly empty group; can't use tail recursion. */
740    
741            RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
742              eptrb, flags, RM48);
743            RRETURN(rrc);
744          }          }
745    
746        /* 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 748  for (;;)
748    
749        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
750          eptrb, flags, RM2);          eptrb, flags, RM2);
751        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
752        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
753        }        }
754      /* Control never reaches here. */      /* Control never reaches here. */
# Line 754  for (;;) Line 796  for (;;)
796          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
797          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
798          }          }
799        else if (rrc != MATCH_NOMATCH)        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
800          {          {
801          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
802          }          }
# Line 766  for (;;) Line 808  for (;;)
808        }        }
809    
810      /* 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,
811      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
812      alternative doesn't exist, we can just plough on. */      match_cbegroup is required for an unlimited repeat of a possibly empty
813        group. If the second alternative doesn't exist, we can just plough on. */
814    
815      if (condition || *ecode == OP_ALT)      if (condition || *ecode == OP_ALT)
816        {        {
817        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
818        flags = match_tail_recursed | ((op == OP_SCOND)? match_cbegroup : 0);        if (op == OP_SCOND)        /* Possibly empty group */
819        goto TAIL_RECURSE;          {
820            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, match_cbegroup, RM49);
821            RRETURN(rrc);
822            }
823          else                       /* Group must match something */
824            {
825            flags = 0;
826            goto TAIL_RECURSE;
827            }
828        }        }
829      else      else                         /* Condition false & no 2nd alternative */
830        {        {
831        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
832        }        }
833      break;      break;
834    
835    
836      /* 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
837      restore the offsets appropriately and continue from after the call. */      recursion, we should restore the offsets appropriately and continue from
838        after the call. */
839    
840        case OP_ACCEPT:
841      case OP_END:      case OP_END:
842      if (md->recursive != NULL && md->recursive->group_num == 0)      if (md->recursive != NULL && md->recursive->group_num == 0)
843        {        {
# Line 805  for (;;) Line 858  for (;;)
858      if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);      if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);
859      md->end_match_ptr = eptr;           /* Record where we ended */      md->end_match_ptr = eptr;           /* Record where we ended */
860      md->end_offset_top = offset_top;    /* and how many extracts were taken */      md->end_offset_top = offset_top;    /* and how many extracts were taken */
861      md->start_match_ptr = mstart;  /* and the start (\K can modify) */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
862      RRETURN(MATCH_MATCH);      RRETURN(MATCH_MATCH);
863    
864      /* Change option settings */      /* Change option settings */
# Line 829  for (;;) Line 882  for (;;)
882        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
883          RM4);          RM4);
884        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
885        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
886        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
887        }        }
888      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 856  for (;;) Line 909  for (;;)
909        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
910          RM5);          RM5);
911        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);
912        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
913        ecode += GET(ecode,1);        ecode += GET(ecode,1);
914        }        }
915      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 880  for (;;) Line 933  for (;;)
933          {          {
934          eptr--;          eptr--;
935          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
936          BACKCHAR(eptr)          BACKCHAR(eptr);
937          }          }
938        }        }
939      else      else
# Line 993  for (;;) Line 1046  for (;;)
1046              (pcre_free)(new_recursive.offset_save);              (pcre_free)(new_recursive.offset_save);
1047            RRETURN(MATCH_MATCH);            RRETURN(MATCH_MATCH);
1048            }            }
1049          else if (rrc != MATCH_NOMATCH)          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1050            {            {
1051            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1052            RRETURN(rrc);            RRETURN(rrc);
# Line 1027  for (;;) Line 1080  for (;;)
1080    
1081      do      do
1082        {        {
1083        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);  
1084        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
1085        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1086        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1087        }        }
1088      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 1073  for (;;) Line 1125  for (;;)
1125    
1126      if (*ecode == OP_KETRMIN)      if (*ecode == OP_KETRMIN)
1127        {        {
1128        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);  
1129        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1130        ecode = prev;        ecode = prev;
1131        flags = match_tail_recursed;        flags = 0;
1132        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1133        }        }
1134      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
# Line 1085  for (;;) Line 1136  for (;;)
1136        RMATCH(eptr, prev, offset_top, md, ims, eptrb, match_cbegroup, RM9);        RMATCH(eptr, prev, offset_top, md, ims, eptrb, match_cbegroup, RM9);
1137        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1138        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1139        flags = match_tail_recursed;        flags = 0;
1140        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1141        }        }
1142      /* Control never gets here */      /* Control never gets here */
# Line 1097  for (;;) Line 1148  for (;;)
1148      do ecode += GET(ecode,1); while (*ecode == OP_ALT);      do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1149      break;      break;
1150    
1151      /* BRAZERO and BRAMINZERO occur just before a bracket group, indicating      /* BRAZERO, BRAMINZERO and SKIPZERO occur just before a bracket group,
1152      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
1153      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
1154      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
1155      preceded by BRAZERO or BRAMINZERO. */      optional ones preceded by BRAZERO or BRAMINZERO. */
