/[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 305 by ph10, Sun Jan 20 20:07:32 2008 UTC revision 475 by ph10, Sat Jan 2 18:21:30 2010 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) 1997-2008 University of Cambridge             Copyright (c) 1997-2010 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 89  static const char rep_max[] = { 0, 0, 0, Line 89  static const char rep_max[] = { 0, 0, 0,
89    
90    
91    
92  #ifdef DEBUG  #ifdef PCRE_DEBUG
93  /*************************************************  /*************************************************
94  *        Debugging function to print chars       *  *        Debugging function to print chars       *
95  *************************************************/  *************************************************/
# Line 141  match_ref(int offset, register USPTR ept Line 141  match_ref(int offset, register USPTR ept
141  {  {
142  USPTR p = md->start_subject + md->offset_vector[offset];  USPTR p = md->start_subject + md->offset_vector[offset];
143    
144  #ifdef DEBUG  #ifdef PCRE_DEBUG
145  if (eptr >= md->end_subject)  if (eptr >= md->end_subject)
146    printf("matching subject <null>");    printf("matching subject <null>");
147  else  else
# Line 158  printf("\n"); Line 158  printf("\n");
158    
159  if (length > md->end_subject - eptr) return FALSE;  if (length > md->end_subject - eptr) return FALSE;
160    
161  /* Separate the caselesss case for speed */  /* Separate the caseless case for speed. In UTF-8 mode we can only do this
162    properly if Unicode properties are supported. Otherwise, we can check only
163    ASCII characters. */
164    
165  if ((ims & PCRE_CASELESS) != 0)  if ((ims & PCRE_CASELESS) != 0)
166    {    {
167    #ifdef SUPPORT_UTF8
168    #ifdef SUPPORT_UCP
169      if (md->utf8)
170        {
171        USPTR endptr = eptr + length;
172        while (eptr < endptr)
173          {
174          int c, d;
175          GETCHARINC(c, eptr);
176          GETCHARINC(d, p);
177          if (c != d && c != UCD_OTHERCASE(d)) return FALSE;
178          }
179        }
180      else
181    #endif
182    #endif
183    
184      /* The same code works when not in UTF-8 mode and in UTF-8 mode when there
185      is no UCP support. */
186    
187    while (length-- > 0)    while (length-- > 0)
188      if (md->lcc[*p++] != md->lcc[*eptr++]) return FALSE;      { if (md->lcc[*p++] != md->lcc[*eptr++]) return FALSE; }
189    }    }
190    
191    /* In the caseful case, we can just compare the bytes, whether or not we
192    are in UTF-8 mode. */
193    
194  else  else
195    { while (length-- > 0) if (*p++ != *eptr++) return FALSE; }    { while (length-- > 0) if (*p++ != *eptr++) return FALSE; }
196    
# Line 228  actuall used in this definition. */ Line 254  actuall used in this definition. */
254  #ifndef NO_RECURSE  #ifndef NO_RECURSE
255  #define REGISTER register  #define REGISTER register
256    
257  #ifdef DEBUG  #ifdef PCRE_DEBUG
258  #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw) \  #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw) \
259    { \    { \
260    printf("match() called in line %d\n", __LINE__); \    printf("match() called in line %d\n", __LINE__); \
# Line 296  typedef struct heapframe { Line 322  typedef struct heapframe {
322    
323    /* Function arguments that may change */    /* Function arguments that may change */
324    
325    const uschar *Xeptr;    USPTR Xeptr;
326    const uschar *Xecode;    const uschar *Xecode;
327    const uschar *Xmstart;    USPTR Xmstart;
328    int Xoffset_top;    int Xoffset_top;
329    long int Xims;    long int Xims;
330    eptrblock *Xeptrb;    eptrblock *Xeptrb;
# Line 307  typedef struct heapframe { Line 333  typedef struct heapframe {
333    
334    /* Function local variables */    /* Function local variables */
335    
336    const uschar *Xcallpat;    USPTR Xcallpat;
337    const uschar *Xcharptr;  #ifdef SUPPORT_UTF8
338    const uschar *Xdata;    USPTR Xcharptr;
339    const uschar *Xnext;  #endif
340    const uschar *Xpp;    USPTR Xdata;
341    const uschar *Xprev;    USPTR Xnext;
342    const uschar *Xsaved_eptr;    USPTR Xpp;
343      USPTR Xprev;
344      USPTR Xsaved_eptr;
345    
346    recursion_info Xnew_recursive;    recursion_info Xnew_recursive;
347    
# Line 334  typedef struct heapframe { Line 362  typedef struct heapframe {
362    uschar Xocchars[8];    uschar Xocchars[8];
363  #endif  #endif
364    
365      int Xcodelink;
366    int Xctype;    int Xctype;
367    unsigned int Xfc;    unsigned int Xfc;
368    int Xfi;    int Xfi;
# Line 369  typedef struct heapframe { Line 398  typedef struct heapframe {
398    
399  /* This function is called recursively in many circumstances. Whenever it  /* This function is called recursively in many circumstances. Whenever it
400  returns a negative (error) response, the outer incarnation must also return the  returns a negative (error) response, the outer incarnation must also return the
401  same response.  same response. */
402    
403    /* These macros pack up tests that are used for partial matching, and which
404    appears several times in the code. We set the "hit end" flag if the pointer is
405    at the end of the subject and also past the start of the subject (i.e.
406    something has been matched). For hard partial matching, we then return
407    immediately. The second one is used when we already know we are past the end of
408    the subject. */
409    
410  Performance note: It might be tempting to extract commonly used fields from the  #define CHECK_PARTIAL()\
411  md structure (e.g. utf8, end_subject) into individual variables to improve    if (md->partial != 0 && eptr >= md->end_subject && eptr > mstart)\
412        {\
413        md->hitend = TRUE;\
414        if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);\
415        }
416    
417    #define SCHECK_PARTIAL()\
418      if (md->partial != 0 && eptr > mstart)\
419        {\
420        md->hitend = TRUE;\
421        if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);\
422        }
423    
424    
425    /* Performance note: It might be tempting to extract commonly used fields from
426    the md structure (e.g. utf8, end_subject) into individual variables to improve
427  performance. Tests using gcc on a SPARC disproved this; in the first case, it  performance. Tests using gcc on a SPARC disproved this; in the first case, it
428  made performance worse.  made performance worse.
429    
# Line 399  Returns: MATCH_MATCH if matched Line 450  Returns: MATCH_MATCH if matched
450  */  */
451    
452  static int  static int
453  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, const uschar *mstart,  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, USPTR mstart,
454    int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,    int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,
455    int flags, unsigned int rdepth)    int flags, unsigned int rdepth)
456  {  {
# Line 413  register unsigned int c; /* Character Line 464  register unsigned int c; /* Character
464  register BOOL utf8;        /* Local copy of UTF-8 flag for speed */  register BOOL utf8;        /* Local copy of UTF-8 flag for speed */
465    
466  BOOL minimize, possessive; /* Quantifier options */  BOOL minimize, possessive; /* Quantifier options */
467    int condcode;
468    
469  /* When recursion is not being used, all "local" variables that have to be  /* When recursion is not being used, all "local" variables that have to be
470  preserved over calls to RMATCH() are part of a "frame" which is obtained from  preserved over calls to RMATCH() are part of a "frame" which is obtained from
# Line 455  HEAP_RECURSE: Line 507  HEAP_RECURSE:
507  #define charptr            frame->Xcharptr  #define charptr            frame->Xcharptr
508  #endif  #endif
509  #define callpat            frame->Xcallpat  #define callpat            frame->Xcallpat
510    #define codelink           frame->Xcodelink
511  #define data               frame->Xdata  #define data               frame->Xdata
512  #define next               frame->Xnext  #define next               frame->Xnext
513  #define pp                 frame->Xpp  #define pp                 frame->Xpp
# Line 535  int oclength; Line 588  int oclength;
588  uschar occhars[8];  uschar occhars[8];
589  #endif  #endif
590    
591    int codelink;
592  int ctype;  int ctype;
593  int length;  int length;
594  int max;  int max;
# Line 568  TAIL_RECURSE: Line 622  TAIL_RECURSE:
622  /* OK, now we can get on with the real code of the function. Recursive calls  /* OK, now we can get on with the real code of the function. Recursive calls
623  are specified by the macro RMATCH and RRETURN is used to return. When  are specified by the macro RMATCH and RRETURN is used to return. When
624  NO_RECURSE is *not* defined, these just turn into a recursive call to match()  NO_RECURSE is *not* defined, these just turn into a recursive call to match()
625  and a "return", respectively (possibly with some debugging if DEBUG is  and a "return", respectively (possibly with some debugging if PCRE_DEBUG is
626  defined). However, RMATCH isn't like a function call because it's quite a  defined). However, RMATCH isn't like a function call because it's quite a
627  complicated macro. It has to be used in one particular way. This shouldn't,  complicated macro. It has to be used in one particular way. This shouldn't,
628  however, impact performance when true recursion is being used. */  however, impact performance when true recursion is being used. */
# Line 610  for (;;) Line 664  for (;;)
664    minimize = possessive = FALSE;    minimize = possessive = FALSE;
665    op = *ecode;    op = *ecode;
666    
   /* For partial matching, remember if we ever hit the end of the subject after  
   matching at least one subject character. */  
   
   if (md->partial &&  
       eptr >= md->end_subject &&  
       eptr > mstart)  
     md->hitend = TRUE;  
   
667    switch(op)    switch(op)
668      {      {
669      case OP_FAIL:      case OP_FAIL:
# Line 667  for (;;) Line 713  for (;;)
713      number = GET2(ecode, 1+LINK_SIZE);      number = GET2(ecode, 1+LINK_SIZE);
714      offset = number << 1;      offset = number << 1;
715    
716  #ifdef DEBUG  #ifdef PCRE_DEBUG
717      printf("start bracket %d\n", number);      printf("start bracket %d\n", number);
718      printf("subject=");      printf("subject=");
719      pchars(eptr, 16, TRUE, md);      pchars(eptr, 16, TRUE, md);
# Line 761  for (;;) Line 807  for (;;)
807    
808      case OP_COND:      case OP_COND:
809      case OP_SCOND:      case OP_SCOND:
810      if (ecode[LINK_SIZE+1] == OP_RREF)         /* Recursion test */      codelink= GET(ecode, 1);
811    
812        /* Because of the way auto-callout works during compile, a callout item is
813        inserted between OP_COND and an assertion condition. */
814    
815        if (ecode[LINK_SIZE+1] == OP_CALLOUT)
816        {        {
817        offset = GET2(ecode, LINK_SIZE + 2);     /* Recursion group number*/        if (pcre_callout != NULL)
818        condition = md->recursive != NULL &&          {
819          (offset == RREF_ANY || offset == md->recursive->group_num);          pcre_callout_block cb;
820        ecode += condition? 3 : GET(ecode, 1);          cb.version          = 1;   /* Version 1 of the callout block */
821            cb.callout_number   = ecode[LINK_SIZE+2];
822            cb.offset_vector    = md->offset_vector;
823            cb.subject          = (PCRE_SPTR)md->start_subject;
824            cb.subject_length   = md->end_subject - md->start_subject;
825            cb.start_match      = mstart - md->start_subject;
826            cb.current_position = eptr - md->start_subject;
827            cb.pattern_position = GET(ecode, LINK_SIZE + 3);
828            cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);
829            cb.capture_top      = offset_top/2;
830            cb.capture_last     = md->capture_last;
831            cb.callout_data     = md->callout_data;
832            if ((rrc = (*pcre_callout)(&cb)) > 0) RRETURN(MATCH_NOMATCH);
833            if (rrc < 0) RRETURN(rrc);
834            }
835          ecode += _pcre_OP_lengths[OP_CALLOUT];
836          }
837    
838        condcode = ecode[LINK_SIZE+1];
839    
840        /* Now see what the actual condition is */
841    
842        if (condcode == OP_RREF || condcode == OP_NRREF)    /* Recursion test */
843          {
844          if (md->recursive == NULL)                /* Not recursing => FALSE */
845            {
846            condition = FALSE;
847            ecode += GET(ecode, 1);
848            }
849          else
850            {
851            int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/
852            condition =  (recno == RREF_ANY || recno == md->recursive->group_num);
853    
854            /* If the test is for recursion into a specific subpattern, and it is
855            false, but the test was set up by name, scan the table to see if the
856            name refers to any other numbers, and test them. The condition is true
857            if any one is set. */
858    
859            if (!condition && condcode == OP_NRREF && recno != RREF_ANY)
860              {
861              uschar *slotA = md->name_table;
862              for (i = 0; i < md->name_count; i++)
863                {
864                if (GET2(slotA, 0) == recno) break;
865                slotA += md->name_entry_size;
866                }
867    
