/[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 294 by ph10, Sat Dec 29 16:55:59 2007 UTC
# 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 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 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 1216  for (;;) Line 1267  for (;;)
1267    
1268      /* 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
1269      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
1270      tail recursion to avoid using another stack frame. */      tail recursion to avoid using another stack frame, unless we have an
1271        unlimited repeat of a group that can match an empty string. */
1272    
1273      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;
1274    
1275      if (*ecode == OP_KETRMIN)      if (*ecode == OP_KETRMIN)
1276        {        {
1277        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);  
1278        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1279          if (flags != 0)    /* Could match an empty string */
1280            {
1281            RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM50);
1282            RRETURN(rrc);
1283            }
1284        ecode = prev;        ecode = prev;
       flags |= match_tail_recursed;  
1285        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1286        }        }
1287      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
# Line 1234  for (;;) Line 1289  for (;;)
1289        RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);        RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);
1290        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1291        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1292        flags = match_tail_recursed;        flags = 0;
1293        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1294        }        }
1295      /* Control never gets here */      /* Control never gets here */
# Line 1471  for (;;) Line 1526  for (;;)
1526        case 0x000d:        case 0x000d:
1527        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
1528        break;        break;
1529    
1530        case 0x000a:        case 0x000a:
1531          break;
1532    
1533        case 0x000b:        case 0x000b:
1534        case 0x000c:        case 0x000c:
1535        case 0x0085:        case 0x0085:
1536        case 0x2028:        case 0x2028:
1537        case 0x2029:        case 0x2029:
1538          if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
1539          break;
1540          }
1541        ecode++;
1542        break;
1543    
1544        case OP_NOT_HSPACE:
1545        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1546        GETCHARINCTEST(c, eptr);
1547        switch(c)
1548          {
1549          default: break;
1550          case 0x09:      /* HT */
1551          case 0x20:      /* SPACE */
1552          case 0xa0:      /* NBSP */
1553          case 0x1680:    /* OGHAM SPACE MARK */
1554          case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
1555          case 0x2000:    /* EN QUAD */
1556          case 0x2001:    /* EM QUAD */
1557          case 0x2002:    /* EN SPACE */
1558          case 0x2003:    /* EM SPACE */
1559          case 0x2004:    /* THREE-PER-EM SPACE */
1560          case 0x2005:    /* FOUR-PER-EM SPACE */
1561          case 0x2006:    /* SIX-PER-EM SPACE */
1562          case 0x2007:    /* FIGURE SPACE */
1563          case 0x2008:    /* PUNCTUATION SPACE */
1564          case 0x2009:    /* THIN SPACE */
1565          case 0x200A:    /* HAIR SPACE */
1566          case 0x202f:    /* NARROW NO-BREAK SPACE */
1567          case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
1568          case 0x3000:    /* IDEOGRAPHIC SPACE */
1569          RRETURN(MATCH_NOMATCH);
1570          }
1571        ecode++;
1572        break;
1573    
1574        case OP_HSPACE:
1575        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1576        GETCHARINCTEST(c, eptr);
1577        switch(c)
1578          {
1579          default: RRETURN(MATCH_NOMATCH);
1580          case 0x09:      /* HT */
1581          case 0x20:      /* SPACE */
1582          case 0xa0:      /* NBSP */
1583          case 0x1680:    /* OGHAM SPACE MARK */
1584          case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
1585          case 0x2000:    /* EN QUAD */
1586          case 0x2001:    /* EM QUAD */
1587          case 0x2002:    /* EN SPACE */
1588          case 0x2003:    /* EM SPACE */
1589          case 0x2004:    /* THREE-PER-EM SPACE */
1590          case 0x2005:    /* FOUR-PER-EM SPACE */
1591          case 0x2006:    /* SIX-PER-EM SPACE */
1592          case 0x2007:    /* FIGURE SPACE */
1593          case 0x2008:    /* PUNCTUATION SPACE */
1594          case 0x2009:    /* THIN SPACE */
1595          case 0x200A:    /* HAIR SPACE */
1596          case 0x202f:    /* NARROW NO-BREAK SPACE */
1597          case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
1598          case 0x3000:    /* IDEOGRAPHIC SPACE */
1599          break;
1600          }
1601        ecode++;
1602        break;
1603    
1604        case OP_NOT_VSPACE:
1605        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1606        GETCHARINCTEST(c, eptr);
1607        switch(c)
1608          {
1609          default: break;
1610          case 0x0a:      /* LF */
1611          case 0x0b:      /* VT */
1612          case 0x0c:      /* FF */
1613          case 0x0d:      /* CR */
1614          case 0x85:      /* NEL */
1615          case 0x2028:    /* LINE SEPARATOR */
1616          case 0x2029:    /* PARAGRAPH SEPARATOR */
1617          RRETURN(MATCH_NOMATCH);
1618          }
1619        ecode++;
1620        break;
1621    
1622        case OP_VSPACE:
1623        if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
1624        GETCHARINCTEST(c, eptr);
1625        switch(c)
1626          {
1627          default: RRETURN(MATCH_NOMATCH);
1628          case 0x0a:      /* LF */