1156    
1157      case OP_BRAZERO:      case OP_BRAZERO:
1158        {        {
# Line 1123  for (;;) Line 1174  for (;;)
1174        }        }
1175      break;      break;
1176    
1177        case OP_SKIPZERO:
1178          {
1179          next = ecode+1;
1180          do next += GET(next,1); while (*next == OP_ALT);
1181          ecode = next + 1 + LINK_SIZE;
1182          }
1183        break;
1184    
1185      /* End of a group, repeated or non-repeating. */      /* End of a group, repeated or non-repeating. */
1186    
1187      case OP_KET:      case OP_KET:
# Line 1216  for (;;) Line 1275  for (;;)
1275    
1276      /* 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
1277      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
1278      tail recursion to avoid using another stack frame. */      tail recursion to avoid using another stack frame, unless we have an
1279        unlimited repeat of a group that can match an empty string. */
1280    
1281      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;
1282    
1283      if (*ecode == OP_KETRMIN)      if (*ecode == OP_KETRMIN)
1284        {        {
1285        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);  
1286        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1287          if (flags != 0)    /* Could match an empty string */
1288            {
1289            RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM50);
1290            RRETURN(rrc);
1291            }
1292        ecode = prev;        ecode = prev;
       flags |= match_tail_recursed;  
1293        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1294        }        }
1295      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
# Line 1234  for (;;) Line 1297  for (;;)
1297        RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);        RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);
1298        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1299        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1300        flags = match_tail_recursed;        flags = 0;
1301        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1302        }        }
1303      /* Control never gets here */      /* Control never gets here */
# Line 1266  for (;;) Line 1329  for (;;)
1329      if (eptr != md->start_subject + md->start_offset) RRETURN(MATCH_NOMATCH);      if (eptr != md->start_subject + md->start_offset) RRETURN(MATCH_NOMATCH);
1330      ecode++;      ecode++;
1331      break;      break;
1332    
1333      /* Reset the start of match point */      /* Reset the start of match point */
1334    
1335      case OP_SET_SOM:      case OP_SET_SOM:
1336      mstart = eptr;      mstart = eptr;
1337      ecode++;      ecode++;
1338      break;      break;
1339    
1340      /* Assert before internal newline if multiline, or before a terminating      /* Assert before internal newline if multiline, or before a terminating
1341      newline unless endonly is set, else end of subject unless noteol is set. */      newline unless endonly is set, else end of subject unless noteol is set. */
# Line 1471  for (;;) Line 1534  for (;;)
1534        case 0x000d:        case 0x000d:
1535        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
1536        break;        break;
1537    
1538        case 0x000a:        case 0x000a:
1539          break;
1540    
1541        case 0x000b:        case 0x000b:
1542        case 0x000c:        case 0x000c:
1543        case 0x0085:        case 0x0085:
1544        case 0x2028:        case 0x2028:
1545        case 0x2029:        case 0x2029:
1546          if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
1547          break;
1548          }
1549        ecode++;
1550        break;
1551    
1552        case OP_NOT_HSPACE:
1553        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1554        GETCHARINCTEST(c, eptr);
1555        switch(c)
1556          {
1557          default: break;
1558          case 0x09:      /* HT */
1559          case 0x20:      /* SPACE */
1560          case 0xa0:      /* NBSP */
1561          case 0x1680:    /* OGHAM SPACE MARK */
1562          case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
1563          case 0x2000:    /* EN QUAD */
1564          case 0x2001:    /* EM QUAD */
1565          case 0x2002:    /* EN SPACE */
1566          case 0x2003:    /* EM SPACE */
1567          case 0x2004:    /* THREE-PER-EM SPACE */
1568          case 0x2005:    /* FOUR-PER-EM SPACE */
1569          case 0x2006:    /* SIX-PER-EM SPACE */
1570          case 0x2007:    /* FIGURE SPACE */
1571          case 0x2008:    /* PUNCTUATION SPACE */
1572          case 0x2009:    /* THIN SPACE */
1573          case 0x200A:    /* HAIR SPACE */
1574          case 0x202f:    /* NARROW NO-BREAK SPACE */
1575          case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
1576          case 0x3000:    /* IDEOGRAPHIC SPACE */
1577          RRETURN(MATCH_NOMATCH);
1578          }
1579        ecode++;
1580        break;
1581    
1582        case OP_HSPACE:
1583        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1584        GETCHARINCTEST(c, eptr);
1585        switch(c)
1586          {
1587          default: RRETURN(MATCH_NOMATCH);
1588          case 0x09:      /* HT */
1589          case 0x20:      /* SPACE */
1590          case 0xa0:      /* NBSP */
1591          case 0x1680:    /* OGHAM SPACE MARK */
1592          case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
1593          case 0x2000:    /* EN QUAD */
1594          case 0x2001:    /* EM QUAD */
1595          case 0x2002:    /* EN SPACE */
1596          case 0x2003:    /* EM SPACE */
1597          case 0x2004:    /* THREE-PER-EM SPACE */
1598          case 0x2005:    /* FOUR-PER-EM SPACE */
1599          case 0x2006:    /* SIX-PER-EM SPACE */
1600          case 0x2007:    /* FIGURE SPACE */
1601          case 0x2008:    /* PUNCTUATION SPACE */
1602          case 0x2009:    /* THIN SPACE */
1603          case 0x200A:    /* HAIR SPACE */
1604          case 0x202f:    /* NARROW NO-BREAK SPACE */
1605          case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
1606          case 0x3000:    /* IDEOGRAPHIC SPACE */
1607          break;
1608          }
1609        ecode++;
1610        break;
1611    
1612        case OP_NOT_VSPACE:
1613        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1614        GETCHARINCTEST(c, eptr);
1615        switch(c)
1616          {
1617          default: break;
1618          case 0x0a:      /* LF */
1619          case 0x0b:      /* VT */
1620          case 0x0c:      /* FF */
1621          case 0x0d:      /* CR */
1622          case 0x85:      /* NEL */
1623          case 0x2028:    /* LINE SEPARATOR */
1624          case 0x2029:    /* PARAGRAPH SEPARATOR */
1625          RRETURN(MATCH_NOMATCH);
1626          }
1627        ecode++;
1628        break;
1629    
1630        case OP_VSPACE:
1631        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1632        GETCHARINCTEST(c, eptr);
1633        switch(c)
1634          {
1635          default: RRETURN(MATCH_NOMATCH);
1636          case 0x0a:      /* LF */
1637          case 0x0b:      /* VT */
1638          case 0x0c:      /* FF */
1639          case 0x0d:      /* CR */
1640          case 0x85:      /* NEL */
1641          case 0x2028:    /* LINE SEPARATOR */
1642          case 0x2029:    /* PARAGRAPH SEPARATOR */
1643        break;        break;
1644        }        }
