/[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 144 by ph10, Mon Apr 2 13:32:07 2007 UTC revision 444 by ph10, Sun Sep 13 16:26:39 2009 UTC
# Line 6  Line 6 
6  and semantics are as close as possible to those of the Perl 5 language.  and semantics are as close as possible to those of the Perl 5 language.
7    
8                         Written by Philip Hazel                         Written by Philip Hazel
9             Copyright (c) 1997-2007 University of Cambridge             Copyright (c) 1997-2009 University of Cambridge
10    
11  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
12  Redistribution and use in source and binary forms, with or without  Redistribution and use in source and binary forms, with or without
# Line 42  POSSIBILITY OF SUCH DAMAGE. Line 42  POSSIBILITY OF SUCH DAMAGE.
42  pattern matching using an NFA algorithm, trying to mimic Perl as closely as  pattern matching using an NFA algorithm, trying to mimic Perl as closely as
43  possible. There are also some static supporting functions. */  possible. There are also some static supporting functions. */
44    
45    #ifdef HAVE_CONFIG_H
46    #include "config.h"
47    #endif
48    
49  #define NLBLOCK md             /* Block containing newline information */  #define NLBLOCK md             /* Block containing newline information */
50  #define PSSTART start_subject  /* Field containing processed string start */  #define PSSTART start_subject  /* Field containing processed string start */
51  #define PSEND   end_subject    /* Field containing processed string end */  #define PSEND   end_subject    /* Field containing processed string end */
# Line 53  possible. There are also some static sup Line 57  possible. There are also some static sup
57  #undef min  #undef min
58  #undef max  #undef max
59    
 /* The chain of eptrblocks for tail recursions uses memory in stack workspace,  
 obtained at top level, the size of which is defined by EPTR_WORK_SIZE. */  
   
 #define EPTR_WORK_SIZE (1000)  
   
60  /* Flag bits for the match() function */  /* Flag bits for the match() function */
61    
62  #define match_condassert     0x01  /* Called to check a condition assertion */  #define match_condassert     0x01  /* Called to check a condition assertion */
63  #define match_cbegroup       0x02  /* Could-be-empty unlimited repeat group */  #define match_cbegroup       0x02  /* Could-be-empty unlimited repeat group */
 #define match_tail_recursed  0x04  /* Tail recursive call */  
64    
65  /* Non-error returns from the match() function. Error returns are externally  /* Non-error returns from the match() function. Error returns are externally
66  defined PCRE_ERROR_xxx codes, which are all negative. */  defined PCRE_ERROR_xxx codes, which are all negative. */
# Line 70  defined PCRE_ERROR_xxx codes, which are Line 68  defined PCRE_ERROR_xxx codes, which are
68  #define MATCH_MATCH        1  #define MATCH_MATCH        1
69  #define MATCH_NOMATCH      0  #define MATCH_NOMATCH      0
70    
71    /* Special internal returns from the match() function. Make them sufficiently
72    negative to avoid the external error codes. */
73    
74    #define MATCH_COMMIT       (-999)
75    #define MATCH_PRUNE        (-998)
76    #define MATCH_SKIP         (-997)
77    #define MATCH_THEN         (-996)
78    
79  /* Maximum number of ints of offset to save on the stack for recursive calls.  /* Maximum number of ints of offset to save on the stack for recursive calls.
80  If the offset vector is bigger, malloc is used. This should be a multiple of 3,  If the offset vector is bigger, malloc is used. This should be a multiple of 3,
81  because the offset vector is always a multiple of 3 long. */  because the offset vector is always a multiple of 3 long. */
# Line 152  printf("\n"); Line 158  printf("\n");
158    
159  if (length > md->end_subject - eptr) return FALSE;  if (length > md->end_subject - eptr) return FALSE;
160    
161  /* Separate the caselesss case for speed */  /* Separate the caseless case for speed. In UTF-8 mode we can only do this
162    properly if Unicode properties are supported. Otherwise, we can check only
163    ASCII characters. */
164    
165  if ((ims & PCRE_CASELESS) != 0)  if ((ims & PCRE_CASELESS) != 0)
166    {    {
167    #ifdef SUPPORT_UTF8
168    #ifdef SUPPORT_UCP
169      if (md->utf8)
170        {
171        USPTR endptr = eptr + length;
172        while (eptr < endptr)
173          {
174          int c, d;
175          GETCHARINC(c, eptr);
176          GETCHARINC(d, p);
177          if (c != d && c != UCD_OTHERCASE(d)) return FALSE;
178          }
179        }
180      else
181    #endif
182    #endif
183    
184      /* The same code works when not in UTF-8 mode and in UTF-8 mode when there
185      is no UCP support. */
186    
187    while (length-- > 0)    while (length-- > 0)
188      if (md->lcc[*p++] != md->lcc[*eptr++]) return FALSE;      { if (md->lcc[*p++] != md->lcc[*eptr++]) return FALSE; }
189    }    }
190    
191    /* In the caseful case, we can just compare the bytes, whether or not we
192    are in UTF-8 mode. */
193    
194  else  else
195    { while (length-- > 0) if (*p++ != *eptr++) return FALSE; }    { while (length-- > 0) if (*p++ != *eptr++) return FALSE; }
196    
# Line 188  calls by keeping local variables that ne Line 220  calls by keeping local variables that ne
220  obtained from malloc() instead instead of on the stack. Macros are used to  obtained from malloc() instead instead of on the stack. Macros are used to
221  achieve this so that the actual code doesn't look very different to what it  achieve this so that the actual code doesn't look very different to what it
222  always used to.  always used to.