1629          case 0x0b:      /* VT */
1630          case 0x0c:      /* FF */
1631          case 0x0d:      /* CR */
1632          case 0x85:      /* NEL */
1633          case 0x2028:    /* LINE SEPARATOR */
1634          case 0x2029:    /* PARAGRAPH SEPARATOR */
1635        break;        break;
1636        }        }
1637      ecode++;      ecode++;
# Line 1937  for (;;) Line 2092  for (;;)
2092            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);
2093            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2094            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
2095            BACKCHAR(eptr)            if (utf8) BACKCHAR(eptr);
2096            }            }
2097          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2098          }          }
# Line 2690  for (;;) Line 2845  for (;;)
2845            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
2846              {              {
2847              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2848              GETCHARINC(c, eptr);              GETCHARINCTEST(c, eptr);
2849              }              }
2850            break;            break;
2851    
# Line 2698  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              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
2858              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
2859                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
# Line 2711  for (;;) Line 2866  for (;;)
2866            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
2867              {              {
2868              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2869              GETCHARINC(c, eptr);              GETCHARINCTEST(c, eptr);
2870              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
2871              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
2872                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
# Line 2722  for (;;) Line 2877  for (;;)
2877            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
2878              {              {
2879              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2880              GETCHARINC(c, eptr);              GETCHARINCTEST(c, eptr);
2881              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
2882              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
2883                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
# Line 2733  for (;;) Line 2888  for (;;)
2888            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
2889              {              {
2890              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2891              GETCHARINC(c, eptr);              GETCHARINCTEST(c, eptr);
2892              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
2893              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
2894                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
# Line 2803  for (;;) Line 2958  for (;;)
2958              case 0x000d:              case 0x000d:
2959              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
2960              break;              break;
2961    
2962              case 0x000a:              case 0x000a:
2963                break;
2964    
2965              case 0x000b:              case 0x000b:
2966              case 0x000c:              case 0x000c:
2967              case 0x0085:              case 0x0085:
2968              case 0x2028:              case 0x2028:
2969              case 0x2029:              case 0x2029:
2970                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
2971                break;
2972                }
2973              }
2974            break;
2975    
2976            case OP_NOT_HSPACE:
2977            for (i = 1; i <= min; i++)
2978              {
2979              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2980              GETCHARINC(c, eptr);
2981              switch(c)
2982                {
2983                default: break;
2984                case 0x09:      /* HT */
2985                case 0x20:      /* SPACE */
2986                case 0xa0:      /* NBSP */
2987                case 0x1680:    /* OGHAM SPACE MARK */
2988                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
2989                case 0x2000:    /* EN QUAD */
2990                case 0x2001:    /* EM QUAD */
2991                case 0x2002:    /* EN SPACE */
2992                case 0x2003:    /* EM SPACE */
2993                case 0x2004:    /* THREE-PER-EM SPACE */
2994                case 0x2005:    /* FOUR-PER-EM SPACE */
2995                case 0x2006:    /* SIX-PER-EM SPACE */
2996                case 0x2007:    /* FIGURE SPACE */
2997                case 0x2008:    /* PUNCTUATION SPACE */
2998                case 0x2009:    /* THIN SPACE */
2999                case 0x200A:    /* HAIR SPACE */
3000                case 0x202f:    /* NARROW NO-BREAK SPACE */
3001                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3002                case 0x3000:    /* IDEOGRAPHIC SPACE */
3003                RRETURN(MATCH_NOMATCH);
3004                }
3005              }
3006            break;
3007    
3008            case OP_HSPACE:
3009            for (i = 1; i <= min; i++)
3010              {
3011              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3012              GETCHARINC(c, eptr);
3013              switch(c)
3014                {
3015                default: RRETURN(MATCH_NOMATCH);
3016                case 0x09:      /* HT */
3017                case 0x20:      /* SPACE */
3018                case 0xa0:      /* NBSP */
3019                case 0x1680:    /* OGHAM SPACE MARK */
3020                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3021                case 0x2000:    /* EN QUAD */
3022                case 0x2001:    /* EM QUAD */
3023                case 0x2002:    /* EN SPACE */