868              /* Found a name for the number - there can be only one; duplicate
869              names for different numbers are allowed, but not vice versa. First
870              scan down for duplicates. */
871    
872              if (i < md->name_count)
873                {
874                uschar *slotB = slotA;
875                while (slotB > md->name_table)
876                  {
877                  slotB -= md->name_entry_size;
878                  if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)
879                    {
880                    condition = GET2(slotB, 0) == md->recursive->group_num;
881                    if (condition) break;
882                    }
883                  else break;
884                  }
885    
886                /* Scan up for duplicates */
887    
888                if (!condition)
889                  {
890                  slotB = slotA;
891                  for (i++; i < md->name_count; i++)
892                    {
893                    slotB += md->name_entry_size;
894                    if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)
895                      {
896                      condition = GET2(slotB, 0) == md->recursive->group_num;
897                      if (condition) break;
898                      }
899                    else break;
900                    }
901                  }
902                }
903              }
904    
905            /* Chose branch according to the condition */
906    
907            ecode += condition? 3 : GET(ecode, 1);
908            }
909        }        }
910    
911      else if (ecode[LINK_SIZE+1] == OP_CREF)    /* Group used test */      else if (condcode == OP_CREF || condcode == OP_NCREF)  /* Group used test */
912        {        {
913        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */
914        condition = offset < offset_top && md->offset_vector[offset] >= 0;        condition = offset < offset_top && md->offset_vector[offset] >= 0;
915    
916          /* If the numbered capture is unset, but the reference was by name,
917          scan the table to see if the name refers to any other numbers, and test
918          them. The condition is true if any one is set. This is tediously similar
919          to the code above, but not close enough to try to amalgamate. */
920    
921          if (!condition && condcode == OP_NCREF)
922            {
923            int refno = offset >> 1;
924            uschar *slotA = md->name_table;
925    
926            for (i = 0; i < md->name_count; i++)
927              {
928              if (GET2(slotA, 0) == refno) break;
929              slotA += md->name_entry_size;
930              }
931    
932            /* Found a name for the number - there can be only one; duplicate names
933            for different numbers are allowed, but not vice versa. First scan down
934            for duplicates. */
935    
936            if (i < md->name_count)
937              {
938              uschar *slotB = slotA;
939              while (slotB > md->name_table)
940                {
941                slotB -= md->name_entry_size;
942                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)
943                  {
944                  offset = GET2(slotB, 0) << 1;
945                  condition = offset < offset_top &&
946                    md->offset_vector[offset] >= 0;
947                  if (condition) break;
948                  }
949                else break;
950                }
951    
952              /* Scan up for duplicates */
953    
954              if (!condition)
955                {
956                slotB = slotA;
957                for (i++; i < md->name_count; i++)
958                  {
959                  slotB += md->name_entry_size;
960                  if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)
961                    {
962                    offset = GET2(slotB, 0) << 1;
963                    condition = offset < offset_top &&
964                      md->offset_vector[offset] >= 0;
965                    if (condition) break;
966                    }
967                  else break;
968                  }
969                }
970              }
971            }
972    
973          /* Chose branch according to the condition */
974    
975        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 3 : GET(ecode, 1);
976        }        }
977    
978      else if (ecode[LINK_SIZE+1] == OP_DEF)     /* DEFINE - always false */      else if (condcode == OP_DEF)     /* DEFINE - always false */
979        {        {
980        condition = FALSE;        condition = FALSE;
981        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
# Line 803  for (;;) Line 1002  for (;;)
1002        else        else
1003          {          {
1004          condition = FALSE;          condition = FALSE;
1005          ecode += GET(ecode, 1);          ecode += codelink;
1006          }          }
1007        }        }
1008    
# Line 826  for (;;) Line 1025  for (;;)
1025          goto TAIL_RECURSE;          goto TAIL_RECURSE;
1026          }          }
1027        }        }
1028      else                         /* Condition false & no 2nd alternative */      else                         /* Condition false & no alternative */
1029        {        {
1030        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1031        }        }
1032      break;      break;
1033    
1034    
1035        /* Before OP_ACCEPT there may be any number of OP_CLOSE opcodes,
1036        to close any currently open capturing brackets. */
1037    
1038        case OP_CLOSE:
1039        number = GET2(ecode, 1);
1040        offset = number << 1;
1041    
1042    #ifdef PCRE_DEBUG
1043          printf("end bracket %d at *ACCEPT", number);
1044          printf("\n");
1045    #endif
1046    
1047        md->capture_last = number;
1048        if (offset >= md->offset_max) md->offset_overflow = TRUE; else
1049          {
1050          md->offset_vector[offset] =
1051            md->offset_vector[md->offset_end - number];
1052          md->offset_vector[offset+1] = eptr - md->start_subject;
1053          if (offset_top <= offset) offset_top = offset + 2;
1054          }
1055        ecode += 3;
1056        break;
1057    
1058    
1059      /* End of the pattern, either real or forced. If we are in a top-level      /* End of the pattern, either real or forced. If we are in a top-level
1060      recursion, we should restore the offsets appropriately and continue from      recursion, we should restore the offsets appropriately and continue from
1061      after the call. */      after the call. */
# Line 846  for (;;) Line 1069  for (;;)
1069        md->recursive = rec->prevrec;        md->recursive = rec->prevrec;
1070        memmove(md->offset_vector, rec->offset_save,        memmove(md->offset_vector, rec->offset_save,
1071          rec->saved_max * sizeof(int));          rec->saved_max * sizeof(int));
1072          offset_top = rec->save_offset_top;
1073        mstart = rec->save_start;        mstart = rec->save_start;
1074        ims = original_ims;        ims = original_ims;
1075        ecode = rec->after_call;        ecode = rec->after_call;
1076        break;        break;
1077        }        }
1078    
1079      /* Otherwise, if PCRE_NOTEMPTY is set, fail if we have matched an empty      /* Otherwise, if we have matched an empty string, fail if PCRE_NOTEMPTY is
1080      string - backtracking will then try other alternatives, if any. */      set, or if PCRE_NOTEMPTY_ATSTART is set and we have matched at the start of
1081        the subject. In both cases, backtracking will then try other alternatives,
1082        if any. */
1083    
1084        if (eptr == mstart &&
1085            (md->notempty ||
1086              (md->notempty_atstart &&
1087                mstart == md->start_subject + md->start_offset)))
1088          RRETURN(MATCH_NOMATCH);
1089    
1090        /* Otherwise, we have a match. */
1091    
     if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);  
1092      md->end_match_ptr = eptr;           /* Record where we ended */      md->end_match_ptr = eptr;           /* Record where we ended */
1093      md->end_offset_top = offset_top;    /* and how many extracts were taken */      md->end_offset_top = offset_top;    /* and how many extracts were taken */
1094      md->start_match_ptr = mstart;       /* and the start (\K can modify) */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
# Line 900  for (;;) Line 1133  for (;;)
1133      offset_top = md->end_offset_top;      offset_top = md->end_offset_top;
1134      continue;      continue;
1135    
1136      /* Negative assertion: all branches must fail to match */      /* Negative assertion: all branches must fail to match. Encountering SKIP,
1137        PRUNE, or COMMIT means we must assume failure without checking subsequent
1138        branches. */
1139    
1140      case OP_ASSERT_NOT:      case OP_ASSERT_NOT:
1141      case OP_ASSERTBACK_NOT:      case OP_ASSERTBACK_NOT:
# Line 909  for (;;) Line 1144  for (;;)
1144        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
1145          RM5);          RM5);
1146        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);
1147          if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)
1148            {
1149            do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1150            break;
1151            }
1152        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1153        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1154        }        }
# Line 946  for (;;) Line 1186  for (;;)
1186        if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);        if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1187        }        }
1188    
1189      /* Skip to next op code */      /* Save the earliest consulted character, then skip to next op code */
1190    
1191        if (eptr < md->start_used_ptr) md->start_used_ptr = eptr;
1192      ecode += 1 + LINK_SIZE;      ecode += 1 + LINK_SIZE;
1193      break;      break;
1194    
# Line 1027  for (;;) Line 1268  for (;;)
1268        memcpy(new_recursive.offset_save, md->offset_vector,        memcpy(new_recursive.offset_save, md->offset_vector,
1269              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
1270        new_recursive.save_start = mstart;        new_recursive.save_start = mstart;
1271          new_recursive.save_offset_top = offset_top;
1272        mstart = eptr;        mstart = eptr;
1273    
1274        /* OK, now we can do the recursion. For each top-level alternative we        /* OK, now we can do the recursion. For each top-level alternative we
# Line 1049  for (;;) Line 1291  for (;;)
1291          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1292            {            {
1293            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1294              if (new_recursive.offset_save != stacksave)
1295                (pcre_free)(new_recursive.offset_save);
1296            RRETURN(rrc);            RRETURN(rrc);
1297            }            }
1298    
# Line 1148  for (;;) Line 1392  for (;;)
1392      do ecode += GET(ecode,1); while (*ecode == OP_ALT);      do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1393      break;      break;
1394    
1395      /* BRAZERO and BRAMINZERO occur just before a bracket group, indicating      /* BRAZERO, BRAMINZERO and SKIPZERO occur just before a bracket group,
1396      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
1397      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
1398      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
1399      preceded by BRAZERO or BRAMINZERO. */      optional ones preceded by BRAZERO or BRAMINZERO. */
1400    
1401      case OP_BRAZERO:      case OP_BRAZERO:
1402        {        {
# Line 1174  for (;;) Line 1418  for (;;)
1418        }        }
1419      break;      break;
1420    
1421        case OP_SKIPZERO:
1422          {
1423          next = ecode+1;
1424          do next += GET(next,1); while (*next == OP_ALT);
1425          ecode = next + 1 + LINK_SIZE;
1426          }
1427        break;
1428    
1429      /* End of a group, repeated or non-repeating. */      /* End of a group, repeated or non-repeating. */
1430    
1431      case OP_KET:      case OP_KET:
# Line 1216  for (;;) Line 1468  for (;;)
1468        number = GET2(prev, 1+LINK_SIZE);        number = GET2(prev, 1+LINK_SIZE);
1469        offset = number << 1;        offset = number << 1;
1470    
1471  #ifdef DEBUG  #ifdef PCRE_DEBUG
1472        printf("end bracket %d", number);        printf("end bracket %d", number);
1473        printf("\n");        printf("\n");
1474  #endif  #endif
# Line 1241  for (;;) Line 1493  for (;;)
1493          mstart = rec->save_start;          mstart = rec->save_start;
1494          memcpy(md->offset_vector, rec->offset_save,          memcpy(md->offset_vector, rec->offset_save,
1495            rec->saved_max * sizeof(int));            rec->saved_max * sizeof(int));
1496            offset_top = rec->save_offset_top;
1497          ecode = rec->after_call;          ecode = rec->after_call;
1498          ims = original_ims;          ims = original_ims;
1499          break;          break;
# Line 1380  for (;;) Line 1633  for (;;)
1633    
1634        /* Find out if the previous and current characters are "word" characters.        /* Find out if the previous and current characters are "word" characters.