1645      ecode++;      ecode++;
# Line 1937  for (;;) Line 2100  for (;;)
2100            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);
2101            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2102            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
2103            BACKCHAR(eptr)            if (utf8) BACKCHAR(eptr);
2104            }            }
2105          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2106          }          }
# Line 2690  for (;;) Line 2853  for (;;)
2853            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
2854              {              {
2855              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2856              GETCHARINC(c, eptr);              GETCHARINCTEST(c, eptr);
2857              }              }
2858            break;            break;
2859    
# Line 2698  for (;;) Line 2861  for (;;)
2861            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
2862              {              {
2863              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2864              GETCHARINC(c, eptr);              GETCHARINCTEST(c, eptr);
2865              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
2866              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
2867                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
# Line 2711  for (;;) Line 2874  for (;;)
2874            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
2875              {              {
2876              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2877              GETCHARINC(c, eptr);              GETCHARINCTEST(c, eptr);
2878              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
2879              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
2880                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
# Line 2722  for (;;) Line 2885  for (;;)
2885            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
2886              {              {
2887              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2888              GETCHARINC(c, eptr);              GETCHARINCTEST(c, eptr);
2889              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
2890              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
2891                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
# Line 2733  for (;;) Line 2896  for (;;)
2896            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
2897              {              {
2898              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2899              GETCHARINC(c, eptr);              GETCHARINCTEST(c, eptr);
2900              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
2901              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
2902                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
# Line 2803  for (;;) Line 2966  for (;;)
2966              case 0x000d:              case 0x000d:
2967              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
2968              break;              break;
2969    
2970              case 0x000a:              case 0x000a:
2971                break;
2972    
2973              case 0x000b:              case 0x000b:
2974              case 0x000c:              case 0x000c:
2975              case 0x0085:              case 0x0085:
2976              case 0x2028:              case 0x2028:
2977              case 0x2029:              case 0x2029:
2978                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
2979                break;
2980                }
2981              }
2982            break;
2983    
2984            case OP_NOT_HSPACE:
2985            for (i = 1; i <= min; i++)
2986              {
2987              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2988              GETCHARINC(c, eptr);
2989              switch(c)
2990                {
2991                default: break;
2992                case 0x09:      /* HT */
2993                case 0x20:      /* SPACE */
2994                case 0xa0:      /* NBSP */
2995                case 0x1680:    /* OGHAM SPACE MARK */
2996                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
2997                case 0x2000:    /* EN QUAD */
2998                case 0x2001:    /* EM QUAD */
2999                case 0x2002:    /* EN SPACE */
3000                case 0x2003:    /* EM SPACE */
3001                case 0x2004:    /* THREE-PER-EM SPACE */
3002                case 0x2005:    /* FOUR-PER-EM SPACE */
3003                case 0x2006:    /* SIX-PER-EM SPACE */
3004                case 0x2007:    /* FIGURE SPACE */
3005                case 0x2008:    /* PUNCTUATION SPACE */
3006                case 0x2009:    /* THIN SPACE */
3007                case 0x200A:    /* HAIR SPACE */
3008                case 0x202f:    /* NARROW NO-BREAK SPACE */
3009                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3010                case 0x3000:    /* IDEOGRAPHIC SPACE */
3011                RRETURN(MATCH_NOMATCH);
3012                }
3013              }
3014            break;
3015    
3016            case OP_HSPACE:
3017            for (i = 1; i <= min; i++)
3018              {
3019              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3020              GETCHARINC(c, eptr);
3021              switch(c)
3022                {
3023                default: RRETURN(MATCH_NOMATCH);
3024                case 0x09:      /* HT */
3025                case 0x20:      /* SPACE */
3026                case 0xa0:      /* NBSP */
3027                case 0x1680:    /* OGHAM SPACE MARK */
3028                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3029                case 0x2000:    /* EN QUAD */
3030                case 0x2001:    /* EM QUAD */
3031                case 0x2002:    /* EN SPACE */
3032                case 0x2003:    /* EM SPACE */
3033                case 0x2004:    /* THREE-PER-EM SPACE */
3034                case 0x2005:    /* FOUR-PER-EM SPACE */
3035                case 0x2006:    /* SIX-PER-EM SPACE */
3036                case 0x2007:    /* FIGURE SPACE */
3037                case 0x2008:    /* PUNCTUATION SPACE */
3038                case 0x2009:    /* THIN SPACE */
3039                case 0x200A:    /* HAIR SPACE */
3040                case 0x202f:    /* NARROW NO-BREAK SPACE */
3041                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3042                case 0x3000:    /* IDEOGRAPHIC SPACE */
3043                break;
3044                }
3045              }
3046            break;
3047    
3048            case OP_NOT_VSPACE:
3049            for (i = 1; i <= min; i++)
3050              {
3051              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3052              GETCHARINC(c, eptr);
3053              switch(c)
3054                {
3055                default: break;
3056                case 0x0a:      /* LF */
3057                case 0x0b:      /* VT */
3058                case 0x0c:      /* FF */
3059                case 0x0d:      /* CR */
3060                case 0x85:      /* NEL */
3061                case 0x2028:    /* LINE SEPARATOR */
3062                case 0x2029:    /* PARAGRAPH SEPARATOR */
3063                RRETURN(MATCH_NOMATCH);
3064                }
3065              }
3066            break;
3067    
3068            case OP_VSPACE:
3069            for (i = 1; i <= min; i++)