223    
224    The original heap-recursive code used longjmp(). However, it seems that this
225    can be very slow on some operating systems. Following a suggestion from Stan
226    Switzer, the use of longjmp() has been abolished, at the cost of having to
227    provide a unique number for each call to RMATCH. There is no way of generating
228    a sequence of numbers at compile time in C. I have given them names, to make
229    them stand out more clearly.
230    
231    Crude tests on x86 Linux show a small speedup of around 5-8%. However, on
232    FreeBSD, avoiding longjmp() more than halves the time taken to run the standard
233    tests. Furthermore, not using longjmp() means that local dynamic variables
234    don't have indeterminate values; this has meant that the frame size can be
235    reduced because the result can be "passed back" by straight setting of the
236    variable instead of being passed in the frame.
237  ****************************************************************************  ****************************************************************************
238  ***************************************************************************/  ***************************************************************************/
239    
240    /* Numbers for RMATCH calls. When this list is changed, the code at HEAP_RETURN
241    below must be updated in sync.  */
242    
243    enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM6,  RM7,  RM8,  RM9,  RM10,
244           RM11,  RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,
245           RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,
246           RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
247           RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
248           RM51,  RM52, RM53, RM54 };
249    
250  /* These versions of the macros use the stack, as normal. There are debugging  /* These versions of the macros use the stack, as normal. There are debugging
251  versions and production versions. */  versions and production versions. Note that the "rw" argument of RMATCH isn't
252    actuall used in this definition. */
253    
254  #ifndef NO_RECURSE  #ifndef NO_RECURSE
255  #define REGISTER register  #define REGISTER register
256    
257  #ifdef DEBUG  #ifdef DEBUG
258  #define RMATCH(rx,ra,rb,rc,rd,re,rf,rg) \  #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__); \
261    rx = match(ra,rb,rc,rd,re,rf,rg,rdepth+1); \    rrc = match(ra,rb,mstart,rc,rd,re,rf,rg,rdepth+1); \
262    printf("to line %d\n", __LINE__); \    printf("to line %d\n", __LINE__); \
263    }    }
264  #define RRETURN(ra) \  #define RRETURN(ra) \
# Line 210  versions and production versions. */ Line 267  versions and production versions. */
267    return ra; \    return ra; \
268    }    }
269  #else  #else
270  #define RMATCH(rx,ra,rb,rc,rd,re,rf,rg) \  #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw) \
271    rx = match(ra,rb,rc,rd,re,rf,rg,rdepth+1)    rrc = match(ra,rb,mstart,rc,rd,re,rf,rg,rdepth+1)
272  #define RRETURN(ra) return ra  #define RRETURN(ra) return ra
273  #endif  #endif
274    
275  #else  #else
276    
277    
278  /* These versions of the macros manage a private stack on the heap. Note  /* These versions of the macros manage a private stack on the heap. Note that
279  that the rd argument of RMATCH isn't actually used. It's the md argument of  the "rd" argument of RMATCH isn't actually used in this definition. It's the md
280  match(), which never changes. */  argument of match(), which never changes. */