1635        It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to        It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to
1636        be "non-word" characters. */        be "non-word" characters. Remember the earliest consulted character for
1637          partial matching. */
1638    
1639  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
1640        if (utf8)        if (utf8)
1641          {          {
1642          if (eptr == md->start_subject) prev_is_word = FALSE; else          if (eptr == md->start_subject) prev_is_word = FALSE; else
1643            {            {
1644            const uschar *lastptr = eptr - 1;            USPTR lastptr = eptr - 1;
1645            while((*lastptr & 0xc0) == 0x80) lastptr--;            while((*lastptr & 0xc0) == 0x80) lastptr--;
1646              if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;
1647            GETCHAR(c, lastptr);            GETCHAR(c, lastptr);
1648            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
1649            }            }
1650          if (eptr >= md->end_subject) cur_is_word = FALSE; else          if (eptr >= md->end_subject)
1651              {
1652              SCHECK_PARTIAL();
1653              cur_is_word = FALSE;
1654              }
1655            else
1656            {            {
1657            GETCHAR(c, eptr);            GETCHAR(c, eptr);
1658            cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;            cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
# Line 1401  for (;;) Line 1661  for (;;)
1661        else        else
1662  #endif  #endif
1663    
1664        /* More streamlined when not in UTF-8 mode */        /* Not in UTF-8 mode */
1665    
1666          {          {
1667          prev_is_word = (eptr != md->start_subject) &&          if (eptr == md->start_subject) prev_is_word = FALSE; else
1668            ((md->ctypes[eptr[-1]] & ctype_word) != 0);            {
1669          cur_is_word = (eptr < md->end_subject) &&            if (eptr <= md->start_used_ptr) md->start_used_ptr = eptr - 1;
1670            ((md->ctypes[*eptr] & ctype_word) != 0);            prev_is_word = ((md->ctypes[eptr[-1]] & ctype_word) != 0);
1671              }
1672            if (eptr >= md->end_subject)
1673              {
1674              SCHECK_PARTIAL();
1675              cur_is_word = FALSE;
1676              }
1677            else cur_is_word = ((md->ctypes[*eptr] & ctype_word) != 0);
1678          }          }
1679    
1680        /* Now see if the situation is what we want */        /* Now see if the situation is what we want */
# Line 1421  for (;;) Line 1688  for (;;)
1688      /* Match a single character type; inline for speed */      /* Match a single character type; inline for speed */
1689    
1690      case OP_ANY:      case OP_ANY:
1691      if ((ims & PCRE_DOTALL) == 0)      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
1692        /* Fall through */
1693    
1694        case OP_ALLANY:
1695        if (eptr++ >= md->end_subject)
1696        {        {
1697        if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);        SCHECK_PARTIAL();
1698          RRETURN(MATCH_NOMATCH);
1699        }        }
1700      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
     if (utf8)  
       while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  
1701      ecode++;      ecode++;
1702      break;      break;
1703    
# Line 1435  for (;;) Line 1705  for (;;)
1705      any byte, even newline, independent of the setting of PCRE_DOTALL. */      any byte, even newline, independent of the setting of PCRE_DOTALL. */
1706    
1707      case OP_ANYBYTE:      case OP_ANYBYTE:
1708      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr++ >= md->end_subject)
1709          {
1710          SCHECK_PARTIAL();
1711          RRETURN(MATCH_NOMATCH);
1712          }
1713      ecode++;      ecode++;
1714      break;      break;
1715    
1716      case OP_NOT_DIGIT:      case OP_NOT_DIGIT:
1717      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1718          {
1719          SCHECK_PARTIAL();
1720          RRETURN(MATCH_NOMATCH);
1721          }
1722      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1723      if (      if (
1724  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1453  for (;;) Line 1731  for (;;)
1731      break;      break;
1732    
1733      case OP_DIGIT:      case OP_DIGIT:
1734      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1735          {
1736          SCHECK_PARTIAL();
1737          RRETURN(MATCH_NOMATCH);
1738          }
1739      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1740      if (      if (
1741  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1466  for (;;) Line 1748  for (;;)
1748      break;      break;
1749    
1750      case OP_NOT_WHITESPACE:      case OP_NOT_WHITESPACE:
1751      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1752          {
1753          SCHECK_PARTIAL();
1754          RRETURN(MATCH_NOMATCH);
1755          }
1756      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1757      if (      if (
1758  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1479  for (;;) Line 1765  for (;;)
1765      break;      break;
1766    
1767      case OP_WHITESPACE:      case OP_WHITESPACE:
1768      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1769          {
1770          SCHECK_PARTIAL();
1771          RRETURN(MATCH_NOMATCH);
1772          }
1773      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1774      if (      if (
1775  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1492  for (;;) Line 1782  for (;;)
1782      break;      break;
1783    
1784      case OP_NOT_WORDCHAR:      case OP_NOT_WORDCHAR:
1785      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1786          {
1787          SCHECK_PARTIAL();
1788          RRETURN(MATCH_NOMATCH);
1789          }
1790      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1791      if (      if (
1792  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1505  for (;;) Line 1799  for (;;)
1799      break;      break;
1800    
1801      case OP_WORDCHAR:      case OP_WORDCHAR:
1802      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1803          {
1804          SCHECK_PARTIAL();
1805          RRETURN(MATCH_NOMATCH);
1806          }
1807      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1808      if (      if (
1809  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1518  for (;;) Line 1816  for (;;)
1816      break;      break;
1817    
1818      case OP_ANYNL:      case OP_ANYNL:
1819      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1820          {
1821          SCHECK_PARTIAL();
1822          RRETURN(MATCH_NOMATCH);
1823          }
1824      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1825      switch(c)      switch(c)
1826        {        {
# Line 1542  for (;;) Line 1844  for (;;)
1844      break;      break;
1845    
1846      case OP_NOT_HSPACE:      case OP_NOT_HSPACE:
1847      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1848          {
1849          SCHECK_PARTIAL();
1850          RRETURN(MATCH_NOMATCH);
1851          }
1852      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1853      switch(c)      switch(c)
1854        {        {
# Line 1572  for (;;) Line 1878  for (;;)
1878      break;      break;
1879    
1880      case OP_HSPACE:      case OP_HSPACE:
1881      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1882          {
1883          SCHECK_PARTIAL();
1884          RRETURN(MATCH_NOMATCH);
1885          }
1886      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1887      switch(c)      switch(c)
1888        {        {
# Line 1602  for (;;) Line 1912  for (;;)
1912      break;      break;
1913    
1914      case OP_NOT_VSPACE:      case OP_NOT_VSPACE:
1915      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1916          {
1917          SCHECK_PARTIAL();
1918          RRETURN(MATCH_NOMATCH);
1919          }
1920      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1921      switch(c)      switch(c)
1922        {        {
# Line 1620  for (;;) Line 1934  for (;;)
1934      break;      break;
1935    
1936      case OP_VSPACE:      case OP_VSPACE:
1937      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1938          {
1939          SCHECK_PARTIAL();
1940          RRETURN(MATCH_NOMATCH);
1941          }
1942      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1943      switch(c)      switch(c)
1944        {        {
# Line 1643  for (;;) Line 1961  for (;;)
1961    
1962      case OP_PROP:      case OP_PROP:
1963      case OP_NOTPROP:      case OP_NOTPROP:
1964      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1965          {
1966          SCHECK_PARTIAL();
1967          RRETURN(MATCH_NOMATCH);
1968          }
1969      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1970        {        {
1971        int chartype, script;        const ucd_record *prop = GET_UCD(c);
       int category = _pcre_ucp_findprop(c, &chartype, &script);  
1972    
1973        switch(ecode[1])        switch(ecode[1])
1974          {          {
# Line 1656  for (;;) Line 1977  for (;;)
1977          break;          break;
1978    
1979          case PT_LAMP:          case PT_LAMP:
1980          if ((chartype == ucp_Lu ||          if ((prop->chartype == ucp_Lu ||
1981               chartype == ucp_Ll ||               prop->chartype == ucp_Ll ||
1982               chartype == ucp_Lt) == (op == OP_NOTPROP))               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))
1983            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1984           break;           break;
1985    
1986          case PT_GC:          case PT_GC:
1987          if ((ecode[2] != category) == (op == OP_PROP))          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))
1988            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1989          break;          break;
1990    
1991          case PT_PC:          case PT_PC:
1992          if ((ecode[2] != chartype) == (op == OP_PROP))          if ((ecode[2] != prop->chartype) == (op == OP_PROP))
1993            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1994          break;          break;
1995    
1996          case PT_SC:          case PT_SC:
1997          if ((ecode[2] != script) == (op == OP_PROP))          if ((ecode[2] != prop->script) == (op == OP_PROP))
1998            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1999          break;          break;
2000    
# Line 1689  for (;;) Line 2010  for (;;)
2010      is in the binary; otherwise a compile-time error occurs. */      is in the binary; otherwise a compile-time error occurs. */
2011    
2012      case OP_EXTUNI:      case OP_EXTUNI:
2013      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2014          {
2015          SCHECK_PARTIAL();
2016          RRETURN(MATCH_NOMATCH);
2017          }
2018      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2019        {        {
2020        int chartype, script;        int category = UCD_CATEGORY(c);
       int category = _pcre_ucp_findprop(c, &chartype, &script);  
2021        if (category == ucp_M) RRETURN(MATCH_NOMATCH);        if (category == ucp_M) RRETURN(MATCH_NOMATCH);
2022        while (eptr < md->end_subject)        while (eptr < md->end_subject)
2023          {          {
# Line 1702  for (;;) Line 2026  for (;;)
2026            {            {
2027            GETCHARLEN(c, eptr, len);            GETCHARLEN(c, eptr, len);
2028            }            }
2029          category = _pcre_ucp_findprop(c, &chartype, &script);          category = UCD_CATEGORY(c);
2030          if (category != ucp_M) break;          if (category != ucp_M) break;
2031          eptr += len;          eptr += len;
2032          }          }
# Line 1723  for (;;) Line 2047  for (;;)
2047      case OP_REF:      case OP_REF:
2048        {        {
2049        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
2050        ecode += 3;                                 /* Advance past item */        ecode += 3;
2051    
2052          /* If the reference is unset, there are two possibilities:
2053    
2054          (a) In the default, Perl-compatible state, set the length to be longer
2055          than the amount of subject left; this ensures that every attempt at a
2056          match fails. We can't just fail here, because of the possibility of
2057          quantifiers with zero minima.
2058    
2059        /* If the reference is unset, set the length to be longer than the amount        (b) If the JavaScript compatibility flag is set, set the length to zero
2060        of subject left; this ensures that every attempt at a match fails. We        so that the back reference matches an empty string.
2061        can't just fail here, because of the possibility of quantifiers with zero  
2062        minima. */        Otherwise, set the length to the length of what was matched by the
2063          referenced subpattern. */
2064        length = (offset >= offset_top || md->offset_vector[offset] < 0)?  
2065          md->end_subject - eptr + 1 :        if (offset >= offset_top || md->offset_vector[offset] < 0)
2066          md->offset_vector[offset+1] - md->offset_vector[offset];          length = (md->jscript_compat)? 0 : md->end_subject - eptr + 1;
2067          else
2068            length = md->offset_vector[offset+1] - md->offset_vector[offset];
2069    
2070        /* Set up for repetition, or handle the non-repeated case */        /* Set up for repetition, or handle the non-repeated case */
2071    
# Line 1761  for (;;) Line 2094  for (;;)
2094          break;          break;
2095    
2096          default:               /* No repeat follows */          default:               /* No repeat follows */
2097          if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH);          if (!match_ref(offset, eptr, length, md, ims))
2098              {
2099              CHECK_PARTIAL();
2100              RRETURN(MATCH_NOMATCH);
2101              }
2102          eptr += length;          eptr += length;
2103          continue;              /* With the main loop */          continue;              /* With the main loop */
2104          }          }
# Line 1777  for (;;) Line 2114  for (;;)
2114    
2115        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2116          {          {
2117          if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH);          if (!match_ref(offset, eptr, length, md, ims))
2118              {
2119              CHECK_PARTIAL();
2120              RRETURN(MATCH_NOMATCH);
2121              }
2122          eptr += length;          eptr += length;
2123          }          }
2124    
# Line 1794  for (;;) Line 2135  for (;;)
2135            {            {
2136            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);
2137            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2138            if (fi >= max || !match_ref(offset, eptr, length, md, ims))            if (fi >= max) RRETURN(MATCH_NOMATCH);
2139              if (!match_ref(offset, eptr, length, md, ims))
2140                {
2141                CHECK_PARTIAL();
2142              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2143                }
2144            eptr += length;            eptr += length;
2145            }            }
2146          /* Control never gets here */          /* Control never gets here */
# Line 1808  for (;;) Line 2153  for (;;)
2153          pp = eptr;          pp = eptr;
2154          for (i = min; i < max; i++)          for (i = min; i < max; i++)
2155            {            {
2156            if (!match_ref(offset, eptr, length, md, ims)) break;            if (!match_ref(offset, eptr, length, md, ims))
2157                {
2158                CHECK_PARTIAL();
2159                break;
2160                }
2161            eptr += length;            eptr += length;
2162            }            }
2163          while (eptr >= pp)          while (eptr >= pp)
# Line 1822  for (;;) Line 2171  for (;;)
2171        }        }
2172      /* Control never gets here */      /* Control never gets here */
2173    
   
   
2174      /* Match a bit-mapped character class, possibly repeatedly. This op code is      /* Match a bit-mapped character class, possibly repeatedly. This op code is
2175      used when all the characters in the class have values in the range 0-255,      used when all the characters in the class have values in the range 0-255,
2176      and either the matching is caseful, or the characters are in the range      and either the matching is caseful, or the characters are in the range
# Line 1878  for (;;) Line 2225  for (;;)
2225          {          {
2226          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2227            {            {
2228            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
2229                {
2230                SCHECK_PARTIAL();
2231                RRETURN(MATCH_NOMATCH);
2232                }
2233            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
2234            if (c > 255)            if (c > 255)
2235              {              {
# Line 1896  for (;;) Line 2247  for (;;)
2247          {          {
2248          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2249            {            {
2250            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
2251                {
2252                SCHECK_PARTIAL();
2253                RRETURN(MATCH_NOMATCH);
2254                }
2255            c = *eptr++;            c = *eptr++;
2256            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2257            }            }
# Line 1920  for (;;) Line 2275  for (;;)
2275              {              {
2276              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);
2277              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2278              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2279                if (eptr >= md->end_subject)
2280                  {
2281                  SCHECK_PARTIAL();
2282                  RRETURN(MATCH_NOMATCH);
2283                  }
2284              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
2285              if (c > 255)              if (c > 255)
2286                {                {
# Line 1940  for (;;) Line 2300  for (;;)
2300              {              {
2301              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);
2302              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2303              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2304                if (eptr >= md->end_subject)
2305                  {
2306                  SCHECK_PARTIAL();
2307                  RRETURN(MATCH_NOMATCH);
2308                  }
2309              c = *eptr++;              c = *eptr++;
2310              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2311              }              }
# Line 1961  for (;;) Line 2326  for (;;)
2326            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2327              {              {
2328              int len = 1;              int len = 1;
2329              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
2330                  {
2331                  SCHECK_PARTIAL();
2332                  break;
2333                  }
2334              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
2335              if (c > 255)              if (c > 255)
2336                {                {
# Line 1987  for (;;) Line 2356  for (;;)
2356            {            {
2357            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2358              {              {
2359              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
2360                  {
2361                  SCHECK_PARTIAL();
2362                  break;
2363                  }
2364              c = *eptr;              c = *eptr;
2365              if ((data[c/8] & (1 << (c&7))) == 0) break;              if ((data[c/8] & (1 << (c&7))) == 0) break;
2366              eptr++;              eptr++;
# Line 2007  for (;;) Line 2380  for (;;)
2380    
2381    
2382      /* Match an extended character class. This opcode is encountered only      /* Match an extended character class. This opcode is encountered only
2383      in UTF-8 mode, because that's the only time it is compiled. */      when UTF-8 mode mode is supported. Nevertheless, we may not be in UTF-8
2384        mode, because Unicode properties are supported in non-UTF-8 mode. */
2385    
2386  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
2387      case OP_XCLASS:      case OP_XCLASS:
# Line 2048  for (;;) Line 2422  for (;;)
2422    
2423        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2424          {          {
2425          if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);          if (eptr >= md->end_subject)
2426          GETCHARINC(c, eptr);            {
2427              SCHECK_PARTIAL();
2428              RRETURN(MATCH_NOMATCH);
2429              }
2430            GETCHARINCTEST(c, eptr);
2431          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);