3070              {
3071              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3072              GETCHARINC(c, eptr);
3073              switch(c)
3074                {
3075                default: RRETURN(MATCH_NOMATCH);
3076                case 0x0a:      /* LF */
3077                case 0x0b:      /* VT */
3078                case 0x0c:      /* FF */
3079                case 0x0d:      /* CR */
3080                case 0x85:      /* NEL */
3081                case 0x2028:    /* LINE SEPARATOR */
3082                case 0x2029:    /* PARAGRAPH SEPARATOR */
3083              break;              break;
3084              }              }
3085            }            }
# Line 2838  for (;;) Line 3109  for (;;)
3109          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3110            {            {
3111            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3112               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_space) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0))
3113              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3114            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3115            }            }
3116          break;          break;
3117    
# Line 2858  for (;;) Line 3129  for (;;)
3129          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3130            {            {
3131            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3132               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_word) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0))
3133              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3134            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3135            }            }
3136          break;          break;
3137    
# Line 2917  for (;;) Line 3188  for (;;)
3188              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3189              break;              break;
3190              case 0x000a:              case 0x000a:
3191                break;
3192    
3193              case 0x000b:              case 0x000b:
3194              case 0x000c:              case 0x000c:
3195              case 0x0085:              case 0x0085:
3196                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3197                break;
3198                }
3199              }
3200            break;
3201    
3202            case OP_NOT_HSPACE:
3203            for (i = 1; i <= min; i++)
3204              {
3205              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3206              switch(*eptr++)
3207                {
3208                default: break;
3209                case 0x09:      /* HT */
3210                case 0x20:      /* SPACE */
3211                case 0xa0:      /* NBSP */
3212                RRETURN(MATCH_NOMATCH);
3213                }
3214              }
3215            break;
3216    
3217            case OP_HSPACE:
3218            for (i = 1; i <= min; i++)
3219              {
3220              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3221              switch(*eptr++)
3222                {
3223                default: RRETURN(MATCH_NOMATCH);
3224                case 0x09:      /* HT */
3225                case 0x20:      /* SPACE */
3226                case 0xa0:      /* NBSP */
3227                break;
3228                }
3229              }
3230            break;
3231    
3232            case OP_NOT_VSPACE:
3233            for (i = 1; i <= min; i++)
3234              {
3235              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3236              switch(*eptr++)
3237                {
3238                default: break;
3239                case 0x0a:      /* LF */
3240                case 0x0b:      /* VT */
3241                case 0x0c:      /* FF */
3242                case 0x0d:      /* CR */
3243                case 0x85:      /* NEL */
3244                RRETURN(MATCH_NOMATCH);
3245                }
3246              }
3247            break;
3248    
3249            case OP_VSPACE:
3250            for (i = 1; i <= min; i++)
3251              {
3252              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3253              switch(*eptr++)
3254                {
3255                default: RRETURN(MATCH_NOMATCH);
3256                case 0x0a:      /* LF */
3257                case 0x0b:      /* VT */
3258                case 0x0c:      /* FF */
3259                case 0x0d:      /* CR */
3260                case 0x85:      /* NEL */
3261              break;              break;
3262              }              }
3263            }            }
# Line 3107  for (;;) Line 3445  for (;;)
3445                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3446                break;                break;
3447                case 0x000a:                case 0x000a:
3448                  break;
3449    
3450                case 0x000b:                case 0x000b:
3451                case 0x000c:                case 0x000c:
3452                case 0x0085:                case 0x0085:
3453                case 0x2028:                case 0x2028:
3454                case 0x2029:                case 0x2029:
3455                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3456                  break;
3457                  }
3458                break;
3459    
3460                case OP_NOT_HSPACE:
3461                switch(c)
3462                  {
3463                  default: break;
3464                  case 0x09:      /* HT */
3465                  case 0x20:      /* SPACE */
3466                  case 0xa0:      /* NBSP */
3467                  case 0x1680:    /* OGHAM SPACE MARK */
3468                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3469                  case 0x2000:    /* EN QUAD */
3470                  case 0x2001:    /* EM QUAD */
3471                  case 0x2002:    /* EN SPACE */
3472                  case 0x2003:    /* EM SPACE */
3473                  case 0x2004:    /* THREE-PER-EM SPACE */
3474                  case 0x2005:    /* FOUR-PER-EM SPACE */
3475                  case 0x2006:    /* SIX-PER-EM SPACE */
3476                  case 0x2007:    /* FIGURE SPACE */
3477                  case 0x2008:    /* PUNCTUATION SPACE */
3478                  case 0x2009:    /* THIN SPACE */
3479                  case 0x200A:    /* HAIR SPACE */
3480                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3481                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3482                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3483                  RRETURN(MATCH_NOMATCH);
3484                  }
3485                break;
3486    
3487                case OP_HSPACE:
3488                switch(c)
3489                  {
3490                  default: RRETURN(MATCH_NOMATCH);
3491                  case 0x09:      /* HT */
3492                  case 0x20:      /* SPACE */
3493                  case 0xa0:      /* NBSP */
3494                  case 0x1680:    /* OGHAM SPACE MARK */
3495                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3496                  case 0x2000:    /* EN QUAD */
3497                  case 0x2001:    /* EM QUAD */
3498                  case 0x2002:    /* EN SPACE */
3499                  case 0x2003:    /* EM SPACE */
3500                  case 0x2004:    /* THREE-PER-EM SPACE */
3501                  case 0x2005:    /* FOUR-PER-EM SPACE */