281    
282  #define REGISTER  #define REGISTER
283    
284  #define RMATCH(rx,ra,rb,rc,rd,re,rf,rg)\  #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw)\
285    {\    {\
286    heapframe *newframe = (pcre_stack_malloc)(sizeof(heapframe));\    heapframe *newframe = (pcre_stack_malloc)(sizeof(heapframe));\
287    if (setjmp(frame->Xwhere) == 0)\    frame->Xwhere = rw; \
288      {\    newframe->Xeptr = ra;\
289      newframe->Xeptr = ra;\    newframe->Xecode = rb;\
290      newframe->Xecode = rb;\    newframe->Xmstart = mstart;\
291      newframe->Xoffset_top = rc;\    newframe->Xoffset_top = rc;\
292      newframe->Xims = re;\    newframe->Xims = re;\
293      newframe->Xeptrb = rf;\    newframe->Xeptrb = rf;\
294      newframe->Xflags = rg;\    newframe->Xflags = rg;\
295      newframe->Xrdepth = frame->Xrdepth + 1;\    newframe->Xrdepth = frame->Xrdepth + 1;\
296      newframe->Xprevframe = frame;\    newframe->Xprevframe = frame;\
297      frame = newframe;\    frame = newframe;\
298      DPRINTF(("restarting from line %d\n", __LINE__));\    DPRINTF(("restarting from line %d\n", __LINE__));\
299      goto HEAP_RECURSE;\    goto HEAP_RECURSE;\
300      }\    L_##rw:\
301    else\    DPRINTF(("jumped back to line %d\n", __LINE__));\
     {\  
     DPRINTF(("longjumped back to line %d\n", __LINE__));\  
     frame = md->thisframe;\  
     rx = frame->Xresult;\  
     }\  
302    }    }
303    
304  #define RRETURN(ra)\  #define RRETURN(ra)\
# Line 256  match(), which never changes. */ Line 308  match(), which never changes. */
308    (pcre_stack_free)(newframe);\    (pcre_stack_free)(newframe);\
309    if (frame != NULL)\    if (frame != NULL)\
310      {\      {\
311      frame->Xresult = ra;\      rrc = ra;\
312      md->thisframe = frame;\      goto HEAP_RETURN;\
     longjmp(frame->Xwhere, 1);\  
313      }\      }\
314    return ra;\    return ra;\
315    }    }
# Line 271  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      USPTR Xmstart;
328    int Xoffset_top;    int Xoffset_top;
329    long int Xims;    long int Xims;
330    eptrblock *Xeptrb;    eptrblock *Xeptrb;
# Line 281  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 308  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 323  typedef struct heapframe { Line 378  typedef struct heapframe {
378    
379    eptrblock Xnewptrb;    eptrblock Xnewptrb;
380    
381    /* Place to pass back result, and where to jump back to */    /* Where to jump back to */
382    
383    int  Xresult;    int Xwhere;
   jmp_buf Xwhere;  
384    
385  } heapframe;  } heapframe;
386    
# Line 344  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    #define CHECK_PARTIAL()\
411      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  Performance note: It might be tempting to extract commonly used fields from the  #define SCHECK_PARTIAL()\
418  md structure (e.g. utf8, end_subject) into individual variables to improve    if (md->partial && eptr > mstart)\
419        {\
420        md->hitend = TRUE;\
421        if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);\
422        }
423    
424    
425    /* Performance note: It might be tempting to extract commonly used fields from
426    the md structure (e.g. utf8, end_subject) into individual variables to improve
427  performance. Tests using gcc on a SPARC disproved this; in the first case, it  performance. Tests using gcc on a SPARC disproved this; in the first case, it
428  made performance worse.  made performance worse.