2432          }          }
2433    
# Line 2067  for (;;) Line 2445  for (;;)
2445            {            {
2446            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);
2447            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2448            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
2449            GETCHARINC(c, eptr);            if (eptr >= md->end_subject)
2450                {
2451                SCHECK_PARTIAL();
2452                RRETURN(MATCH_NOMATCH);
2453                }
2454              GETCHARINCTEST(c, eptr);
2455            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);
2456            }            }
2457          /* Control never gets here */          /* Control never gets here */
# Line 2082  for (;;) Line 2465  for (;;)
2465          for (i = min; i < max; i++)          for (i = min; i < max; i++)
2466            {            {
2467            int len = 1;            int len = 1;
2468            if (eptr >= md->end_subject) break;            if (eptr >= md->end_subject)
2469            GETCHARLEN(c, eptr, len);              {
2470                SCHECK_PARTIAL();
2471                break;
2472                }
2473              GETCHARLENTEST(c, eptr, len);
2474            if (!_pcre_xclass(c, data)) break;            if (!_pcre_xclass(c, data)) break;
2475            eptr += len;            eptr += len;
2476            }            }
# Line 2110  for (;;) Line 2497  for (;;)
2497        length = 1;        length = 1;
2498        ecode++;        ecode++;
2499        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
2500        if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);        if (length > md->end_subject - eptr)
2501            {
2502            CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
2503            RRETURN(MATCH_NOMATCH);
2504            }
2505        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);
2506        }        }
2507      else      else
# Line 2118  for (;;) Line 2509  for (;;)
2509    
2510      /* Non-UTF-8 mode */      /* Non-UTF-8 mode */
2511        {        {
2512        if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH);        if (md->end_subject - eptr < 1)
2513            {
2514            SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
2515            RRETURN(MATCH_NOMATCH);
2516            }
2517        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);
2518        ecode += 2;        ecode += 2;
2519        }        }
# Line 2134  for (;;) Line 2529  for (;;)
2529        ecode++;        ecode++;
2530        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
2531    
2532        if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);        if (length > md->end_subject - eptr)
2533            {
2534            CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
2535            RRETURN(MATCH_NOMATCH);
2536            }
2537    
2538        /* If the pattern character's value is < 128, we have only one byte, and        /* If the pattern character's value is < 128, we have only one byte, and
2539        can use the fast lookup table. */        can use the fast lookup table. */
# Line 2158  for (;;) Line 2557  for (;;)
2557          if (fc != dc)          if (fc != dc)
2558            {            {
2559  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2560            if (dc != _pcre_ucp_othercase(fc))            if (dc != UCD_OTHERCASE(fc))
2561  #endif  #endif
2562              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2563            }            }
# Line 2169  for (;;) Line 2568  for (;;)
2568    
2569      /* Non-UTF-8 mode */      /* Non-UTF-8 mode */
2570        {        {
2571        if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH);        if (md->end_subject - eptr < 1)
2572            {
2573            SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
2574            RRETURN(MATCH_NOMATCH);
2575            }
2576        if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);        if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2577        ecode += 2;        ecode += 2;
2578        }        }
# Line 2223  for (;;) Line 2626  for (;;)
2626      case OP_MINQUERY:      case OP_MINQUERY:
2627      c = *ecode++ - OP_STAR;      c = *ecode++ - OP_STAR;
2628      minimize = (c & 1) != 0;      minimize = (c & 1) != 0;
2629    
2630      min = rep_min[c];                 /* Pick up values from tables; */      min = rep_min[c];                 /* Pick up values from tables; */
2631      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
2632      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
2633    
2634      /* Common code for all repeated single-character matches. We can give      /* Common code for all repeated single-character matches. */
     up quickly if there are fewer than the minimum number of characters left in  
     the subject. */  
2635    
2636      REPEATCHAR:      REPEATCHAR:
2637  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 2238  for (;;) Line 2640  for (;;)
2640        length = 1;        length = 1;
2641        charptr = ecode;        charptr = ecode;
2642        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
       if (min * length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
2643        ecode += length;        ecode += length;
2644    
2645        /* Handle multibyte character matching specially here. There is        /* Handle multibyte character matching specially here. There is
# Line 2249  for (;;) Line 2650  for (;;)
2650  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2651          unsigned int othercase;          unsigned int othercase;
2652          if ((ims & PCRE_CASELESS) != 0 &&          if ((ims & PCRE_CASELESS) != 0 &&
2653              (othercase = _pcre_ucp_othercase(fc)) != NOTACHAR)              (othercase = UCD_OTHERCASE(fc)) != fc)
2654            oclength = _pcre_ord2utf8(othercase, occhars);            oclength = _pcre_ord2utf8(othercase, occhars);
2655          else oclength = 0;          else oclength = 0;
2656  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2657    
2658          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2659            {            {
2660            if (memcmp(eptr, charptr, length) == 0) eptr += length;            if (eptr <= md->end_subject - length &&
2661                memcmp(eptr, charptr, length) == 0) eptr += length;
2662  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2663            /* Need braces because of following else */            else if (oclength > 0 &&
2664            else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                     eptr <= md->end_subject - oclength &&
2665                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2666    #endif  /* SUPPORT_UCP */
2667            else            else
2668              {              {
2669              if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);              CHECK_PARTIAL();
2670              eptr += oclength;              RRETURN(MATCH_NOMATCH);
2671              }              }
 #else   /* without SUPPORT_UCP */  
           else { RRETURN(MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2672            }            }
2673    
2674          if (min == max) continue;          if (min == max) continue;
# Line 2278  for (;;) Line 2679  for (;;)
2679              {              {
2680              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);
2681              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2682              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2683              if (memcmp(eptr, charptr, length) == 0) eptr += length;              if (eptr <= md->end_subject - length &&
2684                  memcmp(eptr, charptr, length) == 0) eptr += length;
2685  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2686              /* Need braces because of following else */              else if (oclength > 0 &&
2687              else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                       eptr <= md->end_subject - oclength &&
2688                         memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2689    #endif  /* SUPPORT_UCP */
2690              else              else
2691                {                {
2692                if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);                CHECK_PARTIAL();
2693                eptr += oclength;                RRETURN(MATCH_NOMATCH);
2694                }                }
 #else   /* without SUPPORT_UCP */  
             else { RRETURN (MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2695              }              }
2696            /* Control never gets here */            /* Control never gets here */
2697            }            }
# Line 2300  for (;;) Line 2701  for (;;)
2701            pp = eptr;            pp = eptr;
2702            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2703              {              {
2704              if (eptr > md->end_subject - length) break;              if (eptr <= md->end_subject - length &&
2705              if (memcmp(eptr, charptr, length) == 0) eptr += length;                  memcmp(eptr, charptr, length) == 0) eptr += length;
2706  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2707              else if (oclength == 0) break;              else if (oclength > 0 &&
2708                         eptr <= md->end_subject - oclength &&
2709                         memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2710    #endif  /* SUPPORT_UCP */
2711              else              else
2712                {                {
2713                if (memcmp(eptr, occhars, oclength) != 0) break;                CHECK_PARTIAL();
2714                eptr += oclength;                break;
2715                }                }
 #else   /* without SUPPORT_UCP */  
             else break;  
 #endif  /* SUPPORT_UCP */  
2716              }              }
2717    
2718            if (possessive) continue;            if (possessive) continue;
2719    
2720            for(;;)            for(;;)
2721             {              {
2722             RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);
2723             if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2724             if (eptr == pp) RRETURN(MATCH_NOMATCH);              if (eptr == pp) { RRETURN(MATCH_NOMATCH); }
2725  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2726             eptr--;              eptr--;
2727             BACKCHAR(eptr);              BACKCHAR(eptr);
2728  #else   /* without SUPPORT_UCP */  #else   /* without SUPPORT_UCP */
2729             eptr -= length;              eptr -= length;
2730  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2731             }              }
2732            }            }
2733          /* Control never gets here */          /* Control never gets here */
2734          }          }
# Line 2339  for (;;) Line 2741  for (;;)
2741  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF8 */
2742    
2743      /* When not in UTF-8 mode, load a single-byte character. */      /* When not in UTF-8 mode, load a single-byte character. */
2744        {  
2745        if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);      fc = *ecode++;
       fc = *ecode++;  
       }  
2746    
2747      /* The value of fc at this point is always less than 256, though we may or      /* The value of fc at this point is always less than 256, though we may or
2748      may not be in UTF-8 mode. The code is duplicated for the caseless and      may not be in UTF-8 mode. The code is duplicated for the caseless and
# Line 2360  for (;;) Line 2760  for (;;)
2760        {        {
2761        fc = md->lcc[fc];        fc = md->lcc[fc];
2762        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2763          if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);          {
2764        if (min == max) continue;          if (eptr >= md->end_subject)
2765              {
2766              SCHECK_PARTIAL();
2767              RRETURN(MATCH_NOMATCH);
2768              }
2769            if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2770            }
2771          if (min == max) continue;
2772        if (minimize)        if (minimize)
2773          {          {
2774          for (fi = min;; fi++)          for (fi = min;; fi++)
2775            {            {
2776            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);
2777            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2778            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) RRETURN(MATCH_NOMATCH);
2779                fc != md->lcc[*eptr++])            if (eptr >= md->end_subject)
2780                {
2781                SCHECK_PARTIAL();
2782              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2783                }
2784              if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2785            }            }
2786          /* Control never gets here */          /* Control never gets here */
2787          }          }
# Line 2379  for (;;) Line 2790  for (;;)
2790          pp = eptr;          pp = eptr;
2791          for (i = min; i < max; i++)          for (i = min; i < max; i++)
2792            {            {
2793            if (eptr >= md->end_subject || fc != md->lcc[*eptr]) break;            if (eptr >= md->end_subject)
2794                {
2795                SCHECK_PARTIAL();
2796                break;
2797                }
2798              if (fc != md->lcc[*eptr]) break;
2799            eptr++;            eptr++;
2800            }            }
2801    
2802          if (possessive) continue;          if (possessive) continue;
2803    
2804          while (eptr >= pp)          while (eptr >= pp)
2805            {            {
2806            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);
# Line 2398  for (;;) Line 2816  for (;;)
2816    
2817      else      else
2818        {        {
2819        for (i = 1; i <= min; i++) if (fc != *eptr++) RRETURN(MATCH_NOMATCH);        for (i = 1; i <= min; i++)
2820            {
2821            if (eptr >= md->end_subject)
2822              {
2823              SCHECK_PARTIAL();
2824              RRETURN(MATCH_NOMATCH);
2825              }
2826            if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
2827            }
2828    
2829        if (min == max) continue;        if (min == max) continue;
2830    
2831        if (minimize)        if (minimize)
2832          {          {
2833          for (fi = min;; fi++)          for (fi = min;; fi++)
2834            {            {
2835            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);
2836            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2837            if (fi >= max || eptr >= md->end_subject || fc != *eptr++)            if (fi >= max) RRETURN(MATCH_NOMATCH);
2838              if (eptr >= md->end_subject)
2839                {
2840                SCHECK_PARTIAL();
2841              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2842                }
2843              if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
2844            }            }
2845          /* Control never gets here */          /* Control never gets here */
2846          }          }
# Line 2416  for (;;) Line 2849  for (;;)
2849          pp = eptr;          pp = eptr;
2850          for (i = min; i < max; i++)          for (i = min; i < max; i++)
2851            {            {
2852            if (eptr >= md->end_subject || fc != *eptr) break;            if (eptr >= md->end_subject)
2853                {
2854                SCHECK_PARTIAL();
2855                break;
2856                }
2857              if (fc != *eptr) break;
2858            eptr++;            eptr++;
2859            }            }
2860          if (possessive) continue;          if (possessive) continue;
2861    
2862          while (eptr >= pp)          while (eptr >= pp)
2863            {            {
2864            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27);
# Line 2435  for (;;) Line 2874  for (;;)
2874      checking can be multibyte. */      checking can be multibyte. */
2875    
2876      case OP_NOT:      case OP_NOT:
2877      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2878          {
2879          SCHECK_PARTIAL();
2880          RRETURN(MATCH_NOMATCH);
2881          }
2882      ecode++;      ecode++;
2883      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2884      if ((ims & PCRE_CASELESS) != 0)      if ((ims & PCRE_CASELESS) != 0)
# Line 2512  for (;;) Line 2955  for (;;)
2955      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
2956      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
2957    
2958      /* Common code for all repeated single-byte matches. We can give up quickly      /* Common code for all repeated single-byte matches. */
     if there are fewer than the minimum number of bytes left in the  
     subject. */  
2959    
2960      REPEATNOTCHAR:      REPEATNOTCHAR:
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
2961      fc = *ecode++;      fc = *ecode++;
2962    
2963      /* The code is duplicated for the caseless and caseful cases, for speed,      /* The code is duplicated for the caseless and caseful cases, for speed,
# Line 2542  for (;;) Line 2982  for (;;)
2982          register unsigned int d;          register unsigned int d;
2983          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2984            {            {
2985              if (eptr >= md->end_subject)
2986                {
2987                SCHECK_PARTIAL();
2988                RRETURN(MATCH_NOMATCH);
2989                }
2990            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
2991            if (d < 256) d = md->lcc[d];            if (d < 256) d = md->lcc[d];
2992            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
# Line 2553  for (;;) Line 2998  for (;;)
2998        /* Not UTF-8 mode */        /* Not UTF-8 mode */
2999          {          {
3000          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3001              {
3002              if (eptr >= md->end_subject)
3003                {
3004                SCHECK_PARTIAL();
3005                RRETURN(MATCH_NOMATCH);
3006                }
3007            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
3008              }
3009          }          }
3010    
3011        if (min == max) continue;        if (min == max) continue;
# Line 2569  for (;;) Line 3021  for (;;)
3021              {              {
3022              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);
3023              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3024                if (fi >= max) RRETURN(MATCH_NOMATCH);
3025                if (eptr >= md->end_subject)
3026                  {
3027                  SCHECK_PARTIAL();
3028                  RRETURN(MATCH_NOMATCH);
3029                  }
3030              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3031              if (d < 256) d = md->lcc[d];              if (d < 256) d = md->lcc[d];
3032              if (fi >= max || eptr >= md->end_subject || fc == d)              if (fc == d) RRETURN(MATCH_NOMATCH);
               RRETURN(MATCH_NOMATCH);  
3033              }              }
3034            }            }
3035          else          else
# Line 2583  for (;;) Line 3040  for (;;)
3040              {              {
3041              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);
3042              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3043              if (fi >= max || eptr >= md->end_subject || fc == md->lcc[*eptr++])              if (fi >= max) RRETURN(MATCH_NOMATCH);
3044                if (eptr >= md->end_subject)
3045                  {
3046                  SCHECK_PARTIAL();
3047                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3048                  }
3049                if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
3050              }              }
3051            }            }
3052          /* Control never gets here */          /* Control never gets here */
# Line 2604  for (;;) Line 3066  for (;;)
3066            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3067              {              {
3068              int len = 1;              int len = 1;
3069              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
3070                  {
3071                  SCHECK_PARTIAL();
3072                  break;
3073                  }
3074              GETCHARLEN(d, eptr, len);              GETCHARLEN(d, eptr, len);
3075              if (d < 256) d = md->lcc[d];              if (d < 256) d = md->lcc[d];
3076              if (fc == d) break;              if (fc == d) break;
# Line 2625  for (;;) Line 3091  for (;;)
3091            {            {
3092            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3093              {              {
3094              if (eptr >= md->end_subject || fc == md->lcc[*eptr]) break;              if (eptr >= md->end_subject)
3095                  {