3024                case 0x2003:    /* EM SPACE */
3025                case 0x2004:    /* THREE-PER-EM SPACE */
3026                case 0x2005:    /* FOUR-PER-EM SPACE */
3027                case 0x2006:    /* SIX-PER-EM SPACE */
3028                case 0x2007:    /* FIGURE SPACE */
3029                case 0x2008:    /* PUNCTUATION SPACE */
3030                case 0x2009:    /* THIN SPACE */
3031                case 0x200A:    /* HAIR SPACE */
3032                case 0x202f:    /* NARROW NO-BREAK SPACE */
3033                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3034                case 0x3000:    /* IDEOGRAPHIC SPACE */
3035                break;
3036                }
3037              }
3038            break;
3039    
3040            case OP_NOT_VSPACE:
3041            for (i = 1; i <= min; i++)
3042              {
3043              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3044              GETCHARINC(c, eptr);
3045              switch(c)
3046                {
3047                default: break;
3048                case 0x0a:      /* LF */
3049                case 0x0b:      /* VT */
3050                case 0x0c:      /* FF */
3051                case 0x0d:      /* CR */
3052                case 0x85:      /* NEL */
3053                case 0x2028:    /* LINE SEPARATOR */
3054                case 0x2029:    /* PARAGRAPH SEPARATOR */
3055                RRETURN(MATCH_NOMATCH);
3056                }
3057              }
3058            break;
3059    
3060            case OP_VSPACE:
3061            for (i = 1; i <= min; i++)
3062              {
3063              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3064              GETCHARINC(c, eptr);
3065              switch(c)
3066                {
3067                default: RRETURN(MATCH_NOMATCH);
3068                case 0x0a:      /* LF */
3069                case 0x0b:      /* VT */
3070                case 0x0c:      /* FF */
3071                case 0x0d:      /* CR */
3072                case 0x85:      /* NEL */
3073                case 0x2028:    /* LINE SEPARATOR */
3074                case 0x2029:    /* PARAGRAPH SEPARATOR */
3075              break;              break;
3076              }              }
3077            }            }
# Line 2838  for (;;) Line 3101  for (;;)
3101          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3102            {            {
3103            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3104               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_space) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0))
3105              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3106            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3107            }            }
3108          break;          break;
3109    
# Line 2858  for (;;) Line 3121  for (;;)
3121          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3122            {            {
3123            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3124               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_word) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0))
3125              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3126            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3127            }            }
3128          break;          break;
3129    
# Line 2917  for (;;) Line 3180  for (;;)
3180              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3181              break;              break;
3182              case 0x000a:              case 0x000a:
3183                break;
3184    
3185              case 0x000b:              case 0x000b:
3186              case 0x000c:              case 0x000c:
3187              case 0x0085:              case 0x0085:
3188                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3189                break;
3190                }
3191              }
3192            break;
3193    
3194            case OP_NOT_HSPACE:
3195            for (i = 1; i <= min; i++)
3196              {
3197              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3198              switch(*eptr++)
3199                {
3200                default: break;
3201                case 0x09:      /* HT */
3202                case 0x20:      /* SPACE */
3203                case 0xa0:      /* NBSP */
3204                RRETURN(MATCH_NOMATCH);
3205                }
3206              }
3207            break;
3208    
3209            case OP_HSPACE:
3210            for (i = 1; i <= min; i++)
3211              {
3212              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3213              switch(*eptr++)
3214                {
3215                default: RRETURN(MATCH_NOMATCH);
3216                case 0x09:      /* HT */
3217                case 0x20:      /* SPACE */
3218                case 0xa0:      /* NBSP */
3219                break;
3220                }
3221              }
3222            break;
3223    
3224            case OP_NOT_VSPACE:
3225            for (i = 1; i <= min; i++)
3226              {
3227              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3228              switch(*eptr++)
3229                {
3230                default: break;
3231                case 0x0a:      /* LF */
3232                case 0x0b:      /* VT */
3233                case 0x0c:      /* FF */
3234                case 0x0d:      /* CR */
3235                case 0x85:      /* NEL */
3236                RRETURN(MATCH_NOMATCH);
3237                }
3238              }
3239            break;
3240    
3241            case OP_VSPACE:
3242            for (i = 1; i <= min; i++)
3243              {
3244              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
3245              switch(*eptr++)
3246                {
3247                default: RRETURN(MATCH_NOMATCH);
3248                case 0x0a:      /* LF */
3249                case 0x0b:      /* VT */