3502                  case 0x2006:    /* SIX-PER-EM SPACE */
3503                  case 0x2007:    /* FIGURE SPACE */
3504                  case 0x2008:    /* PUNCTUATION SPACE */
3505                  case 0x2009:    /* THIN SPACE */
3506                  case 0x200A:    /* HAIR SPACE */
3507                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3508                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3509                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3510                  break;
3511                  }
3512                break;
3513    
3514                case OP_NOT_VSPACE:
3515                switch(c)
3516                  {
3517                  default: break;
3518                  case 0x0a:      /* LF */
3519                  case 0x0b:      /* VT */
3520                  case 0x0c:      /* FF */
3521                  case 0x0d:      /* CR */
3522                  case 0x85:      /* NEL */
3523                  case 0x2028:    /* LINE SEPARATOR */
3524                  case 0x2029:    /* PARAGRAPH SEPARATOR */
3525                  RRETURN(MATCH_NOMATCH);
3526                  }
3527                break;
3528    
3529                case OP_VSPACE:
3530                switch(c)
3531                  {
3532                  default: RRETURN(MATCH_NOMATCH);
3533                  case 0x0a:      /* LF */
3534                  case 0x0b:      /* VT */
3535                  case 0x0c:      /* FF */
3536                  case 0x0d:      /* CR */
3537                  case 0x85:      /* NEL */
3538                  case 0x2028:    /* LINE SEPARATOR */
3539                  case 0x2029:    /* PARAGRAPH SEPARATOR */
3540                break;                break;
3541                }                }
3542              break;              break;
# Line 3179  for (;;) Line 3604  for (;;)
3604                case 0x000d:                case 0x000d:
3605                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3606                break;                break;
3607    
3608                case 0x000a:                case 0x000a:
3609                  break;
3610    
3611                case 0x000b:                case 0x000b:
3612                case 0x000c:                case 0x000c:
3613                case 0x0085:                case 0x0085:
3614                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3615                  break;
3616                  }
3617                break;
3618    
3619                case OP_NOT_HSPACE:
3620                switch(c)
3621                  {
3622                  default: break;
3623                  case 0x09:      /* HT */
3624                  case 0x20:      /* SPACE */
3625                  case 0xa0:      /* NBSP */
3626                  RRETURN(MATCH_NOMATCH);
3627                  }
3628                break;
3629    
3630                case OP_HSPACE:
3631                switch(c)
3632                  {
3633                  default: RRETURN(MATCH_NOMATCH);
3634                  case 0x09:      /* HT */
3635                  case 0x20:      /* SPACE */
3636                  case 0xa0:      /* NBSP */
3637                  break;
3638                  }
3639                break;
3640    
3641                case OP_NOT_VSPACE:
3642                switch(c)
3643                  {
3644                  default: break;
3645                  case 0x0a:      /* LF */
3646                  case 0x0b:      /* VT */
3647                  case 0x0c:      /* FF */
3648                  case 0x0d:      /* CR */
3649                  case 0x85:      /* NEL */
3650                  RRETURN(MATCH_NOMATCH);
3651                  }
3652                break;
3653    
3654                case OP_VSPACE:
3655                switch(c)
3656                  {
3657                  default: RRETURN(MATCH_NOMATCH);
3658                  case 0x0a:      /* LF */
3659                  case 0x0b:      /* VT */
3660                  case 0x0c:      /* FF */
3661                  case 0x0d:      /* CR */
3662                  case 0x85:      /* NEL */
3663                break;                break;
3664                }                }
3665              break;              break;
# Line 3306  for (;;) Line 3783  for (;;)
3783            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);
3784            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3785            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
3786            BACKCHAR(eptr);            if (utf8) BACKCHAR(eptr);
3787            }            }
3788          }          }
3789    
# Line 3345  for (;;) Line 3822  for (;;)
3822            for (;;)                        /* Move back over one extended */            for (;;)                        /* Move back over one extended */
3823              {              {
3824              int len = 1;              int len = 1;
             BACKCHAR(eptr);  
3825              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr; else
3826                {                {
3827                  BACKCHAR(eptr);
3828                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
3829                }                }
3830              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
# Line 3368  for (;;) Line 3845  for (;;)
3845          switch(ctype)          switch(ctype)
3846            {            {
3847            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. */  
   
3848            if (max < INT_MAX)            if (max < INT_MAX)
3849              {              {
3850              if ((ims & PCRE_DOTALL) == 0)              if ((ims & PCRE_DOTALL) == 0)
# Line 3405  for (;;) Line 3877  for (;;)
3877                  {                  {
3878                  if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;                  if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
3879                  eptr++;                  eptr++;
3880                    while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3881                  }                  }
               break;  
3882                }                }
3883              else              else
3884                {                {
3885                c = max - min;                eptr = md->end_subject;
               if (c > (unsigned int)(md->end_subject - eptr))  
                 c = md->end_subject - eptr;  
               eptr += c;  
3886                }                }
3887              }              }
3888            break;            break;
# Line 3440  for (;;) Line 3909  for (;;)
3909                }                }
3910              else              else
3911                {                {
3912                if (c != 0x000a && c != 0x000b && c != 0x000c &&                if (c != 0x000a &&
3913                    c != 0x0085 && c != 0x2028 && c != 0x2029)                    (md->bsr_anycrlf ||
3914                       (c != 0x000b && c != 0x000c &&
3915                        c != 0x0085 && c != 0x2028 && c != 0x2029)))
3916                  break;                  break;
3917                eptr += len;                eptr += len;
3918                }                }
3919              }              }
3920            break;            break;
3921    
3922              case OP_NOT_HSPACE:
3923              case OP_HSPACE:
3924              for (i = min; i < max; i++)