3096                  SCHECK_PARTIAL();
3097                  break;
3098                  }
3099                if (fc == md->lcc[*eptr]) break;
3100              eptr++;              eptr++;
3101              }              }
3102            if (possessive) continue;            if (possessive) continue;
# Line 2653  for (;;) Line 3124  for (;;)
3124          register unsigned int d;          register unsigned int d;
3125          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3126            {            {
3127              if (eptr >= md->end_subject)
3128                {
3129                SCHECK_PARTIAL();
3130                RRETURN(MATCH_NOMATCH);
3131                }
3132            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3133            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
3134            }            }
# Line 2662  for (;;) Line 3138  for (;;)
3138        /* Not UTF-8 mode */        /* Not UTF-8 mode */
3139          {          {
3140          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3141              {
3142              if (eptr >= md->end_subject)
3143                {
3144                SCHECK_PARTIAL();
3145                RRETURN(MATCH_NOMATCH);
3146                }
3147            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
3148              }
3149          }          }
3150    
3151        if (min == max) continue;        if (min == max) continue;
# Line 2678  for (;;) Line 3161  for (;;)
3161              {              {
3162              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);
3163              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3164              GETCHARINC(d, eptr);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3165              if (fi >= max || eptr >= md->end_subject || fc == d)              if (eptr >= md->end_subject)
3166                  {
3167                  SCHECK_PARTIAL();
3168                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3169                  }
3170                GETCHARINC(d, eptr);
3171                if (fc == d) RRETURN(MATCH_NOMATCH);
3172              }              }
3173            }            }
3174          else          else
# Line 2691  for (;;) Line 3179  for (;;)
3179              {              {
3180              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);
3181              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3182              if (fi >= max || eptr >= md->end_subject || fc == *eptr++)              if (fi >= max) RRETURN(MATCH_NOMATCH);
3183                if (eptr >= md->end_subject)
3184                  {
3185                  SCHECK_PARTIAL();
3186                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3187                  }
3188                if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
3189              }              }
3190            }            }
3191          /* Control never gets here */          /* Control never gets here */
# Line 2712  for (;;) Line 3205  for (;;)
3205            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3206              {              {
3207              int len = 1;              int len = 1;
3208              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
3209                  {
3210                  SCHECK_PARTIAL();
3211                  break;
3212                  }
3213              GETCHARLEN(d, eptr, len);              GETCHARLEN(d, eptr, len);
3214              if (fc == d) break;              if (fc == d) break;
3215              eptr += len;              eptr += len;
# Line 2732  for (;;) Line 3229  for (;;)
3229            {            {
3230            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3231              {              {
3232              if (eptr >= md->end_subject || fc == *eptr) break;              if (eptr >= md->end_subject)
3233                  {
3234                  SCHECK_PARTIAL();
3235                  break;
3236                  }
3237                if (fc == *eptr) break;
3238              eptr++;              eptr++;
3239              }              }
3240            if (possessive) continue;            if (possessive) continue;
# Line 2826  for (;;) Line 3328  for (;;)
3328    
3329      /* First, ensure the minimum number of matches are present. Use inline      /* First, ensure the minimum number of matches are present. Use inline
3330      code for maximizing the speed, and do the type test once at the start      code for maximizing the speed, and do the type test once at the start
3331      (i.e. keep it out of the loop). Also we can test that there are at least      (i.e. keep it out of the loop). Separate the UTF-8 code completely as that
     the minimum number of bytes before we start. This isn't as effective in  
     UTF-8 mode, but it does no harm. Separate the UTF-8 code completely as that  
3332      is tidier. Also separate the UCP code, which can be the same for both UTF-8      is tidier. Also separate the UCP code, which can be the same for both UTF-8
3333      and single-bytes. */      and single-bytes. */
3334    
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
3335      if (min > 0)      if (min > 0)
3336        {        {
3337  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 2844  for (;;) Line 3343  for (;;)
3343            if (prop_fail_result) RRETURN(MATCH_NOMATCH);            if (prop_fail_result) RRETURN(MATCH_NOMATCH);
3344            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3345              {              {
3346              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3347                  {
3348                  SCHECK_PARTIAL();
3349                  RRETURN(MATCH_NOMATCH);
3350                  }
3351              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3352              }              }
3353            break;            break;
# Line 2852  for (;;) Line 3355  for (;;)
3355            case PT_LAMP:            case PT_LAMP:
3356            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3357              {              {
3358              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3359                  {
3360                  SCHECK_PARTIAL();
3361                  RRETURN(MATCH_NOMATCH);
3362                  }
3363              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3364              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
3365              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
3366                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
3367                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 2865  for (;;) Line 3372  for (;;)
3372            case PT_GC:            case PT_GC:
3373            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3374              {              {
3375              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3376                  {
3377                  SCHECK_PARTIAL();
3378                  RRETURN(MATCH_NOMATCH);
3379                  }
3380              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3381              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
3382              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
3383                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3384              }              }
# Line 2876  for (;;) Line 3387  for (;;)
3387            case PT_PC:            case PT_PC:
3388            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3389              {              {
3390              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3391                  {
3392                  SCHECK_PARTIAL();
3393                  RRETURN(MATCH_NOMATCH);
3394                  }
3395              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3396              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
3397              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
3398                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3399              }              }
# Line 2887  for (;;) Line 3402  for (;;)
3402            case PT_SC:            case PT_SC:
3403            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3404              {              {
3405              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3406                  {
3407                  SCHECK_PARTIAL();
3408                  RRETURN(MATCH_NOMATCH);
3409                  }
3410              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
3411              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_script = UCD_SCRIPT(c);
3412              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
3413                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3414              }              }
# Line 2907  for (;;) Line 3426  for (;;)
3426          {          {
3427          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3428            {            {
3429              if (eptr >= md->end_subject)
3430                {
3431                SCHECK_PARTIAL();
3432                RRETURN(MATCH_NOMATCH);
3433                }
3434            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3435            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
3436            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);
3437            while (eptr < md->end_subject)            while (eptr < md->end_subject)
3438              {              {
3439              int len = 1;              int len = 1;
3440              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
3441                {                else { GETCHARLEN(c, eptr, len); }
3442                GETCHARLEN(c, eptr, len);              prop_category = UCD_CATEGORY(c);
               }  
             prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);  
3443              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
3444              eptr += len;              eptr += len;
3445              }              }
# Line 2935  for (;;) Line 3457  for (;;)
3457          case OP_ANY:          case OP_ANY:
3458          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3459            {            {
3460            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3461                 ((ims & PCRE_DOTALL) == 0 && IS_NEWLINE(eptr)))              {
3462                SCHECK_PARTIAL();
3463                RRETURN(MATCH_NOMATCH);
3464                }
3465              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
3466              eptr++;
3467              while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3468              }
3469            break;
3470    
3471            case OP_ALLANY:
3472            for (i = 1; i <= min; i++)
3473              {
3474              if (eptr >= md->end_subject)
3475                {
3476                SCHECK_PARTIAL();
3477              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3478                }
3479            eptr++;            eptr++;
3480            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3481            }            }
3482          break;          break;
3483    
3484          case OP_ANYBYTE:          case OP_ANYBYTE:
3485            if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
3486          eptr += min;          eptr += min;
3487          break;          break;
3488    
3489          case OP_ANYNL:          case OP_ANYNL:
3490          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3491            {            {
3492            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3493                {
3494                SCHECK_PARTIAL();
3495                RRETURN(MATCH_NOMATCH);
3496                }
3497            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3498            switch(c)            switch(c)
3499              {              {
# Line 2976  for (;;) Line 3519  for (;;)
3519          case OP_NOT_HSPACE:          case OP_NOT_HSPACE:
3520          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3521            {            {
3522            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3523                {
3524                SCHECK_PARTIAL();
3525                RRETURN(MATCH_NOMATCH);
3526                }
3527            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3528            switch(c)            switch(c)
3529              {              {
# Line 3008  for (;;) Line 3555  for (;;)
3555          case OP_HSPACE:          case OP_HSPACE:
3556          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3557            {            {
3558            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3559                {
3560                SCHECK_PARTIAL();
3561                RRETURN(MATCH_NOMATCH);
3562                }
3563            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3564            switch(c)            switch(c)
3565              {              {
# Line 3040  for (;;) Line 3591  for (;;)
3591          case OP_NOT_VSPACE:          case OP_NOT_VSPACE:
3592          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3593            {            {
3594            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3595                {
3596                SCHECK_PARTIAL();
3597                RRETURN(MATCH_NOMATCH);
3598                }
3599            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3600            switch(c)            switch(c)
3601              {              {
# Line 3060  for (;;) Line 3615  for (;;)
3615          case OP_VSPACE:          case OP_VSPACE:
3616          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3617            {            {
3618            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3619                {
3620                SCHECK_PARTIAL();
3621                RRETURN(MATCH_NOMATCH);
3622                }
3623            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3624            switch(c)            switch(c)
3625              {              {
# Line 3080  for (;;) Line 3639  for (;;)
3639          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
3640          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3641            {            {
3642            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3643                {
3644                SCHECK_PARTIAL();
3645                RRETURN(MATCH_NOMATCH);
3646                }
3647            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3648            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)
3649              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
# Line 3090  for (;;) Line 3653  for (;;)
3653          case OP_DIGIT:          case OP_DIGIT:
3654          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3655            {            {
3656            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3657               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)              {
3658                SCHECK_PARTIAL();
3659                RRETURN(MATCH_NOMATCH);
3660                }
3661              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)
3662              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3663            /* No need to skip more bytes - we know it's a 1-byte character */            /* No need to skip more bytes - we know it's a 1-byte character */
3664            }            }
# Line 3100  for (;;) Line 3667  for (;;)
3667          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
3668          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3669            {            {
3670            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3671               (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0))              {
3672                SCHECK_PARTIAL();
3673                RRETURN(MATCH_NOMATCH);
3674                }
3675              if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)
3676              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3677            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3678            }            }
# Line 3110  for (;;) Line 3681  for (;;)
3681          case OP_WHITESPACE:          case OP_WHITESPACE:
3682          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3683            {            {
3684            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3685               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)              {
3686                SCHECK_PARTIAL();
3687                RRETURN(MATCH_NOMATCH);
3688                }
3689              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)
3690              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3691            /* No need to skip more bytes - we know it's a 1-byte character */            /* No need to skip more bytes - we know it's a 1-byte character */
3692            }            }
# Line 3130  for (;;) Line 3705  for (;;)
3705          case OP_WORDCHAR:          case OP_WORDCHAR:
3706          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3707            {            {
3708            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3709               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)              {
3710                SCHECK_PARTIAL();
3711                RRETURN(MATCH_NOMATCH);
3712                }
3713              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)
3714              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3715            /* No need to skip more bytes - we know it's a 1-byte character */            /* No need to skip more bytes - we know it's a 1-byte character */
3716            }            }
# Line 3145  for (;;) Line 3724  for (;;)
3724  #endif     /* SUPPORT_UTF8 */  #endif     /* SUPPORT_UTF8 */
3725    
3726        /* Code for the non-UTF-8 case for minimum matching of operators other        /* Code for the non-UTF-8 case for minimum matching of operators other
3727        than OP_PROP and OP_NOTPROP. We can assume that there are the minimum        than OP_PROP and OP_NOTPROP. */
       number of bytes present, as this was tested above. */  
3728    
3729        switch(ctype)        switch(ctype)
3730          {          {
3731          case OP_ANY:          case OP_ANY:
3732          if ((ims & PCRE_DOTALL) == 0)          for (i = 1; i <= min; i++)
3733            {            {
3734            for (i = 1; i <= min; i++)            if (eptr >= md->end_subject)
3735              {              {
3736              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
3737              eptr++;              RRETURN(MATCH_NOMATCH);
3738              }              }
3739              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
3740              eptr++;
3741            }            }
         else eptr += min;  
3742          break;          break;
3743    
3744          case OP_ANYBYTE:          case OP_ALLANY:
3745            if (eptr > md->end_subject - min)
3746              {
3747              SCHECK_PARTIAL();
3748              RRETURN(MATCH_NOMATCH);
3749              }
3750          eptr += min;          eptr += min;
3751          break;          break;
3752    
3753          /* Because of the CRLF case, we can't assume the minimum number of          case OP_ANYBYTE:
3754          bytes are present in this case. */          if (eptr > md->end_subject - min)
3755              {
3756              SCHECK_PARTIAL();
3757              RRETURN(MATCH_NOMATCH);
3758              }
3759            eptr += min;
3760            break;
3761    
3762          case OP_ANYNL:          case OP_ANYNL:
3763          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3764            {            {
3765            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3766                {
3767                SCHECK_PARTIAL();
3768                RRETURN(MATCH_NOMATCH);
3769                }
3770            switch(*eptr++)            switch(*eptr++)
3771              {              {
3772              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
# Line 3194  for (;;) Line 3788  for (;;)
3788          case OP_NOT_HSPACE:          case OP_NOT_HSPACE:
3789          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3790            {            {
3791            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3792                {
3793                SCHECK_PARTIAL();
3794                RRETURN(MATCH_NOMATCH);
3795                }
3796            switch(*eptr++)            switch(*eptr++)
3797              {              {
3798              default: break;              default: break;
# Line 3209  for (;;) Line 3807  for (;;)
3807          case OP_HSPACE:          case OP_HSPACE:
3808          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3809            {            {
3810            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3811                {
3812                SCHECK_PARTIAL();
3813                RRETURN(MATCH_NOMATCH);
3814                }
3815            switch(*eptr++)            switch(*eptr++)