3250                case 0x0c:      /* FF */
3251                case 0x0d:      /* CR */
3252                case 0x85:      /* NEL */
3253              break;              break;
3254              }              }
3255            }            }
# Line 3107  for (;;) Line 3437  for (;;)
3437                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3438                break;                break;
3439                case 0x000a:                case 0x000a:
3440                  break;
3441    
3442                case 0x000b:                case 0x000b:
3443                case 0x000c:                case 0x000c:
3444                case 0x0085:                case 0x0085:
3445                case 0x2028:                case 0x2028:
3446                case 0x2029:                case 0x2029:
3447                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3448                  break;
3449                  }
3450                break;
3451    
3452                case OP_NOT_HSPACE:
3453                switch(c)
3454                  {
3455                  default: break;
3456                  case 0x09:      /* HT */
3457                  case 0x20:      /* SPACE */
3458                  case 0xa0:      /* NBSP */
3459                  case 0x1680:    /* OGHAM SPACE MARK */
3460                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3461                  case 0x2000:    /* EN QUAD */
3462                  case 0x2001:    /* EM QUAD */
3463                  case 0x2002:    /* EN SPACE */
3464                  case 0x2003:    /* EM SPACE */
3465                  case 0x2004:    /* THREE-PER-EM SPACE */
3466                  case 0x2005:    /* FOUR-PER-EM SPACE */
3467                  case 0x2006:    /* SIX-PER-EM SPACE */
3468                  case 0x2007:    /* FIGURE SPACE */
3469                  case 0x2008:    /* PUNCTUATION SPACE */
3470                  case 0x2009:    /* THIN SPACE */
3471                  case 0x200A:    /* HAIR SPACE */
3472                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3473                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3474                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3475                  RRETURN(MATCH_NOMATCH);
3476                  }
3477                break;
3478    
3479                case OP_HSPACE:
3480                switch(c)
3481                  {
3482                  default: RRETURN(MATCH_NOMATCH);
3483                  case 0x09:      /* HT */
3484                  case 0x20:      /* SPACE */
3485                  case 0xa0:      /* NBSP */
3486                  case 0x1680:    /* OGHAM SPACE MARK */
3487                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3488                  case 0x2000:    /* EN QUAD */
3489                  case 0x2001:    /* EM QUAD */
3490                  case 0x2002:    /* EN SPACE */
3491                  case 0x2003:    /* EM SPACE */
3492                  case 0x2004:    /* THREE-PER-EM SPACE */
3493                  case 0x2005:    /* FOUR-PER-EM SPACE */
3494                  case 0x2006:    /* SIX-PER-EM SPACE */
3495                  case 0x2007:    /* FIGURE SPACE */
3496                  case 0x2008:    /* PUNCTUATION SPACE */
3497                  case 0x2009:    /* THIN SPACE */
3498                  case 0x200A:    /* HAIR SPACE */
3499                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3500                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3501                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3502                  break;
3503                  }
3504                break;
3505    
3506                case OP_NOT_VSPACE:
3507                switch(c)
3508                  {
3509                  default: break;
3510                  case 0x0a:      /* LF */
3511                  case 0x0b:      /* VT */
3512                  case 0x0c:      /* FF */
3513                  case 0x0d:      /* CR */
3514                  case 0x85:      /* NEL */
3515                  case 0x2028:    /* LINE SEPARATOR */
3516                  case 0x2029:    /* PARAGRAPH SEPARATOR */
3517                  RRETURN(MATCH_NOMATCH);
3518                  }
3519                break;
3520    
3521                case OP_VSPACE:
3522                switch(c)
3523                  {
3524                  default: RRETURN(MATCH_NOMATCH);
3525                  case 0x0a:      /* LF */
3526                  case 0x0b:      /* VT */
3527                  case 0x0c:      /* FF */
3528                  case 0x0d:      /* CR */
3529                  case 0x85:      /* NEL */
3530                  case 0x2028:    /* LINE SEPARATOR */
3531                  case 0x2029:    /* PARAGRAPH SEPARATOR */
3532                break;                break;
3533                }                }
3534              break;              break;
# Line 3179  for (;;) Line 3596  for (;;)
3596                case 0x000d:                case 0x000d:
3597                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3598                break;                break;
3599    
3600                case 0x000a:                case 0x000a:
3601                  break;
3602    
3603                case 0x000b:                case 0x000b:
3604                case 0x000c:                case 0x000c:
3605                case 0x0085:                case 0x0085:
3606                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3607                  break;
3608                  }
3609                break;
3610    
3611                case OP_NOT_HSPACE:
3612                switch(c)
3613                  {
3614                  default: break;
3615                  case 0x09:      /* HT */
3616                  case 0x20:      /* SPACE */
3617                  case 0xa0:      /* NBSP */
3618                  RRETURN(MATCH_NOMATCH);
3619                  }
3620                break;
3621    