429    
430  Arguments:  Arguments:
431     eptr        pointer to current character in subject     eptr        pointer to current character in subject
432     ecode       pointer to current position in compiled code     ecode       pointer to current position in compiled code
433       mstart      pointer to the current match start position (can be modified
434                     by encountering \K)
435     offset_top  current top pointer     offset_top  current top pointer
436     md          pointer to "static" info for the match     md          pointer to "static" info for the match
437     ims         current /i, /m, and /s options     ims         current /i, /m, and /s options
# Line 363  Arguments: Line 441  Arguments:
441                   match_condassert - this is an assertion condition                   match_condassert - this is an assertion condition
442                   match_cbegroup - this is the start of an unlimited repeat                   match_cbegroup - this is the start of an unlimited repeat
443                     group that can match an empty string                     group that can match an empty string
                  match_tail_recursed - this is a tail_recursed group  
444     rdepth      the recursion depth     rdepth      the recursion depth
445    
446  Returns:       MATCH_MATCH if matched            )  these values are >= 0  Returns:       MATCH_MATCH if matched            )  these values are >= 0
# Line 373  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,  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 387  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 401  frame->Xprevframe = NULL;            /* Line 479  frame->Xprevframe = NULL;            /*
479    
480  frame->Xeptr = eptr;  frame->Xeptr = eptr;
481  frame->Xecode = ecode;  frame->Xecode = ecode;
482    frame->Xmstart = mstart;
483  frame->Xoffset_top = offset_top;  frame->Xoffset_top = offset_top;
484  frame->Xims = ims;  frame->Xims = ims;
485  frame->Xeptrb = eptrb;  frame->Xeptrb = eptrb;
# Line 415  HEAP_RECURSE: Line 494  HEAP_RECURSE:
494    
495  #define eptr               frame->Xeptr  #define eptr               frame->Xeptr
496  #define ecode              frame->Xecode  #define ecode              frame->Xecode
497    #define mstart             frame->Xmstart
498  #define offset_top         frame->Xoffset_top  #define offset_top         frame->Xoffset_top
499  #define ims                frame->Xims  #define ims                frame->Xims
500  #define eptrb              frame->Xeptrb  #define eptrb              frame->Xeptrb
# Line 427  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 507  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 545  defined). However, RMATCH isn't like a f Line 627  defined). However, RMATCH isn't like a f
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. */
629    
630    #ifdef SUPPORT_UTF8
631    utf8 = md->utf8;       /* Local copy of the flag */
632    #else
633    utf8 = FALSE;
634    #endif
635    
636  /* First check that we haven't called match() too many times, or that we  /* First check that we haven't called match() too many times, or that we
637  haven't exceeded the recursive call limit. */  haven't exceeded the recursive call limit. */
638    
# Line 553  if (rdepth >= md->match_limit_recursion) Line 641  if (rdepth >= md->match_limit_recursion)
641    
642  original_ims = ims;    /* Save for resetting on ')' */  original_ims = ims;    /* Save for resetting on ')' */
643    
 #ifdef SUPPORT_UTF8  
 utf8 = md->utf8;       /* Local copy of the flag */  
 #else  
 utf8 = FALSE;  
 #endif  
   
644  /* At the start of a group with an unlimited repeat that may match an empty  /* At the start of a group with an unlimited repeat that may match an empty
645  string, the match_cbegroup flag is set. When this is the case, add the current  string, the match_cbegroup flag is set. When this is the case, add the current
646  subject pointer to the chain of such remembered pointers, to be checked when we  subject pointer to the chain of such remembered pointers, to be checked when we
647  hit the closing ket, in order to break infinite loops that match no characters.  hit the closing ket, in order to break infinite loops that match no characters.
648  When match() is called in other circumstances, don't add to the chain. If this  When match() is called in other circumstances, don't add to the chain. The
649  is a tail recursion, use a block from the workspace, as the one on the stack is  match_cbegroup flag must NOT be used with tail recursion, because the memory
650  already used. */  block that is used is on the stack, so a new one may be required for each
651    match(). */
652    
653  if ((flags & match_cbegroup) != 0)  if ((flags & match_cbegroup) != 0)
654    {    {
655    eptrblock *p;    newptrb.epb_saved_eptr = eptr;
656    if ((flags & match_tail_recursed) != 0)    newptrb.epb_prev = eptrb;
657      {    eptrb = &newptrb;
     if (md->eptrn >= EPTR_WORK_SIZE) RRETURN(PCRE_ERROR_NULLWSLIMIT);  
     p = md->eptrchain + md->eptrn++;  
     }  
   else p = &newptrb;  
   p->epb_saved_eptr = eptr;  
   p->epb_prev = eptrb;  
   eptrb = p;  
658    }    }
659    
660  /* Now start processing the opcodes. */  /* Now start processing the opcodes. */
# Line 588  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 > md->start_match)  
     md->hitend = TRUE;  
   
667    switch(op)    switch(op)
668      {      {
669        case OP_FAIL:
670        RRETURN(MATCH_NOMATCH);
671    
672        case OP_PRUNE:
673        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
674          ims, eptrb, flags, RM51);
675        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
676        RRETURN(MATCH_PRUNE);
677    
678        case OP_COMMIT:
679        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
680          ims, eptrb, flags, RM52);
681        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
682        RRETURN(MATCH_COMMIT);
683    
684        case OP_SKIP:
685        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
686          ims, eptrb, flags, RM53);
687        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
688        md->start_match_ptr = eptr;   /* Pass back current position */
689        RRETURN(MATCH_SKIP);
690    
691        case OP_THEN:
692        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
693          ims, eptrb, flags, RM54);
694        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
695        RRETURN(MATCH_THEN);
696    
697      /* Handle a capturing bracket. If there is space in the offset vector, save      /* Handle a capturing bracket. If there is space in the offset vector, save
698      the current subject position in the working slot at the top of the vector.      the current subject position in the working slot at the top of the vector.