3816              {              {
3817              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
# Line 3224  for (;;) Line 3826  for (;;)
3826          case OP_NOT_VSPACE:          case OP_NOT_VSPACE:
3827          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3828            {            {
3829            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3830                {
3831                SCHECK_PARTIAL();
3832                RRETURN(MATCH_NOMATCH);
3833                }
3834            switch(*eptr++)            switch(*eptr++)
3835              {              {
3836              default: break;              default: break;
# Line 3241  for (;;) Line 3847  for (;;)
3847          case OP_VSPACE:          case OP_VSPACE:
3848          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3849            {            {
3850            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3851                {
3852                SCHECK_PARTIAL();
3853                RRETURN(MATCH_NOMATCH);
3854                }
3855            switch(*eptr++)            switch(*eptr++)
3856              {              {
3857              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
# Line 3257  for (;;) Line 3867  for (;;)
3867    
3868          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
3869          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3870              {
3871              if (eptr >= md->end_subject)
3872                {
3873                SCHECK_PARTIAL();
3874                RRETURN(MATCH_NOMATCH);
3875                }
3876            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);
3877              }
3878          break;          break;
3879    
3880          case OP_DIGIT:          case OP_DIGIT:
3881          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3882              {
3883              if (eptr >= md->end_subject)
3884                {
3885                SCHECK_PARTIAL();
3886                RRETURN(MATCH_NOMATCH);
3887                }
3888            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);
3889              }
3890          break;          break;
3891    
3892          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
3893          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3894              {
3895              if (eptr >= md->end_subject)
3896                {
3897                SCHECK_PARTIAL();
3898                RRETURN(MATCH_NOMATCH);
3899                }
3900            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);
3901              }
3902          break;          break;
3903    
3904          case OP_WHITESPACE:          case OP_WHITESPACE:
3905          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3906              {
3907              if (eptr >= md->end_subject)
3908                {
3909                SCHECK_PARTIAL();
3910                RRETURN(MATCH_NOMATCH);
3911                }
3912            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);
3913              }
3914          break;          break;
3915    
3916          case OP_NOT_WORDCHAR:          case OP_NOT_WORDCHAR:
3917          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3918              {
3919              if (eptr >= md->end_subject)
3920                {
3921                SCHECK_PARTIAL();
3922                RRETURN(MATCH_NOMATCH);
3923                }
3924            if ((md->ctypes[*eptr++] & ctype_word) != 0)            if ((md->ctypes[*eptr++] & ctype_word) != 0)
3925              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3926              }
3927          break;          break;
3928    
3929          case OP_WORDCHAR:          case OP_WORDCHAR:
3930          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3931              {
3932              if (eptr >= md->end_subject)
3933                {
3934                SCHECK_PARTIAL();
3935                RRETURN(MATCH_NOMATCH);
3936                }
3937            if ((md->ctypes[*eptr++] & ctype_word) == 0)            if ((md->ctypes[*eptr++] & ctype_word) == 0)
3938              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3939              }
3940          break;          break;
3941    
3942          default:          default:
# Line 3312  for (;;) Line 3964  for (;;)
3964              {              {
3965              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);
3966              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3967              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3968                if (eptr >= md->end_subject)
3969                  {
3970                  SCHECK_PARTIAL();
3971                  RRETURN(MATCH_NOMATCH);
3972                  }
3973              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3974              if (prop_fail_result) RRETURN(MATCH_NOMATCH);              if (prop_fail_result) RRETURN(MATCH_NOMATCH);
3975              }              }
# Line 3323  for (;;) Line 3980  for (;;)
3980              {              {
3981              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);
3982              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3983              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3984                if (eptr >= md->end_subject)
3985                  {
3986                  SCHECK_PARTIAL();
3987                  RRETURN(MATCH_NOMATCH);
3988                  }
3989              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3990              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
3991              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
3992                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
3993                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 3338  for (;;) Line 4000  for (;;)
4000              {              {
4001              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);
4002              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4003              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4004                if (eptr >= md->end_subject)
4005                  {
4006                  SCHECK_PARTIAL();
4007                  RRETURN(MATCH_NOMATCH);
4008                  }
4009              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
4010              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4011              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
4012                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4013              }              }
# Line 3351  for (;;) Line 4018  for (;;)
4018              {              {
4019              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);
4020              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4021              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4022                if (eptr >= md->end_subject)
4023                  {
4024                  SCHECK_PARTIAL();
4025                  RRETURN(MATCH_NOMATCH);
4026                  }
4027              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
4028              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
4029              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
4030                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4031              }              }
# Line 3364  for (;;) Line 4036  for (;;)
4036              {              {
4037              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);
4038              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4039              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
4040                if (eptr >= md->end_subject)
4041                  {
4042                  SCHECK_PARTIAL();
4043                  RRETURN(MATCH_NOMATCH);
4044                  }
4045              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
4046              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_script = UCD_SCRIPT(c);
4047              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
4048                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4049              }              }
# Line 3386  for (;;) Line 4063  for (;;)
4063            {            {
4064            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);
4065            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4066            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
4067              if (eptr >= md->end_subject)
4068                {
4069                SCHECK_PARTIAL();
4070                RRETURN(MATCH_NOMATCH);
4071                }
4072            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
4073            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
4074            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);
4075            while (eptr < md->end_subject)            while (eptr < md->end_subject)
4076              {              {
4077              int len = 1;              int len = 1;
4078              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
4079                {                else { GETCHARLEN(c, eptr, len); }
4080                GETCHARLEN(c, eptr, len);              prop_category = UCD_CATEGORY(c);
               }  
             prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);  
4081              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
4082              eptr += len;              eptr += len;
4083              }              }
# Line 3415  for (;;) Line 4095  for (;;)
4095            {            {
4096            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);
4097            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4098            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) RRETURN(MATCH_NOMATCH);
4099                 (ctype == OP_ANY && (ims & PCRE_DOTALL) == 0 &&            if (eptr >= md->end_subject)
4100                  IS_NEWLINE(eptr)))              {
4101                SCHECK_PARTIAL();
4102                RRETURN(MATCH_NOMATCH);
4103                }
4104              if (ctype == OP_ANY && IS_NEWLINE(eptr))
4105              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
   
4106            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4107            switch(ctype)            switch(ctype)
4108              {              {
4109              case OP_ANY:        /* This is the DOTALL case */              case OP_ANY:        /* This is the non-NL case */
4110              break;              case OP_ALLANY:
   
4111              case OP_ANYBYTE:              case OP_ANYBYTE:
4112              break;              break;
4113    
# Line 3576  for (;;) Line 4258  for (;;)
4258            {            {
4259            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);
4260            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4261            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) RRETURN(MATCH_NOMATCH);
4262                 ((ims & PCRE_DOTALL) == 0 && IS_NEWLINE(eptr)))            if (eptr >= md->end_subject)
4263                {
4264                SCHECK_PARTIAL();
4265                RRETURN(MATCH_NOMATCH);
4266                }
4267              if (ctype == OP_ANY && IS_NEWLINE(eptr))
4268              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
   
4269            c = *eptr++;            c = *eptr++;
4270            switch(ctype)            switch(ctype)
4271              {              {
4272              case OP_ANY:   /* This is the DOTALL case */              case OP_ANY:     /* This is the non-NL case */
4273              break;              case OP_ALLANY:
   
4274              case OP_ANYBYTE:              case OP_ANYBYTE:
4275              break;              break;
4276    
# Line 3705  for (;;) Line 4390  for (;;)
4390            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4391              {              {
4392              int len = 1;              int len = 1;
4393              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4394                  {
4395                  SCHECK_PARTIAL();
4396                  break;
4397                  }
4398              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4399              if (prop_fail_result) break;              if (prop_fail_result) break;
4400              eptr+= len;              eptr+= len;
# Line 3716  for (;;) Line 4405  for (;;)
4405            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4406              {              {
4407              int len = 1;              int len = 1;
4408              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4409                  {
4410                  SCHECK_PARTIAL();
4411                  break;
4412                  }
4413              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4414              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
4415              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
4416                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
4417                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 3731  for (;;) Line 4424  for (;;)
4424            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4425              {              {
4426              int len = 1;              int len = 1;
4427              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4428                  {
4429                  SCHECK_PARTIAL();
4430                  break;
4431                  }
4432              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4433              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4434              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
4435                break;                break;
4436              eptr+= len;              eptr+= len;
# Line 3744  for (;;) Line 4441  for (;;)
4441            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4442              {              {
4443              int len = 1;              int len = 1;
4444              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4445                  {
4446                  SCHECK_PARTIAL();
4447                  break;
4448                  }
4449              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4450              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
4451              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
4452                break;                break;
4453              eptr+= len;              eptr+= len;
# Line 3757  for (;;) Line 4458  for (;;)
4458            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4459              {              {
4460              int len = 1;              int len = 1;
4461              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4462                  {
4463                  SCHECK_PARTIAL();
4464                  break;
4465                  }
4466              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4467              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_script = UCD_SCRIPT(c);
4468              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
4469                break;                break;
4470              eptr+= len;              eptr+= len;
# Line 3786  for (;;) Line 4491  for (;;)
4491          {          {
4492          for (i = min; i < max; i++)          for (i = min; i < max; i++)
4493            {            {
4494            if (eptr >= md->end_subject) break;            if (eptr >= md->end_subject)
4495                {
4496                SCHECK_PARTIAL();
4497                break;
4498                }
4499            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
4500            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
4501            if (prop_category == ucp_M) break;            if (prop_category == ucp_M) break;
4502            while (eptr < md->end_subject)            while (eptr < md->end_subject)
4503              {              {
# Line 3797  for (;;) Line 4506  for (;;)
4506                {                {
4507                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
4508                }                }
4509              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4510              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
4511              eptr += len;              eptr += len;
4512              }              }
# Line 3806  for (;;) Line 4515  for (;;)
4515          /* eptr is now past the end of the maximum run */          /* eptr is now past the end of the maximum run */
4516    
4517          if (possessive) continue;          if (possessive) continue;
4518    
4519          for(;;)          for(;;)
4520            {            {
4521            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM45);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM45);
# Line 3819  for (;;) Line 4529  for (;;)
4529                BACKCHAR(eptr);                BACKCHAR(eptr);
4530                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
4531                }                }
4532              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4533              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
4534              eptr--;              eptr--;
4535              }              }
# Line 3839  for (;;) Line 4549  for (;;)
4549            case OP_ANY:            case OP_ANY:
4550            if (max < INT_MAX)            if (max < INT_MAX)
4551              {              {
4552              if ((ims & PCRE_DOTALL) == 0)              for (i = min; i < max; i++)
               {  
               for (i = min; i < max; i++)  
                 {  
                 if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;  
                 eptr++;  
                 while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  
                 }  
               }  
             else  
4553                {                {
4554                for (i = min; i < max; i++)                if (eptr >= md->end_subject)
4555                  {                  {
4556                  if (eptr >= md->end_subject) break;                  SCHECK_PARTIAL();
4557                  eptr++;                  break;
                 while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  
4558                  }                  }
4559                  if (IS_NEWLINE(eptr)) break;
4560                  eptr++;
4561                  while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
4562                }                }
4563              }              }
4564    
# Line 3863  for (;;) Line 4566  for (;;)
4566    
4567            else            else
4568              {              {
4569              if ((ims & PCRE_DOTALL) == 0)              for (i = min; i < max; i++)
4570                {                {
4571                for (i = min; i < max; i++)                if (eptr >= md->end_subject)
4572                  {                  {
4573                  if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;                  SCHECK_PARTIAL();
4574                  eptr++;                  break;
                 while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  
4575                  }                  }
4576                  if (IS_NEWLINE(eptr)) break;
4577                  eptr++;
4578                  while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
4579                }                }
4580              else              }
4581              break;
4582    
4583              case OP_ALLANY:
4584              if (max < INT_MAX)
4585                {
4586                for (i = min; i < max; i++)
4587                {                {
4588                eptr = md->end_subject;                if (eptr >= md->end_subject)
4589                    {
4590                    SCHECK_PARTIAL();
4591                    break;
4592                    }
4593                  eptr++;
4594                  while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
4595                }                }
4596              }              }
4597              else eptr = md->end_subject;   /* Unlimited UTF-8 repeat */
4598            break;            break;
4599    
4600            /* The byte case is the same as non-UTF8 */            /* The byte case is the same as non-UTF8 */
# Line 3884  for (;;) Line 4602  for (;;)
4602            case OP_ANYBYTE:            case OP_ANYBYTE:
4603            c = max - min;            c = max - min;
4604            if (c > (unsigned int)(md->end_subject - eptr))            if (c > (unsigned int)(md->end_subject - eptr))
4605              c = md->end_subject - eptr;              {
4606            eptr += c;              eptr = md->end_subject;
4607                SCHECK_PARTIAL();
4608                }
4609              else eptr += c;
4610            break;            break;
4611    
4612            case OP_ANYNL:            case OP_ANYNL:
4613            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4614              {              {