3622                case OP_HSPACE:
3623                switch(c)
3624                  {
3625                  default: RRETURN(MATCH_NOMATCH);
3626                  case 0x09:      /* HT */
3627                  case 0x20:      /* SPACE */
3628                  case 0xa0:      /* NBSP */
3629                  break;
3630                  }
3631                break;
3632    
3633                case OP_NOT_VSPACE:
3634                switch(c)
3635                  {
3636                  default: break;
3637                  case 0x0a:      /* LF */
3638                  case 0x0b:      /* VT */
3639                  case 0x0c:      /* FF */
3640                  case 0x0d:      /* CR */
3641                  case 0x85:      /* NEL */
3642                  RRETURN(MATCH_NOMATCH);
3643                  }
3644                break;
3645    
3646                case OP_VSPACE:
3647                switch(c)
3648                  {
3649                  default: RRETURN(MATCH_NOMATCH);
3650                  case 0x0a:      /* LF */
3651                  case 0x0b:      /* VT */
3652                  case 0x0c:      /* FF */
3653                  case 0x0d:      /* CR */
3654                  case 0x85:      /* NEL */
3655                break;                break;
3656                }                }
3657              break;              break;
# Line 3306  for (;;) Line 3775  for (;;)
3775            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);
3776            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3777            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
3778            BACKCHAR(eptr);            if (utf8) BACKCHAR(eptr);
3779            }            }
3780          }          }
3781    
# Line 3345  for (;;) Line 3814  for (;;)
3814            for (;;)                        /* Move back over one extended */            for (;;)                        /* Move back over one extended */
3815              {              {
3816              int len = 1;              int len = 1;
             BACKCHAR(eptr);  
3817              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr; else
3818                {                {
3819                  BACKCHAR(eptr);
3820                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
3821                }                }
3822              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
# Line 3368  for (;;) Line 3837  for (;;)
3837          switch(ctype)          switch(ctype)
3838            {            {
3839            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. */  
   
3840            if (max < INT_MAX)            if (max < INT_MAX)
3841              {              {
3842              if ((ims & PCRE_DOTALL) == 0)              if ((ims & PCRE_DOTALL) == 0)
# Line 3405  for (;;) Line 3869  for (;;)
3869                  {                  {
3870                  if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;                  if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
3871                  eptr++;                  eptr++;
3872                    while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3873                  }                  }
               break;  
3874                }                }
3875              else              else
3876                {                {
3877                c = max - min;                eptr = md->end_subject;
               if (c > (unsigned int)(md->end_subject - eptr))  
                 c = md->end_subject - eptr;  
               eptr += c;  
3878                }                }
3879              }              }
3880            break;            break;
# Line 3440  for (;;) Line 3901  for (;;)
3901                }                }
3902              else              else
3903                {                {
3904                if (c != 0x000a && c != 0x000b && c != 0x000c &&                if (c != 0x000a &&
3905                    c != 0x0085 && c != 0x2028 && c != 0x2029)                    (md->bsr_anycrlf ||
3906                       (c != 0x000b && c != 0x000c &&
3907                        c != 0x0085 && c != 0x2028 && c != 0x2029)))
3908                  break;                  break;
3909                eptr += len;                eptr += len;
3910                }                }
3911              }              }
3912            break;            break;
3913    
3914              case OP_NOT_HSPACE:
3915              case OP_HSPACE:
3916              for (i = min; i < max; i++)
3917                {
3918                BOOL gotspace;
3919                int len = 1;
3920                if (eptr >= md->end_subject) break;
3921                GETCHARLEN(c, eptr, len);
3922                switch(c)
3923                  {
3924                  default: gotspace = FALSE; break;
3925                  case 0x09:      /* HT */
3926                  case 0x20:      /* SPACE */
3927                  case 0xa0:      /* NBSP */
3928                  case 0x1680:    /* OGHAM SPACE MARK */
3929                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3930                  case 0x2000:    /* EN QUAD */
3931                  case 0x2001:    /* EM QUAD */
3932                  case 0x2002:    /* EN SPACE */
3933                  case 0x2003:    /* EM SPACE */
3934                  case 0x2004:    /* THREE-PER-EM SPACE */
3935                  case 0x2005:    /* FOUR-PER-EM SPACE */
3936                  case 0x2006:    /* SIX-PER-EM SPACE */
3937                  case 0x2007:    /* FIGURE SPACE */
3938                  case 0x2008:    /* PUNCTUATION SPACE */
3939                  case 0x2009:    /* THIN SPACE */
3940                  case 0x200A:    /* HAIR SPACE */
3941                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3942                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3943                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3944                  gotspace = TRUE;