3925                {
3926                BOOL gotspace;
3927                int len = 1;
3928                if (eptr >= md->end_subject) break;
3929                GETCHARLEN(c, eptr, len);
3930                switch(c)
3931                  {
3932                  default: gotspace = FALSE; break;
3933                  case 0x09:      /* HT */
3934                  case 0x20:      /* SPACE */
3935                  case 0xa0:      /* NBSP */
3936                  case 0x1680:    /* OGHAM SPACE MARK */
3937                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3938                  case 0x2000:    /* EN QUAD */
3939                  case 0x2001:    /* EM QUAD */
3940                  case 0x2002:    /* EN SPACE */
3941                  case 0x2003:    /* EM SPACE */
3942                  case 0x2004:    /* THREE-PER-EM SPACE */
3943                  case 0x2005:    /* FOUR-PER-EM SPACE */
3944                  case 0x2006:    /* SIX-PER-EM SPACE */
3945                  case 0x2007:    /* FIGURE SPACE */
3946                  case 0x2008:    /* PUNCTUATION SPACE */
3947                  case 0x2009:    /* THIN SPACE */
3948                  case 0x200A:    /* HAIR SPACE */
3949                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3950                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3951                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3952                  gotspace = TRUE;
3953                  break;
3954                  }
3955                if (gotspace == (ctype == OP_NOT_HSPACE)) break;
3956                eptr += len;
3957                }
3958              break;
3959    
3960              case OP_NOT_VSPACE:
3961              case OP_VSPACE:
3962              for (i = min; i < max; i++)
3963                {
3964                BOOL gotspace;
3965                int len = 1;
3966                if (eptr >= md->end_subject) break;
3967                GETCHARLEN(c, eptr, len);
3968                switch(c)
3969                  {
3970                  default: gotspace = FALSE; break;
3971                  case 0x0a:      /* LF */
3972                  case 0x0b:      /* VT */
3973                  case 0x0c:      /* FF */
3974                  case 0x0d:      /* CR */
3975                  case 0x85:      /* NEL */
3976                  case 0x2028:    /* LINE SEPARATOR */
3977                  case 0x2029:    /* PARAGRAPH SEPARATOR */
3978                  gotspace = TRUE;
3979                  break;
3980                  }
3981                if (gotspace == (ctype == OP_NOT_VSPACE)) break;
3982                eptr += len;
3983                }
3984              break;
3985    
3986            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
3987            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3988              {              {
# Line 3530  for (;;) Line 4065  for (;;)
4065            }            }
4066          }          }
4067        else        else
4068  #endif  #endif  /* SUPPORT_UTF8 */
4069    
4070        /* Not UTF-8 mode */        /* Not UTF-8 mode */
4071          {          {
# Line 3567  for (;;) Line 4102  for (;;)
4102                }                }
4103              else              else
4104                {                {
4105                if (c != 0x000a && c != 0x000b && c != 0x000c && c != 0x0085)                if (c != 0x000a &&
4106                      (md->bsr_anycrlf ||
4107                        (c != 0x000b && c != 0x000c && c != 0x0085)))
4108                  break;                  break;
4109                eptr++;                eptr++;
4110                }                }
4111              }              }
4112            break;            break;
4113    
4114              case OP_NOT_HSPACE:
4115              for (i = min; i < max; i++)
4116                {
4117                if (eptr >= md->end_subject) break;
4118                c = *eptr;
4119                if (c == 0x09 || c == 0x20 || c == 0xa0) break;
4120                eptr++;
4121                }
4122              break;
4123    
4124              case OP_HSPACE:
4125              for (i = min; i < max; i++)
4126                {
4127                if (eptr >= md->end_subject) break;
4128                c = *eptr;
4129                if (c != 0x09 && c != 0x20 && c != 0xa0) break;
4130                eptr++;
4131                }
4132              break;
4133    
4134              case OP_NOT_VSPACE:
4135              for (i = min; i < max; i++)
4136                {
4137                if (eptr >= md->end_subject) break;
4138                c = *eptr;
4139                if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)
4140                  break;
4141                eptr++;
4142                }
4143              break;
4144    
4145              case OP_VSPACE:
4146              for (i = min; i < max; i++)
4147                {
4148                if (eptr >= md->end_subject) break;
4149                c = *eptr;
4150                if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85)
4151                  break;
4152                eptr++;
4153                }
4154              break;
4155    
4156            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
4157            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4158              {              {
# Line 3675  HEAP_RETURN: Line 4254  HEAP_RETURN:
4254  switch (frame->Xwhere)  switch (frame->Xwhere)
4255    {    {
4256    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)
4257    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)
4258    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)
4259    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)
4260    LBL(33) LBL(34) LBL(35) LBL(36) LBL(37) LBL(38) LBL(39) LBL(40)    LBL(53) LBL(54)
4261    LBL(41) LBL(42) LBL(43) LBL(44) LBL(45) LBL(46) LBL(47)  #ifdef SUPPORT_UTF8
4262      LBL(16) LBL(18) LBL(20) LBL(21) LBL(22) LBL(23) LBL(28) LBL(30)
4263      LBL(32) LBL(34) LBL(42) LBL(46)
4264    #ifdef SUPPORT_UCP
4265      LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
4266    #endif  /* SUPPORT_UCP */
4267    #endif  /* SUPPORT_UTF8 */
4268    default:    default:
4269    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));
4270    return PCRE_ERROR_INTERNAL;    return PCRE_ERROR_INTERNAL;
# Line 3796  const uschar *start_bits = NULL; Line 4381  const uschar *start_bits = NULL;
4381  USPTR start_match = (USPTR)subject + start_offset;  USPTR start_match = (USPTR)subject + start_offset;
4382  USPTR end_subject;  USPTR end_subject;
4383  USPTR req_byte_ptr = start_match - 1;  USPTR req_byte_ptr = start_match - 1;
 eptrblock eptrchain[EPTR_WORK_SIZE];  
4384    
4385  pcre_study_data internal_study;  pcre_study_data internal_study;
4386  const pcre_study_data *study;  const pcre_study_data *study;
# Line 3859  if (re->magic_number != MAGIC_NUMBER) Line 4443  if (re->magic_number != MAGIC_NUMBER)
4443  /* Set up other data */  /* Set up other data */
4444    
4445  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
4446  startline = (re->options & PCRE_STARTLINE) != 0;  startline = (re->flags & PCRE_STARTLINE) != 0;
4447  firstline = (re->options & PCRE_FIRSTLINE) != 0;  firstline = (re->options & PCRE_FIRSTLINE) != 0;
4448    
4449  /* 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 3882  md->partial = (options & PCRE_PARTIAL) ! Line 4466  md->partial = (options & PCRE_PARTIAL) !