699      We mustn't change the current values of the data slot, because they may be      We mustn't change the current values of the data slot, because they may be
# Line 637  for (;;) Line 733  for (;;)
733        flags = (op == OP_SCBRA)? match_cbegroup : 0;        flags = (op == OP_SCBRA)? match_cbegroup : 0;
734        do        do
735          {          {
736          RMATCH(rrc, eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
737            ims, eptrb, flags);            ims, eptrb, flags, RM1);
738          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
739          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
740          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
741          }          }
# Line 654  for (;;) Line 750  for (;;)
750        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
751        }        }
752    
753      /* Insufficient room for saving captured contents. Treat as a non-capturing      /* FALL THROUGH ... Insufficient room for saving captured contents. Treat
754      bracket. */      as a non-capturing bracket. */
755    
756        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
757        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
758    
759      DPRINTF(("insufficient capture room: treat as non-capturing\n"));      DPRINTF(("insufficient capture room: treat as non-capturing\n"));
760    
761        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
762        /* VVVVVVVVVVVVVVVVVVVVVVVVV */
763    
764      /* Non-capturing bracket. Loop for all the alternatives. When we get to the      /* Non-capturing bracket. Loop for all the alternatives. When we get to the
765      final alternative within the brackets, we would return the result of a      final alternative within the brackets, we would return the result of a
766      recursive call to match() whatever happened. We can reduce stack usage by      recursive call to match() whatever happened. We can reduce stack usage by
767      turning this into a tail recursion. */      turning this into a tail recursion, except in the case when match_cbegroup
768        is set.*/
769    
770      case OP_BRA:      case OP_BRA:
771      case OP_SBRA:      case OP_SBRA:
# Line 670  for (;;) Line 773  for (;;)
773      flags = (op >= OP_SBRA)? match_cbegroup : 0;      flags = (op >= OP_SBRA)? match_cbegroup : 0;
774      for (;;)      for (;;)
775        {        {
776        if (ecode[GET(ecode, 1)] != OP_ALT)        if (ecode[GET(ecode, 1)] != OP_ALT)   /* Final alternative */
777          {          {
778          ecode += _pcre_OP_lengths[*ecode];          if (flags == 0)    /* Not a possibly empty group */
779          flags |= match_tail_recursed;            {
780          DPRINTF(("bracket 0 tail recursion\n"));            ecode += _pcre_OP_lengths[*ecode];
781          goto TAIL_RECURSE;            DPRINTF(("bracket 0 tail recursion\n"));
782              goto TAIL_RECURSE;
783              }
784    
785            /* Possibly empty group; can't use tail recursion. */
786    
787            RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
788              eptrb, flags, RM48);
789            RRETURN(rrc);
790          }          }
791    
792        /* For non-final alternatives, continue the loop for a NOMATCH result;        /* For non-final alternatives, continue the loop for a NOMATCH result;
793        otherwise return. */        otherwise return. */
794    
795        RMATCH(rrc, eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
796          eptrb, flags);          eptrb, flags, RM2);
797        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
798        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
799        }        }
800      /* Control never reaches here. */      /* Control never reaches here. */
# Line 696  for (;;) Line 807  for (;;)
807    
808      case OP_COND:      case OP_COND:
809      case OP_SCOND:      case OP_SCOND:
810      if (ecode[LINK_SIZE+1] == OP_RREF)         /* Recursion test */      codelink= GET(ecode, 1);
811    
812        /* Because of the way auto-callout works during compile, a callout item is
813        inserted between OP_COND and an assertion condition. */
814    
815        if (ecode[LINK_SIZE+1] == OP_CALLOUT)
816          {
817          if (pcre_callout != NULL)
818            {
819            pcre_callout_block cb;
820            cb.version          = 1;   /* Version 1 of the callout block */
821            cb.callout_number   = ecode[LINK_SIZE+2];
822            cb.offset_vector    = md->offset_vector;
823            cb.subject          = (PCRE_SPTR)md->start_subject;
824            cb.subject_length   = md->end_subject - md->start_subject;
825            cb.start_match      = mstart - md->start_subject;
826            cb.current_position = eptr - md->start_subject;
827            cb.pattern_position = GET(ecode, LINK_SIZE + 3);
828            cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);
829            cb.capture_top      = offset_top/2;
830            cb.capture_last     = md->capture_last;
831            cb.callout_data     = md->callout_data;
832            if ((rrc = (*pcre_callout)(&cb)) > 0) RRETURN(MATCH_NOMATCH);
833            if (rrc < 0) RRETURN(rrc);
834            }
835          ecode += _pcre_OP_lengths[OP_CALLOUT];
836          }
837    
838        condcode = ecode[LINK_SIZE+1];
839    
840        /* Now see what the actual condition is */
841    