3945                  break;
3946                  }
3947                if (gotspace == (ctype == OP_NOT_HSPACE)) break;
3948                eptr += len;
3949                }
3950              break;
3951    
3952              case OP_NOT_VSPACE:
3953              case OP_VSPACE:
3954              for (i = min; i < max; i++)
3955                {
3956                BOOL gotspace;
3957                int len = 1;
3958                if (eptr >= md->end_subject) break;
3959                GETCHARLEN(c, eptr, len);
3960                switch(c)
3961                  {
3962                  default: gotspace = FALSE; break;
3963                  case 0x0a:      /* LF */
3964                  case 0x0b:      /* VT */
3965                  case 0x0c:      /* FF */
3966                  case 0x0d:      /* CR */
3967                  case 0x85:      /* NEL */
3968                  case 0x2028:    /* LINE SEPARATOR */
3969                  case 0x2029:    /* PARAGRAPH SEPARATOR */
3970                  gotspace = TRUE;
3971                  break;
3972                  }
3973                if (gotspace == (ctype == OP_NOT_VSPACE)) break;
3974                eptr += len;
3975                }
3976              break;
3977    
3978            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
3979            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3980              {              {
# Line 3530  for (;;) Line 4057  for (;;)
4057            }            }
4058          }          }
4059        else        else
4060  #endif  #endif  /* SUPPORT_UTF8 */
4061    
4062        /* Not UTF-8 mode */        /* Not UTF-8 mode */
4063          {          {
# Line 3567  for (;;) Line 4094  for (;;)
4094                }                }
4095              else              else
4096                {                {
4097                if (c != 0x000a && c != 0x000b && c != 0x000c && c != 0x0085)                if (c != 0x000a &&
4098                      (md->bsr_anycrlf ||
4099                        (c != 0x000b && c != 0x000c && c != 0x0085)))
4100                  break;                  break;
4101                eptr++;                eptr++;
4102                }                }
4103              }              }
4104            break;            break;
4105    
4106              case OP_NOT_HSPACE:
4107              for (i = min; i < max; i++)
4108                {
4109                if (eptr >= md->end_subject) break;
4110                c = *eptr;
4111                if (c == 0x09 || c == 0x20 || c == 0xa0) break;
4112                eptr++;
4113                }
4114              break;
4115    
4116              case OP_HSPACE:
4117              for (i = min; i < max; i++)
4118                {
4119                if (eptr >= md->end_subject) break;
4120                c = *eptr;
4121                if (c != 0x09 && c != 0x20 && c != 0xa0) break;
4122                eptr++;
4123                }
4124              break;
4125    
4126              case OP_NOT_VSPACE:
4127              for (i = min; i < max; i++)
4128                {
4129                if (eptr >= md->end_subject) break;
4130                c = *eptr;
4131                if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)
4132                  break;
4133                eptr++;
4134                }
4135              break;
4136    
4137              case OP_VSPACE:
4138              for (i = min; i < max; i++)
4139                {
4140                if (eptr >= md->end_subject) break;
4141                c = *eptr;
4142                if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85)
4143                  break;
4144                eptr++;
4145                }
4146              break;
4147    
4148            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
4149            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4150              {              {
# Line 3675  HEAP_RETURN: Line 4246  HEAP_RETURN:
4246  switch (frame->Xwhere)  switch (frame->Xwhere)
4247    {    {
4248    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)
4249    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)
4250    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)
4251    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)
4252    LBL(33) LBL(34) LBL(35) LBL(36) LBL(37) LBL(38) LBL(39) LBL(40)    LBL(53) LBL(54)
4253    LBL(41) LBL(42) LBL(43) LBL(44) LBL(45) LBL(46) LBL(47)  #ifdef SUPPORT_UTF8
4254      LBL(16) LBL(18) LBL(20) LBL(21) LBL(22) LBL(23) LBL(28) LBL(30)
4255      LBL(32) LBL(34) LBL(42) LBL(46)
4256    #ifdef SUPPORT_UCP
4257      LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
4258    #endif  /* SUPPORT_UCP */
4259    #endif  /* SUPPORT_UTF8 */
4260    default:    default:
4261    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));
4262    return PCRE_ERROR_INTERNAL;    return PCRE_ERROR_INTERNAL;
# Line 3796  const uschar *start_bits = NULL; Line 4373  const uschar *start_bits = NULL;
4373  USPTR start_match = (USPTR)subject + start_offset;  USPTR start_match = (USPTR)subject + start_offset;
4374  USPTR end_subject;  USPTR end_subject;
4375  USPTR req_byte_ptr = start_match - 1;  USPTR req_byte_ptr = start_match - 1;
 eptrblock eptrchain[EPTR_WORK_SIZE];  
4376    
4377  pcre_study_data internal_study;  pcre_study_data internal_study;
4378  const pcre_study_data *study;  const pcre_study_data *study;
# Line 3859  if (re->magic_number != MAGIC_NUMBER) Line 4435  if (re->magic_number != MAGIC_NUMBER)
4435  /* Set up other data */  /* Set up other data */
4436    
4437  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
4438  startline = (re->options & PCRE_STARTLINE) != 0;  startline = (re->flags & PCRE_STARTLINE) != 0;
4439  firstline = (re->options & PCRE_FIRSTLINE) != 0;  firstline = (re->options & PCRE_FIRSTLINE) != 0;
4440    
4441  /* 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 4458  md->partial = (options & PCRE_PARTIAL) !