4466  md->hitend = FALSE;  md->hitend = FALSE;
4467    
4468  md->recursive = NULL;                   /* No recursion at top level */  md->recursive = NULL;                   /* No recursion at top level */
 md->eptrchain = eptrchain;              /* Make workspace generally available */  
4469    
4470  md->lcc = tables + lcc_offset;  md->lcc = tables + lcc_offset;
4471  md->ctypes = tables + ctypes_offset;  md->ctypes = tables + ctypes_offset;
4472    
4473    /* Handle different \R options. */
4474    
4475    switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
4476      {
4477      case 0:
4478      if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
4479        md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0;
4480      else
4481    #ifdef BSR_ANYCRLF
4482      md->bsr_anycrlf = TRUE;
4483    #else
4484      md->bsr_anycrlf = FALSE;
4485    #endif
4486      break;
4487    
4488      case PCRE_BSR_ANYCRLF:
4489      md->bsr_anycrlf = TRUE;
4490      break;
4491    
4492      case PCRE_BSR_UNICODE:
4493      md->bsr_anycrlf = FALSE;
4494      break;
4495    
4496      default: return PCRE_ERROR_BADNEWLINE;
4497      }
4498    
4499  /* Handle different types of newline. The three bits give eight cases. If  /* Handle different types of newline. The three bits give eight cases. If
4500  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. */
4501    
4502  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
4503         PCRE_NEWLINE_BITS)          (pcre_uint32)options) & PCRE_NEWLINE_BITS)
4504    {    {
4505    case 0: newline = NEWLINE; break;   /* Compile-time default */    case 0: newline = NEWLINE; break;   /* Compile-time default */
4506    case PCRE_NEWLINE_CR: newline = '\r'; break;    case PCRE_NEWLINE_CR: newline = '\r'; break;
# Line 3930  else Line 4539  else
4539  /* Partial matching is supported only for a restricted set of regexes at the  /* Partial matching is supported only for a restricted set of regexes at the
4540  moment. */  moment. */
4541    
4542  if (md->partial && (re->options & PCRE_NOPARTIAL) != 0)  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
4543    return PCRE_ERROR_BADPARTIAL;    return PCRE_ERROR_BADPARTIAL;
4544    
4545  /* 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 4007  studied, there may be a bitmap of possib Line 4616  studied, there may be a bitmap of possib
4616    
4617  if (!anchored)  if (!anchored)
4618    {    {
4619    if ((re->options & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
4620      {      {
4621      first_byte = re->first_byte & 255;      first_byte = re->first_byte & 255;
4622      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 4631  if (!anchored)
4631  /* For anchored or unanchored matches, there may be a "last known required  /* For anchored or unanchored matches, there may be a "last known required
4632  character" set. */  character" set. */
4633    
4634  if ((re->options & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
4635    {    {
4636    req_byte = re->req_byte & 255;    req_byte = re->req_byte & 255;
4637    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 4647  the loop runs just once. */
4647  for(;;)  for(;;)
4648    {    {
4649    USPTR save_end_subject = end_subject;    USPTR save_end_subject = end_subject;
4650      USPTR new_start_match;
4651    
4652    /* Reset the maximum number of extractions we might see. */    /* Reset the maximum number of extractions we might see. */
4653    
# Line 4068  for(;;) Line 4678  for(;;)
4678      if (first_byte_caseless)      if (first_byte_caseless)
4679        while (start_match < end_subject &&        while (start_match < end_subject &&
4680               md->lcc[*start_match] != first_byte)               md->lcc[*start_match] != first_byte)
4681          start_match++;          { NEXTCHAR(start_match); }
4682      else      else
4683        while (start_match < end_subject && *start_match != first_byte)        while (start_match < end_subject && *start_match != first_byte)
4684          start_match++;          { NEXTCHAR(start_match); }
4685      }      }
4686    
4687    /* Or to just after a linebreak for a multiline match if possible */    /* Or to just after a linebreak for a multiline match if possible */
# Line 4081  for(;;) Line 4691  for(;;)
4691      if (start_match > md->start_subject + start_offset)      if (start_match > md->start_subject + start_offset)
4692        {        {
4693        while (start_match <= end_subject && !WAS_NEWLINE(start_match))        while (start_match <= end_subject && !WAS_NEWLINE(start_match))
4694          start_match++;          { NEXTCHAR(start_match); }
4695    
4696        /* 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,
4697        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.