4615              int len = 1;              int len = 1;
4616              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4617                  {
4618                  SCHECK_PARTIAL();
4619                  break;
4620                  }
4621              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4622              if (c == 0x000d)              if (c == 0x000d)
4623                {                {
# Line 3917  for (;;) Line 4642  for (;;)
4642              {              {
4643              BOOL gotspace;              BOOL gotspace;
4644              int len = 1;              int len = 1;
4645              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4646                  {
4647                  SCHECK_PARTIAL();
4648                  break;
4649                  }
4650              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4651              switch(c)              switch(c)
4652                {                {
# Line 3955  for (;;) Line 4684  for (;;)
4684              {              {
4685              BOOL gotspace;              BOOL gotspace;
4686              int len = 1;              int len = 1;
4687              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4688                  {
4689                  SCHECK_PARTIAL();
4690                  break;
4691                  }
4692              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4693              switch(c)              switch(c)
4694                {                {
# Line 3979  for (;;) Line 4712  for (;;)
4712            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4713              {              {
4714              int len = 1;              int len = 1;
4715              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4716                  {
4717                  SCHECK_PARTIAL();
4718                  break;
4719                  }
4720              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4721              if (c < 256 && (md->ctypes[c] & ctype_digit) != 0) break;              if (c < 256 && (md->ctypes[c] & ctype_digit) != 0) break;
4722              eptr+= len;              eptr+= len;
# Line 3990  for (;;) Line 4727  for (;;)
4727            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4728              {              {
4729              int len = 1;              int len = 1;
4730              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4731                  {
4732                  SCHECK_PARTIAL();
4733                  break;
4734                  }
4735              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4736              if (c >= 256 ||(md->ctypes[c] & ctype_digit) == 0) break;              if (c >= 256 ||(md->ctypes[c] & ctype_digit) == 0) break;
4737              eptr+= len;              eptr+= len;
# Line 4001  for (;;) Line 4742  for (;;)
4742            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4743              {              {
4744              int len = 1;              int len = 1;
4745              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4746                  {
4747                  SCHECK_PARTIAL();
4748                  break;
4749                  }
4750              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4751              if (c < 256 && (md->ctypes[c] & ctype_space) != 0) break;              if (c < 256 && (md->ctypes[c] & ctype_space) != 0) break;
4752              eptr+= len;              eptr+= len;
# Line 4012  for (;;) Line 4757  for (;;)
4757            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4758              {              {
4759              int len = 1;              int len = 1;
4760              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4761                  {
4762                  SCHECK_PARTIAL();
4763                  break;
4764                  }
4765              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4766              if (c >= 256 ||(md->ctypes[c] & ctype_space) == 0) break;              if (c >= 256 ||(md->ctypes[c] & ctype_space) == 0) break;
4767              eptr+= len;              eptr+= len;
# Line 4023  for (;;) Line 4772  for (;;)
4772            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4773              {              {
4774              int len = 1;              int len = 1;
4775              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4776                  {
4777                  SCHECK_PARTIAL();
4778                  break;
4779                  }
4780              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4781              if (c < 256 && (md->ctypes[c] & ctype_word) != 0) break;              if (c < 256 && (md->ctypes[c] & ctype_word) != 0) break;
4782              eptr+= len;              eptr+= len;
# Line 4034  for (;;) Line 4787  for (;;)
4787            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4788              {              {
4789              int len = 1;              int len = 1;
4790              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4791                  {
4792                  SCHECK_PARTIAL();
4793                  break;
4794                  }
4795              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4796              if (c >= 256 || (md->ctypes[c] & ctype_word) == 0) break;              if (c >= 256 || (md->ctypes[c] & ctype_word) == 0) break;
4797              eptr+= len;              eptr+= len;
# Line 4064  for (;;) Line 4821  for (;;)
4821          switch(ctype)          switch(ctype)
4822            {            {
4823            case OP_ANY:            case OP_ANY:
4824            if ((ims & PCRE_DOTALL) == 0)            for (i = min; i < max; i++)
4825              {              {
4826              for (i = min; i < max; i++)              if (eptr >= md->end_subject)
4827                {                {
4828                if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;                SCHECK_PARTIAL();
4829                eptr++;                break;
4830                }                }
4831              break;              if (IS_NEWLINE(eptr)) break;
4832                eptr++;
4833              }              }
4834            /* For DOTALL case, fall through and treat as \C */            break;
4835    
4836              case OP_ALLANY:
4837            case OP_ANYBYTE:            case OP_ANYBYTE:
4838            c = max - min;            c = max - min;
4839            if (c > (unsigned int)(md->end_subject - eptr))            if (c > (unsigned int)(md->end_subject - eptr))
4840              c = md->end_subject - eptr;              {
4841            eptr += c;              eptr = md->end_subject;
4842                SCHECK_PARTIAL();
4843                }
4844              else eptr += c;
4845            break;            break;
4846    
4847            case OP_ANYNL:            case OP_ANYNL:
4848            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4849              {              {
4850              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4851                  {
4852                  SCHECK_PARTIAL();
4853                  break;
4854                  }
4855              c = *eptr;              c = *eptr;
4856              if (c == 0x000d)              if (c == 0x000d)
4857                {                {
# Line 4106  for (;;) Line 4872  for (;;)
4872            case OP_NOT_HSPACE:            case OP_NOT_HSPACE:
4873            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4874              {              {
4875              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4876                  {
4877                  SCHECK_PARTIAL();
4878                  break;
4879                  }
4880              c = *eptr;              c = *eptr;
4881              if (c == 0x09 || c == 0x20 || c == 0xa0) break;              if (c == 0x09 || c == 0x20 || c == 0xa0) break;
4882              eptr++;              eptr++;
# Line 4116  for (;;) Line 4886  for (;;)
4886            case OP_HSPACE:            case OP_HSPACE:
4887            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4888              {              {
4889              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4890                  {
4891                  SCHECK_PARTIAL();
4892                  break;
4893                  }
4894              c = *eptr;              c = *eptr;
4895              if (c != 0x09 && c != 0x20 && c != 0xa0) break;              if (c != 0x09 && c != 0x20 && c != 0xa0) break;
4896              eptr++;              eptr++;
# Line 4126  for (;;) Line 4900  for (;;)
4900            case OP_NOT_VSPACE:            case OP_NOT_VSPACE:
4901            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4902              {              {
4903              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4904                  {
4905                  SCHECK_PARTIAL();
4906                  break;
4907                  }
4908              c = *eptr;              c = *eptr;
4909              if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)              if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)
4910                break;                break;
# Line 4137  for (;;) Line 4915  for (;;)
4915            case OP_VSPACE:            case OP_VSPACE:
4916            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4917              {              {
4918              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject)
4919                  {
4920                  SCHECK_PARTIAL();
4921                  break;
4922                  }
4923              c = *eptr;              c = *eptr;
4924              if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85)              if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85)
4925                break;                break;
# Line 4148  for (;;) Line 4930  for (;;)
4930            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
4931            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4932              {              {
4933              if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_digit) != 0)              if (eptr >= md->end_subject)
4934                  {
4935                  SCHECK_PARTIAL();
4936                break;                break;
4937                  }
4938                if ((md->ctypes[*eptr] & ctype_digit) != 0) break;
4939              eptr++;              eptr++;
4940              }              }
4941            break;            break;
# Line 4157  for (;;) Line 4943  for (;;)
4943            case OP_DIGIT:            case OP_DIGIT:
4944            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4945              {              {
4946              if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_digit) == 0)              if (eptr >= md->end_subject)
4947                  {
4948                  SCHECK_PARTIAL();
4949                break;                break;
4950                  }
4951                if ((md->ctypes[*eptr] & ctype_digit) == 0) break;
4952              eptr++;              eptr++;
4953              }              }
4954            break;            break;
# Line 4166  for (;;) Line 4956  for (;;)
4956            case OP_NOT_WHITESPACE:            case OP_NOT_WHITESPACE:
4957            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4958              {              {
4959              if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_space) != 0)              if (eptr >= md->end_subject)
4960                  {
4961                  SCHECK_PARTIAL();
4962                break;                break;
4963                  }
4964                if ((md->ctypes[*eptr] & ctype_space) != 0) break;
4965              eptr++;              eptr++;
4966              }              }
4967            break;            break;
# Line 4175  for (;;) Line 4969  for (;;)
4969            case OP_WHITESPACE:            case OP_WHITESPACE:
4970            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4971              {              {
4972              if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_space) == 0)              if (eptr >= md->end_subject)
4973                  {
4974                  SCHECK_PARTIAL();
4975                break;                break;
4976                  }
4977                if ((md->ctypes[*eptr] & ctype_space) == 0) break;
4978              eptr++;              eptr++;
4979              }              }
4980            break;            break;
# Line 4184  for (;;) Line 4982  for (;;)
4982            case OP_NOT_WORDCHAR:            case OP_NOT_WORDCHAR:
4983            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4984              {              {
4985              if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_word) != 0)              if (eptr >= md->end_subject)
4986                  {
4987                  SCHECK_PARTIAL();
4988                break;                break;
4989                  }
4990                if ((md->ctypes[*eptr] & ctype_word) != 0) break;
4991              eptr++;              eptr++;
4992              }              }
4993            break;            break;
# Line 4193  for (;;) Line 4995  for (;;)
4995            case OP_WORDCHAR:            case OP_WORDCHAR:
4996            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4997              {              {
4998              if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_word) == 0)              if (eptr >= md->end_subject)
4999                  {
5000                  SCHECK_PARTIAL();
5001                break;                break;
5002                  }
5003                if ((md->ctypes[*eptr] & ctype_word) == 0) break;
5004              eptr++;              eptr++;
5005              }              }
5006            break;            break;
# Line 4348  Returns: > 0 => success; value Line 5154  Returns: > 0 => success; value
5154                   < -1 => some kind of unexpected problem                   < -1 => some kind of unexpected problem
5155  */  */
5156    
5157  PCRE_EXP_DEFN int  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
5158  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,
5159    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,
5160    int offsetcount)    int offsetcount)
# Line 4372  const uschar *tables; Line 5178  const uschar *tables;
5178  const uschar *start_bits = NULL;  const uschar *start_bits = NULL;
5179  USPTR start_match = (USPTR)subject + start_offset;  USPTR start_match = (USPTR)subject + start_offset;
5180  USPTR end_subject;  USPTR end_subject;
5181    USPTR start_partial = NULL;
5182  USPTR req_byte_ptr = start_match - 1;  USPTR req_byte_ptr = start_match - 1;
5183    
5184  pcre_study_data internal_study;  pcre_study_data internal_study;
# Line 4388  if (re == NULL || subject == NULL || Line 5195  if (re == NULL || subject == NULL ||
5195     (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL;     (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL;
5196  if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;  if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;
5197    
5198    /* This information is for finding all the numbers associated with a given
5199    name, for condition testing. */
5200    
5201    md->name_table = (uschar *)re + re->name_table_offset;
5202    md->name_count = re->name_count;
5203    md->name_entry_size = re->name_entry_size;
5204    
5205  /* Fish out the optional data from the extra_data structure, first setting  /* Fish out the optional data from the extra_data structure, first setting
5206  the default values. */  the default values. */
5207    
# Line 4450  end_subject = md->end_subject; Line 5264  end_subject = md->end_subject;
5264    
5265  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;
5266  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;
5267    md->jscript_compat = (re->options & PCRE_JAVASCRIPT_COMPAT) != 0;
5268    
5269  md->notbol = (options & PCRE_NOTBOL) != 0;  md->notbol = (options & PCRE_NOTBOL) != 0;
5270  md->noteol = (options & PCRE_NOTEOL) != 0;  md->noteol = (options & PCRE_NOTEOL) != 0;
5271  md->notempty = (options & PCRE_NOTEMPTY) != 0;  md->notempty = (options & PCRE_NOTEMPTY) != 0;
5272  md->partial = (options & PCRE_PARTIAL) != 0;  md->notempty_atstart = (options & PCRE_NOTEMPTY_ATSTART) != 0;
5273    md->partial = ((options & PCRE_PARTIAL_HARD) != 0)? 2 :
5274                  ((options & PCRE_PARTIAL_SOFT) != 0)? 1 : 0;
5275  md->hitend = FALSE;  md->hitend = FALSE;
5276    
5277  md->recursive = NULL;                   /* No recursion at top level */  md->recursive = NULL;                   /* No recursion at top level */
# Line 4495  switch ((((options & PCRE_NEWLINE_BITS) Line 5312  switch ((((options & PCRE_NEWLINE_BITS)
5312          (pcre_uint32)options) & PCRE_NEWLINE_BITS)          (pcre_uint32)options) & PCRE_NEWLINE_BITS)
5313    {    {
5314    case 0: newline = NEWLINE; break;   /* Compile-time default */    case 0: newline = NEWLINE; break;   /* Compile-time default */
5315    case PCRE_NEWLINE_CR: newline = '\r'; break;    case PCRE_NEWLINE_CR: newline = CHAR_CR; break;
5316    case PCRE_NEWLINE_LF: newline = '\n'; break;    case PCRE_NEWLINE_LF: newline = CHAR_NL; break;
5317    case PCRE_NEWLINE_CR+    case PCRE_NEWLINE_CR+
5318         PCRE_NEWLINE_LF: newline = ('\r' << 8) | '\n'; break;         PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break;
5319    case PCRE_NEWLINE_ANY: newline = -1; break;    case PCRE_NEWLINE_ANY: newline = -1; break;
5320    case PCRE_NEWLINE_ANYCRLF: newline = -2; break;    case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
5321    default: return PCRE_ERROR_BADNEWLINE;    default: return PCRE_ERROR_BADNEWLINE;
# Line 4528  else Line 5345  else
5345      }      }
5346    }    }
5347    
5348  /* Partial matching is supported only for a restricted set of regexes at the  /* Partial matching was originally supported only for a restricted set of
5349  moment. */  regexes; from release 8.00 there are no restrictions, but the bits are still
5350    defined (though never set). So there's no harm in leaving this code. */
5351    
5352  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
5353    return PCRE_ERROR_BADPARTIAL;    return PCRE_ERROR_BADPARTIAL;
# Line 4540  back the character offset. */ Line 5358  back the character offset. */
5358  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
5359  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)
5360    {    {
5361    if (_pcre_valid_utf8((uschar *)subject, length) >= 0)    if (_pcre_valid_utf8((USPTR)subject, length) >= 0)
5362      return PCRE_ERROR_BADUTF8;      return PCRE_ERROR_BADUTF8;
5363    if (start_offset > 0 && start_offset < length)    if (start_offset > 0 && start_offset < length)
5364      {      {
5365      int tb = ((uschar *)subject)[start_offset];      int tb = ((USPTR)subject)[start_offset];
5366      if (tb > 127)      if (tb > 127)
5367        {        {
5368        tb &= 0xc0;        tb &= 0xc0;
# Line 4616  if (!anchored) Line 5434  if (!anchored)
5434      }      }
5435    else    else
5436      if (!startline && study != NULL &&      if (!startline && study != NULL &&
5437        (study->options & PCRE_STUDY_MAPPED) != 0)        (study->flags & PCRE_STUDY_MAPPED) != 0)
5438          start_bits = study->start_bits;          start_bits = study->start_bits;
5439    }    }
5440    
# Line 4650  for(;;) Line 5468  for(;;)
5468      while (iptr < iend) *iptr++ = -1;      while (iptr < iend) *iptr++ = -1;
5469      }      }
5470    
5471    /* Advance to a unique first char if possible. If firstline is TRUE, the    /* If firstline is TRUE, the start of the match is constrained to the first
5472    start of the match is constrained to the first line of a multiline string.    line of a multiline string. That is, the match must be before or at the first
5473    That is, the match must be before or at the first newline. Implement this by    newline. Implement this by temporarily adjusting end_subject so that we stop
5474    temporarily adjusting end_subject so that we stop scanning at a newline. If    scanning at a newline. If the match fails at the newline, later code breaks
5475    the match fails at the newline, later code breaks this loop. */    this loop. */
5476    
5477    if (firstline)    if (firstline)
5478      {      {
5479      USPTR t = start_match;      USPTR t = start_match;
5480    #ifdef SUPPORT_UTF8
5481        if (utf8)
5482          {
5483          while (t < md->end_subject && !IS_NEWLINE(t))
5484            {
5485            t++;
5486            while (t < end_subject && (*t & 0xc0) == 0x80) t++;
5487            }
5488          }
5489        else
5490    #endif
5491      while (t < md->end_subject && !IS_NEWLINE(t)) t++;      while (t < md->end_subject && !IS_NEWLINE(t)) t++;
5492      end_subject = t;      end_subject = t;
5493      }      }
5494    
5495    /* Now test for a unique first byte */    /* There are some optimizations that avoid running the match if a known
5496      starting point is not found, or if a known later character is not present.