4458  md->hitend = FALSE;  md->hitend = FALSE;
4459    
4460  md->recursive = NULL;                   /* No recursion at top level */  md->recursive = NULL;                   /* No recursion at top level */
 md->eptrchain = eptrchain;              /* Make workspace generally available */  
4461    
4462  md->lcc = tables + lcc_offset;  md->lcc = tables + lcc_offset;
4463  md->ctypes = tables + ctypes_offset;  md->ctypes = tables + ctypes_offset;
4464    
4465    /* Handle different \R options. */
4466    
4467    switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
4468      {
4469      case 0:
4470      if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
4471        md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0;
4472      else
4473    #ifdef BSR_ANYCRLF
4474      md->bsr_anycrlf = TRUE;
4475    #else
4476      md->bsr_anycrlf = FALSE;
4477    #endif
4478      break;
4479    
4480      case PCRE_BSR_ANYCRLF:
4481      md->bsr_anycrlf = TRUE;
4482      break;
4483    
4484      case PCRE_BSR_UNICODE:
4485      md->bsr_anycrlf = FALSE;
4486      break;
4487    
4488      default: return PCRE_ERROR_BADNEWLINE;
4489      }
4490    
4491  /* Handle different types of newline. The three bits give eight cases. If  /* Handle different types of newline. The three bits give eight cases. If
4492  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. */
4493    
4494  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
4495         PCRE_NEWLINE_BITS)          (pcre_uint32)options) & PCRE_NEWLINE_BITS)
4496    {    {
4497    case 0: newline = NEWLINE; break;   /* Compile-time default */    case 0: newline = NEWLINE; break;   /* Compile-time default */
4498    case PCRE_NEWLINE_CR: newline = '\r'; break;    case PCRE_NEWLINE_CR: newline = '\r'; break;
# Line 3930  else Line 4531  else
4531  /* 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
4532  moment. */  moment. */
4533    
4534  if (md->partial && (re->options & PCRE_NOPARTIAL) != 0)  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
4535    return PCRE_ERROR_BADPARTIAL;    return PCRE_ERROR_BADPARTIAL;
4536    
4537  /* 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 4608  studied, there may be a bitmap of possib
4608    
4609  if (!anchored)  if (!anchored)
4610    {    {
4611    if ((re->options & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
4612      {      {
4613      first_byte = re->first_byte & 255;      first_byte = re->first_byte & 255;
4614      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 4623  if (!anchored)
4623  /* For anchored or unanchored matches, there may be a "last known required  /* For anchored or unanchored matches, there may be a "last known required
4624  character" set. */  character" set. */
4625    
4626  if ((re->options & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
4627    {    {
4628    req_byte = re->req_byte & 255;    req_byte = re->req_byte & 255;
4629    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 4639  the loop runs just once. */
4639  for(;;)  for(;;)
4640    {    {
4641    USPTR save_end_subject = end_subject;    USPTR save_end_subject = end_subject;
4642      USPTR new_start_match;
4643    
4644    /* Reset the maximum number of extractions we might see. */    /* Reset the maximum number of extractions we might see. */
4645    
# Line 4068  for(;;) Line 4670  for(;;)
4670      if (first_byte_caseless)      if (first_byte_caseless)
4671        while (start_match < end_subject &&        while (start_match < end_subject &&
4672               md->lcc[*start_match] != first_byte)               md->lcc[*start_match] != first_byte)
4673          start_match++;          { NEXTCHAR(start_match); }
4674      else      else
4675        while (start_match < end_subject && *start_match != first_byte)        while (start_match < end_subject && *start_match != first_byte)
4676          start_match++;          { NEXTCHAR(start_match); }
4677      }      }
4678    
4679    /* 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 4683  for(;;)
4683      if (start_match > md->start_subject + start_offset)      if (start_match > md->start_subject + start_offset)
4684        {        {
4685        while (start_match <= end_subject && !WAS_NEWLINE(start_match))        while (start_match <= end_subject && !WAS_NEWLINE(start_match))
4686          start_match++;          { NEXTCHAR(start_match); }
4687    
4688        /* 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,
4689        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 4704  for(;;)
4704      while (start_match < end_subject)      while (start_match < end_subject)
4705        {        {
4706        register unsigned int c = *start_match;        register unsigned int c = *start_match;