# Line 4102  for(;;) Line 4712  for(;;)
4712      while (start_match < end_subject)      while (start_match < end_subject)
4713        {        {
4714        register unsigned int c = *start_match;        register unsigned int c = *start_match;
4715        if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; else break;        if ((start_bits[c/8] & (1 << (c&7))) == 0)
4716            { NEXTCHAR(start_match); }
4717          else break;
4718        }        }
4719      }      }
4720    
# Line 4178  for(;;) Line 4790  for(;;)
4790    
4791    /* OK, we can now run the match. */    /* OK, we can now run the match. */
4792    
4793    md->start_match_ptr = start_match;      /* Insurance */    md->start_match_ptr = start_match;
4794    md->match_call_count = 0;    md->match_call_count = 0;
4795    md->eptrn = 0;                          /* Next free eptrchain slot */    rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0);
4796    rc = match(start_match, md->start_code, start_match, 2, md,  
4797      ims, NULL, 0, 0);    switch(rc)
4798        {
4799        /* NOMATCH and PRUNE advance by one character. THEN at this level acts
4800        exactly like PRUNE. */
4801    
4802    /* Any return other than MATCH_NOMATCH breaks the loop. */      case MATCH_NOMATCH:
4803        case MATCH_PRUNE:
4804        case MATCH_THEN:
4805        new_start_match = start_match + 1;
4806    #ifdef SUPPORT_UTF8
4807        if (utf8)
4808          while(new_start_match < end_subject && (*new_start_match & 0xc0) == 0x80)
4809            new_start_match++;
4810    #endif
4811        break;
4812    
4813        /* SKIP passes back the next starting point explicitly. */
4814    
4815        case MATCH_SKIP:
4816        new_start_match = md->start_match_ptr;
4817        break;
4818    
4819    if (rc != MATCH_NOMATCH) break;      /* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */
4820    
4821        case MATCH_COMMIT:
4822        rc = MATCH_NOMATCH;
4823        goto ENDLOOP;
4824    
4825        /* Any other return is some kind of error. */
4826    
4827        default:
4828        goto ENDLOOP;
4829        }
4830    
4831      /* Control reaches here for the various types of "no match at this point"
4832      result. Reset the code to MATCH_NOMATCH for subsequent checking. */
4833    
4834      rc = MATCH_NOMATCH;
4835    
4836    /* 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
4837    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 4839  for(;;)
4839    
4840    if (firstline && IS_NEWLINE(start_match)) break;    if (firstline && IS_NEWLINE(start_match)) break;
4841    
4842    /* Advance the match position by one character. */    /* Advance to new matching position */
4843    
4844    start_match++;    start_match = new_start_match;
 #ifdef SUPPORT_UTF8  
   if (utf8)  
     while(start_match < end_subject && (*start_match & 0xc0) == 0x80)  
       start_match++;  
 #endif  
4845    
4846    /* 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
4847    the subject. */    the subject. */
4848    
4849    if (anchored || start_match > end_subject) break;    if (anchored || start_match > end_subject) break;
4850    
4851    /* 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
4852    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
4853    character. */    or ANY or ANYCRLF, advance the match position by one more character. */
4854    
4855    if (start_match[-1] == '\r' &&    if (start_match[-1] == '\r' &&
4856         (md->nltype == NLTYPE_ANY ||        start_match < end_subject &&
4857          md->nltype == NLTYPE_ANYCRLF ||        *start_match == '\n' &&
4858          md->nllen == 2) &&        (re->flags & PCRE_HASCRORLF) == 0 &&
4859         start_match < end_subject &&          (md->nltype == NLTYPE_ANY ||
4860         *start_match == '\n')           md->nltype == NLTYPE_ANYCRLF ||
4861             md->nllen == 2))
4862      start_match++;      start_match++;
4863    
4864    }   /* End of for(;;) "bumpalong" loop */    }   /* End of for(;;) "bumpalong" loop */
# Line 4227  for(;;) Line 4868  for(;;)
4868  /* 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
4869  conditions is true:  conditions is true:
4870    
4871  (1) The pattern is anchored;  (1) The pattern is anchored or the match was failed by (*COMMIT);
4872    
4873  (2) We are past the end of the subject;  (2) We are past the end of the subject;
4874    
# Line 4242  processing, copy those that we can. In t Line 4883  processing, copy those that we can. In t
4883  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
4884  capturing parentheses than vector slots. */  capturing parentheses than vector slots. */
4885    
4886    ENDLOOP:
4887    
4888  if (rc == MATCH_MATCH)  if (rc == MATCH_MATCH)
4889    {    {
4890    if (using_temporary_offsets)    if (using_temporary_offsets)
# Line 4263  if (rc == MATCH_MATCH) Line 4906  if (rc == MATCH_MATCH)
4906    rc = md->offset_overflow? 0 : md->end_offset_top/2;    rc = md->offset_overflow? 0 : md->end_offset_top/2;
4907    
4908    /* If there is space, set up the whole thing as substring 0. The value of    /* If there is space, set up the whole thing as substring 0. The value of
4909    md->start_match_ptr might be modified if \K was encountered on the success    md->start_match_ptr might be modified if \K was encountered on the success
4910    matching path. */    matching path. */
4911    
4912    if (offsetcount < 2) rc = 0; else    if (offsetcount < 2) rc = 0; else

Legend:
Removed from v.168  
changed lines
  Added in v.335

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12