5497      However, there is an option that disables these, for testing and for ensuring
5498      that all callouts do actually occur. */
5499    
5500    if (first_byte >= 0)    if ((options & PCRE_NO_START_OPTIMIZE) == 0)
5501      {      {
5502      if (first_byte_caseless)      /* Advance to a unique first byte if there is one. */
5503        while (start_match < end_subject &&  
5504               md->lcc[*start_match] != first_byte)      if (first_byte >= 0)
5505          { NEXTCHAR(start_match); }        {
5506      else        if (first_byte_caseless)
5507        while (start_match < end_subject && *start_match != first_byte)          while (start_match < end_subject && md->lcc[*start_match] != first_byte)
5508          { NEXTCHAR(start_match); }            start_match++;
5509      }        else
5510            while (start_match < end_subject && *start_match != first_byte)
5511              start_match++;
5512          }
5513    
5514    /* Or to just after a linebreak for a multiline match if possible */      /* Or to just after a linebreak for a multiline match */
5515    
5516    else if (startline)      else if (startline)
     {  
     if (start_match > md->start_subject + start_offset)  
5517        {        {
5518        while (start_match <= end_subject && !WAS_NEWLINE(start_match))        if (start_match > md->start_subject + start_offset)
5519          { NEXTCHAR(start_match); }          {
5520    #ifdef SUPPORT_UTF8
5521            if (utf8)
5522              {
5523              while (start_match < end_subject && !WAS_NEWLINE(start_match))
5524                {
5525                start_match++;
5526                while(start_match < end_subject && (*start_match & 0xc0) == 0x80)
5527                  start_match++;
5528                }
5529              }
5530            else
5531    #endif
5532            while (start_match < end_subject && !WAS_NEWLINE(start_match))
5533              start_match++;
5534    
5535            /* If we have just passed a CR and the newline option is ANY or ANYCRLF,
5536            and we are now at a LF, advance the match position by one more character.
5537            */
5538    
5539        /* If we have just passed a CR and the newline option is ANY or ANYCRLF,          if (start_match[-1] == CHAR_CR &&
5540        and we are now at a LF, advance the match position by one more character.               (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
5541        */               start_match < end_subject &&
5542                 *start_match == CHAR_NL)
5543        if (start_match[-1] == '\r' &&            start_match++;
5544             (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&          }
            start_match < end_subject &&  
            *start_match == '\n')  
         start_match++;  
5545        }        }
     }  
5546    
5547    /* Or to a non-unique first char after study */      /* Or to a non-unique first byte after study */
5548    
5549    else if (start_bits != NULL)      else if (start_bits != NULL)
     {  
     while (start_match < end_subject)  
5550        {        {
5551        register unsigned int c = *start_match;        while (start_match < end_subject)
5552        if ((start_bits[c/8] & (1 << (c&7))) == 0)          {
5553          { NEXTCHAR(start_match); }          register unsigned int c = *start_match;
5554        else break;          if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++;
5555              else break;
5556            }
5557        }        }
5558      }      }   /* Starting optimizations */
5559    
5560    /* Restore fudged end_subject */    /* Restore fudged end_subject */
5561    
5562    end_subject = save_end_subject;    end_subject = save_end_subject;
5563    
5564  #ifdef DEBUG  /* Sigh. Some compilers never learn. */    /* The following two optimizations are disabled for partial matching or if
5565    printf(">>>> Match against: ");    disabling is explicitly requested. */
   pchars(start_match, end_subject - start_match, TRUE, md);  
   printf("\n");  
 #endif  
5566    
5567    /* If req_byte is set, we know that that character must appear in the subject    if ((options & PCRE_NO_START_OPTIMIZE) == 0 && !md->partial)
   for the match to succeed. If the first character is set, req_byte must be  
   later in the subject; otherwise the test starts at the match point. This  
   optimization can save a huge amount of backtracking in patterns with nested  
   unlimited repeats that aren't going to match. Writing separate code for  
   cased/caseless versions makes it go faster, as does using an autoincrement  
   and backing off on a match.  
   
   HOWEVER: when the subject string is very, very long, searching to its end can  
   take a long time, and give bad performance on quite ordinary patterns. This  
   showed up when somebody was matching something like /^\d+C/ on a 32-megabyte  
   string... so we don't do this when the string is sufficiently long.  
   
   ALSO: this processing is disabled when partial matching is requested.  
   */  
   
   if (req_byte >= 0 &&  
       end_subject - start_match < REQ_BYTE_MAX &&  
       !md->partial)  
5568      {      {
5569      register USPTR p = start_match + ((first_byte >= 0)? 1 : 0);      /* If the pattern was studied, a minimum subject length may be set. This is
5570        a lower bound; no actual string of that length may actually match the
5571        pattern. Although the value is, strictly, in characters, we treat it as
5572        bytes to avoid spending too much time in this optimization. */
5573    
5574        if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 &&
5575            end_subject - start_match < study->minlength)
5576          {
5577          rc = MATCH_NOMATCH;
5578          break;
5579          }
5580    
5581        /* If req_byte is set, we know that that character must appear in the
5582        subject for the match to succeed. If the first character is set, req_byte
5583        must be later in the subject; otherwise the test starts at the match point.
5584        This optimization can save a huge amount of backtracking in patterns with
5585        nested unlimited repeats that aren't going to match. Writing separate code
5586        for cased/caseless versions makes it go faster, as does using an
5587        autoincrement and backing off on a match.
5588    
5589      /* We don't need to repeat the search if we haven't yet reached the      HOWEVER: when the subject string is very, very long, searching to its end
5590      place we found it at last time. */      can take a long time, and give bad performance on quite ordinary patterns.
5591        This showed up when somebody was matching something like /^\d+C/ on a
5592        32-megabyte string... so we don't do this when the string is sufficiently
5593        long. */
5594    
5595      if (p > req_byte_ptr)      if (req_byte >= 0 && end_subject - start_match < REQ_BYTE_MAX)
5596        {        {
5597        if (req_byte_caseless)        register USPTR p = start_match + ((first_byte >= 0)? 1 : 0);
5598    
5599          /* We don't need to repeat the search if we haven't yet reached the
5600          place we found it at last time. */
5601    
5602          if (p > req_byte_ptr)
5603          {          {
5604          while (p < end_subject)          if (req_byte_caseless)
5605            {            {
5606            register int pp = *p++;            while (p < end_subject)
5607            if (pp == req_byte || pp == req_byte2) { p--; break; }              {
5608                register int pp = *p++;
5609                if (pp == req_byte || pp == req_byte2) { p--; break; }
5610                }
5611            }            }
5612          }          else
       else  
         {  
         while (p < end_subject)  
5613            {            {
5614            if (*p++ == req_byte) { p--; break; }            while (p < end_subject)
5615                {
5616                if (*p++ == req_byte) { p--; break; }
5617                }
5618            }            }
         }  
5619    
5620        /* If we can't find the required character, break the matching loop,          /* If we can't find the required character, break the matching loop,
5621        forcing a match failure. */          forcing a match failure. */
5622    
5623        if (p >= end_subject)          if (p >= end_subject)
5624          {            {
5625          rc = MATCH_NOMATCH;            rc = MATCH_NOMATCH;
5626          break;            break;
5627          }            }
5628    
5629        /* If we have found the required character, save the point where we          /* If we have found the required character, save the point where we
5630        found it, so that we don't search again next time round the loop if          found it, so that we don't search again next time round the loop if
5631        the start hasn't passed this character yet. */          the start hasn't passed this character yet. */
5632    
5633        req_byte_ptr = p;          req_byte_ptr = p;
5634            }
5635        }        }
5636      }      }
5637    
5638    /* OK, we can now run the match. */  #ifdef PCRE_DEBUG  /* Sigh. Some compilers never learn. */
5639      printf(">>>> Match against: ");
5640      pchars(start_match, end_subject - start_match, TRUE, md);
5641      printf("\n");
5642    #endif
5643    
5644      /* OK, we can now run the match. If "hitend" is set afterwards, remember the
5645      first starting point for which a partial match was found. */
5646    
5647    md->start_match_ptr = start_match;    md->start_match_ptr = start_match;
5648      md->start_used_ptr = start_match;
5649    md->match_call_count = 0;    md->match_call_count = 0;
5650    rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0);    rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0);
5651      if (md->hitend && start_partial == NULL) start_partial = md->start_used_ptr;
5652    
5653    switch(rc)    switch(rc)
5654      {      {
# Line 4814  for(;;) Line 5678  for(;;)
5678      rc = MATCH_NOMATCH;      rc = MATCH_NOMATCH;
5679      goto ENDLOOP;      goto ENDLOOP;
5680    
5681      /* Any other return is some kind of error. */      /* Any other return is either a match, or some kind of error. */
5682    
5683      default:      default:
5684      goto ENDLOOP;      goto ENDLOOP;
# Line 4844  for(;;) Line 5708  for(;;)
5708    not contain any explicit matches for \r or \n, and the newline option is CRLF    not contain any explicit matches for \r or \n, and the newline option is CRLF
5709    or ANY or ANYCRLF, advance the match position by one more character. */    or ANY or ANYCRLF, advance the match position by one more character. */
5710    
5711    if (start_match[-1] == '\r' &&    if (start_match[-1] == CHAR_CR &&
5712        start_match < end_subject &&        start_match < end_subject &&
5713        *start_match == '\n' &&        *start_match == CHAR_NL &&
5714        (re->flags & PCRE_HASCRORLF) == 0 &&        (re->flags & PCRE_HASCRORLF) == 0 &&
5715          (md->nltype == NLTYPE_ANY ||          (md->nltype == NLTYPE_ANY ||
5716           md->nltype == NLTYPE_ANYCRLF ||           md->nltype == NLTYPE_ANYCRLF ||
# Line 4920  if (using_temporary_offsets) Line 5784  if (using_temporary_offsets)
5784    (pcre_free)(md->offset_vector);    (pcre_free)(md->offset_vector);
5785    }    }
5786    
5787  if (rc != MATCH_NOMATCH)  if (rc != MATCH_NOMATCH && rc != PCRE_ERROR_PARTIAL)
5788    {    {
5789    DPRINTF((">>>> error: returning %d\n", rc));    DPRINTF((">>>> error: returning %d\n", rc));
5790    return rc;    return rc;
5791    }    }
5792  else if (md->partial && md->hitend)  else if (start_partial != NULL)
5793    {    {
5794    DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n"));    DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n"));
5795      if (offsetcount > 1)
5796        {
5797        offsets[0] = start_partial - (USPTR)subject;
5798        offsets[1] = end_subject - (USPTR)subject;
5799        }
5800    return PCRE_ERROR_PARTIAL;    return PCRE_ERROR_PARTIAL;
5801    }    }
5802  else  else
Legend:
Removed from v.305  
changed lines
  Added in v.475
webmaster@exim.org
 ViewVC Help
Powered by ViewVC 1.1.12