4707        if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; else break;        if ((start_bits[c/8] & (1 << (c&7))) == 0)
4708            { NEXTCHAR(start_match); }
4709          else break;
4710        }        }
4711      }      }
4712    
# Line 4178  for(;;) Line 4782  for(;;)
4782    
4783    /* OK, we can now run the match. */    /* OK, we can now run the match. */
4784    
4785    md->start_match_ptr = start_match;      /* Insurance */    md->start_match_ptr = start_match;
4786    md->match_call_count = 0;    md->match_call_count = 0;
4787    md->eptrn = 0;                          /* Next free eptrchain slot */    rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0);
4788    rc = match(start_match, md->start_code, start_match, 2, md,  
4789      ims, NULL, 0, 0);    switch(rc)
4790        {
4791        /* NOMATCH and PRUNE advance by one character. THEN at this level acts
4792        exactly like PRUNE. */
4793    
4794        case MATCH_NOMATCH:
4795        case MATCH_PRUNE:
4796        case MATCH_THEN:
4797        new_start_match = start_match + 1;
4798    #ifdef SUPPORT_UTF8
4799        if (utf8)
4800          while(new_start_match < end_subject && (*new_start_match & 0xc0) == 0x80)
4801            new_start_match++;
4802    #endif
4803        break;
4804    
4805        /* SKIP passes back the next starting point explicitly. */
4806    
4807        case MATCH_SKIP:
4808        new_start_match = md->start_match_ptr;
4809        break;
4810    
4811        /* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */
4812    
4813        case MATCH_COMMIT:
4814        rc = MATCH_NOMATCH;
4815        goto ENDLOOP;
4816    
4817        /* Any other return is some kind of error. */
4818    
4819        default:
4820        goto ENDLOOP;
4821        }
4822    
4823    /* Any return other than MATCH_NOMATCH breaks the loop. */    /* Control reaches here for the various types of "no match at this point"
4824      result. Reset the code to MATCH_NOMATCH for subsequent checking. */
4825    
4826    if (rc != MATCH_NOMATCH) break;    rc = MATCH_NOMATCH;
4827    
4828    /* 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
4829    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 4831  for(;;)
4831    
4832    if (firstline && IS_NEWLINE(start_match)) break;    if (firstline && IS_NEWLINE(start_match)) break;
4833    
4834    /* Advance the match position by one character. */    /* Advance to new matching position */
4835    
4836    start_match++;    start_match = new_start_match;
 #ifdef SUPPORT_UTF8  
   if (utf8)  
     while(start_match < end_subject && (*start_match & 0xc0) == 0x80)  
       start_match++;  
 #endif  
4837    
4838    /* 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
4839    the subject. */    the subject. */
4840    
4841    if (anchored || start_match > end_subject) break;    if (anchored || start_match > end_subject) break;
4842    
4843    /* 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
4844    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
4845    character. */    or ANY or ANYCRLF, advance the match position by one more character. */
4846    
4847    if (start_match[-1] == '\r' &&    if (start_match[-1] == '\r' &&
4848         (md->nltype == NLTYPE_ANY ||        start_match < end_subject &&
4849          md->nltype == NLTYPE_ANYCRLF ||        *start_match == '\n' &&
4850          md->nllen == 2) &&        (re->flags & PCRE_HASCRORLF) == 0 &&
4851         start_match < end_subject &&          (md->nltype == NLTYPE_ANY ||
4852         *start_match == '\n')           md->nltype == NLTYPE_ANYCRLF ||
4853             md->nllen == 2))
4854      start_match++;      start_match++;
4855    
4856    }   /* End of for(;;) "bumpalong" loop */    }   /* End of for(;;) "bumpalong" loop */
# Line 4227  for(;;) Line 4860  for(;;)
4860  /* 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
4861  conditions is true:  conditions is true:
4862    
4863  (1) The pattern is anchored;  (1) The pattern is anchored or the match was failed by (*COMMIT);
4864    
4865  (2) We are past the end of the subject;  (2) We are past the end of the subject;
4866    
# Line 4242  processing, copy those that we can. In t Line 4875  processing, copy those that we can. In t
4875  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
4876  capturing parentheses than vector slots. */  capturing parentheses than vector slots. */
4877    
4878    ENDLOOP:
4879    
4880  if (rc == MATCH_MATCH)  if (rc == MATCH_MATCH)
4881    {    {
4882    if (using_temporary_offsets)    if (using_temporary_offsets)

Legend:
Removed from v.172  
changed lines
  Added in v.294

  ViewVC Help
Powered by ViewVC 1.1.5