842        if (condcode == OP_RREF)         /* Recursion test */
843        {        {
844        offset = GET2(ecode, LINK_SIZE + 2);     /* Recursion group number*/        offset = GET2(ecode, LINK_SIZE + 2);     /* Recursion group number*/
845        condition = md->recursive != NULL &&        condition = md->recursive != NULL &&
# Line 704  for (;;) Line 847  for (;;)
847        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 3 : GET(ecode, 1);
848        }        }
849    
850      else if (ecode[LINK_SIZE+1] == OP_CREF)    /* Group used test */      else if (condcode == OP_CREF)    /* Group used test */
851        {        {
852        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */        offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */
853        condition = offset < offset_top && md->offset_vector[offset] >= 0;        condition = offset < offset_top && md->offset_vector[offset] >= 0;
854        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 3 : GET(ecode, 1);
855        }        }
856    
857      else if (ecode[LINK_SIZE+1] == OP_DEF)     /* DEFINE - always false */      else if (condcode == OP_DEF)     /* DEFINE - always false */
858        {        {
859        condition = FALSE;        condition = FALSE;
860        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
# Line 723  for (;;) Line 866  for (;;)
866    
867      else      else
868        {        {
869        RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL,
870            match_condassert);            match_condassert, RM3);
871        if (rrc == MATCH_MATCH)        if (rrc == MATCH_MATCH)
872          {          {
873          condition = TRUE;          condition = TRUE;
874          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
875          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
876          }          }
877        else if (rrc != MATCH_NOMATCH)        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
878          {          {
879          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
880          }          }
881        else        else
882          {          {
883          condition = FALSE;          condition = FALSE;
884          ecode += GET(ecode, 1);          ecode += codelink;
885          }          }
886        }        }
887    
888      /* We are now at the branch that is to be obeyed. As there is only one,      /* We are now at the branch that is to be obeyed. As there is only one,
889      we can use tail recursion to avoid using another stack frame. If the second      we can use tail recursion to avoid using another stack frame, except when
890      alternative doesn't exist, we can just plough on. */      match_cbegroup is required for an unlimited repeat of a possibly empty
891        group. If the second alternative doesn't exist, we can just plough on. */
892    
893      if (condition || *ecode == OP_ALT)      if (condition || *ecode == OP_ALT)
894        {        {
895        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
896        flags = match_tail_recursed | ((op == OP_SCOND)? match_cbegroup : 0);        if (op == OP_SCOND)        /* Possibly empty group */
897        goto TAIL_RECURSE;          {
898            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, match_cbegroup, RM49);
899            RRETURN(rrc);
900            }
901          else                       /* Group must match something */
902            {
903            flags = 0;
904            goto TAIL_RECURSE;
905            }
906        }        }
907      else      else                         /* Condition false & no alternative */
908        {        {
909        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
910        }        }
911      break;      break;
912    
913    
914      /* End of the pattern. If we are in a top-level recursion, we should      /* End of the pattern, either real or forced. If we are in a top-level
915      restore the offsets appropriately and continue from after the call. */      recursion, we should restore the offsets appropriately and continue from
916        after the call. */
917    
918        case OP_ACCEPT:
919      case OP_END:      case OP_END:
920      if (md->recursive != NULL && md->recursive->group_num == 0)      if (md->recursive != NULL && md->recursive->group_num == 0)
921        {        {
# Line 770  for (;;) Line 924  for (;;)
924        md->recursive = rec->prevrec;        md->recursive = rec->prevrec;
925        memmove(md->offset_vector, rec->offset_save,        memmove(md->offset_vector, rec->offset_save,
926          rec->saved_max * sizeof(int));          rec->saved_max * sizeof(int));
927        md->start_match = rec->save_start;        mstart = rec->save_start;
928        ims = original_ims;        ims = original_ims;
929        ecode = rec->after_call;        ecode = rec->after_call;
930        break;        break;
931        }        }
932    
933      /* 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
934      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
935        the subject. In both cases, backtracking will then try other alternatives,
936        if any. */
937    
938        if (eptr == mstart &&
939            (md->notempty ||
940              (md->notempty_atstart &&
941                mstart == md->start_subject + md->start_offset)))
942          RRETURN(MATCH_NOMATCH);
943    
944        /* Otherwise, we have a match. */
945    
946      if (md->notempty && eptr == md->start_match) RRETURN(MATCH_NOMATCH);      md->end_match_ptr = eptr;           /* Record where we ended */
947      md->end_match_ptr = eptr;          /* Record where we ended */      md->end_offset_top = offset_top;    /* and how many extracts were taken */
948      md->end_offset_top = offset_top;   /* and how many extracts were taken */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
949      RRETURN(MATCH_MATCH);      RRETURN(MATCH_MATCH);
950    
951      /* Change option settings */      /* Change option settings */
# Line 802  for (;;) Line 966  for (;;)
966      case OP_ASSERTBACK:      case OP_ASSERTBACK:
967      do      do
968        {        {
969        RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
970            RM4);
971        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
972        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
973        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
974        }        }
975      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 828  for (;;) Line 993  for (;;)
993      case OP_ASSERTBACK_NOT:      case OP_ASSERTBACK_NOT:
994      do      do
995        {        {
996        RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
997            RM5);
998        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);
999        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1000        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1001        }        }
1002      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 854  for (;;) Line 1020  for (;;)
1020          {          {
1021          eptr--;          eptr--;
1022          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1023          BACKCHAR(eptr)          BACKCHAR(eptr);
1024          }          }
1025        }        }
1026      else      else
# Line 867  for (;;) Line 1033  for (;;)
1033        if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);        if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1034        }        }
1035    
1036      /* Skip to next op code */      /* Save the earliest consulted character, then skip to next op code */
1037    
1038        if (eptr < md->start_used_ptr) md->start_used_ptr = eptr;
1039      ecode += 1 + LINK_SIZE;      ecode += 1 + LINK_SIZE;
1040      break;      break;
1041    
# Line 885  for (;;) Line 1052  for (;;)
1052        cb.offset_vector    = md->offset_vector;        cb.offset_vector    = md->offset_vector;
1053        cb.subject          = (PCRE_SPTR)md->start_subject;        cb.subject          = (PCRE_SPTR)md->start_subject;
1054        cb.subject_length   = md->end_subject - md->start_subject;        cb.subject_length   = md->end_subject - md->start_subject;
1055        cb.start_match      = md->start_match - md->start_subject;        cb.start_match      = mstart - md->start_subject;
1056        cb.current_position = eptr - md->start_subject;        cb.current_position = eptr - md->start_subject;
1057        cb.pattern_position = GET(ecode, 2);        cb.pattern_position = GET(ecode, 2);
1058        cb.next_item_length = GET(ecode, 2 + LINK_SIZE);        cb.next_item_length = GET(ecode, 2 + LINK_SIZE);
# Line 947  for (;;) Line 1114  for (;;)
1114    
1115        memcpy(new_recursive.offset_save, md->offset_vector,        memcpy(new_recursive.offset_save, md->offset_vector,
1116              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
1117        new_recursive.save_start = md->start_match;        new_recursive.save_start = mstart;
1118        md->start_match = eptr;        mstart = eptr;
1119    
1120        /* 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
1121        restore the offset and recursion data. */        restore the offset and recursion data. */
# Line 957  for (;;) Line 1124  for (;;)
1124        flags = (*callpat >= OP_SBRA)? match_cbegroup : 0;        flags = (*callpat >= OP_SBRA)? match_cbegroup : 0;
1125        do        do
1126          {          {
1127          RMATCH(rrc, eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,          RMATCH(eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,
1128            md, ims, eptrb, flags);            md, ims, eptrb, flags, RM6);
1129          if (rrc == MATCH_MATCH)          if (rrc == MATCH_MATCH)
1130            {            {
1131            DPRINTF(("Recursion matched\n"));            DPRINTF(("Recursion matched\n"));
# Line 967  for (;;) Line 1134  for (;;)
1134              (pcre_free)(new_recursive.offset_save);              (pcre_free)(new_recursive.offset_save);
1135            RRETURN(MATCH_MATCH);            RRETURN(MATCH_MATCH);
1136            }            }
1137          else if (rrc != MATCH_NOMATCH)          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1138            {            {
1139            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1140              if (new_recursive.offset_save != stacksave)
1141                (pcre_free)(new_recursive.offset_save);
1142            RRETURN(rrc);            RRETURN(rrc);
1143            }            }
1144    
# Line 1001  for (;;) Line 1170  for (;;)
1170    
1171      do      do
1172        {        {
1173        RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM7);
         eptrb, 0);  
1174        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
1175        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1176        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1177        }        }
1178      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 1047  for (;;) Line 1215  for (;;)
1215    
1216      if (*ecode == OP_KETRMIN)      if (*ecode == OP_KETRMIN)
1217        {        {
1218        RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM8);
1219        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1220        ecode = prev;        ecode = prev;
1221        flags = match_tail_recursed;        flags = 0;
1222        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1223        }        }
1224      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
1225        {        {
1226        RMATCH(rrc, eptr, prev, offset_top, md, ims, eptrb, match_cbegroup);        RMATCH(eptr, prev, offset_top, md, ims, eptrb, match_cbegroup, RM9);
1227        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1228        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1229        flags = match_tail_recursed;        flags = 0;
1230        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1231        }        }
1232      /* Control never gets here */      /* Control never gets here */
# Line 1070  for (;;) Line 1238  for (;;)
1238      do ecode += GET(ecode,1); while (*ecode == OP_ALT);      do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1239      break;      break;
1240    
1241      /* BRAZERO and BRAMINZERO occur just before a bracket group, indicating      /* BRAZERO, BRAMINZERO and SKIPZERO occur just before a bracket group,
1242      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
1243      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
1244      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
1245      preceded by BRAZERO or BRAMINZERO. */      optional ones preceded by BRAZERO or BRAMINZERO. */
1246    
1247      case OP_BRAZERO:      case OP_BRAZERO:
1248        {        {
1249        next = ecode+1;        next = ecode+1;
1250        RMATCH(rrc, eptr, next, offset_top, md, ims, eptrb, 0);        RMATCH(eptr, next, offset_top, md, ims, eptrb, 0, RM10);
1251        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1252        do next += GET(next,1); while (*next == OP_ALT);        do next += GET(next,1); while (*next == OP_ALT);
1253        ecode = next + 1 + LINK_SIZE;        ecode = next + 1 + LINK_SIZE;
# Line 1090  for (;;) Line 1258  for (;;)
1258        {        {
1259        next = ecode+1;        next = ecode+1;
1260        do next += GET(next, 1); while (*next == OP_ALT);        do next += GET(next, 1); while (*next == OP_ALT);
1261        RMATCH(rrc, eptr, next + 1+LINK_SIZE, offset_top, md, ims, eptrb, 0);        RMATCH(eptr, next + 1+LINK_SIZE, offset_top, md, ims, eptrb, 0, RM11);
1262        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1263        ecode++;        ecode++;
1264        }        }
1265      break;      break;
1266    
1267        case OP_SKIPZERO:
1268          {
1269          next = ecode+1;
1270          do next += GET(next,1); while (*next == OP_ALT);
1271          ecode = next + 1 + LINK_SIZE;
1272          }
1273        break;
1274    
1275      /* End of a group, repeated or non-repeating. */      /* End of a group, repeated or non-repeating. */
1276    
1277      case OP_KET:      case OP_KET:
# Line 1160  for (;;) Line 1336  for (;;)
1336          recursion_info *rec = md->recursive;          recursion_info *rec = md->recursive;
1337          DPRINTF(("Recursion (%d) succeeded - continuing\n", number));          DPRINTF(("Recursion (%d) succeeded - continuing\n", number));
1338          md->recursive = rec->prevrec;          md->recursive = rec->prevrec;
1339          md->start_match = rec->save_start;          mstart = rec->save_start;
1340          memcpy(md->offset_vector, rec->offset_save,          memcpy(md->offset_vector, rec->offset_save,
1341            rec->saved_max * sizeof(int));            rec->saved_max * sizeof(int));
1342          ecode = rec->after_call;          ecode = rec->after_call;
# Line 1189  for (;;) Line 1365  for (;;)
1365    
1366      /* The repeating kets try the rest of the pattern or restart from the      /* The repeating kets try the rest of the pattern or restart from the
1367      preceding bracket, in the appropriate order. In the second case, we can use      preceding bracket, in the appropriate order. In the second case, we can use
1368      tail recursion to avoid using another stack frame. */      tail recursion to avoid using another stack frame, unless we have an
1369        unlimited repeat of a group that can match an empty string. */
1370    
1371      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;      flags = (*prev >= OP_SBRA)? match_cbegroup : 0;
1372    
1373      if (*ecode == OP_KETRMIN)      if (*ecode == OP_KETRMIN)
1374        {        {
1375        RMATCH(rrc, eptr, ecode + 1+LINK_SIZE, offset_top, md, ims, eptrb, 0);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM12);
1376        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1377          if (flags != 0)    /* Could match an empty string */
1378            {
1379            RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM50);
1380            RRETURN(rrc);
1381            }
1382        ecode = prev;        ecode = prev;
       flags |= match_tail_recursed;  
1383        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1384        }        }
1385      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
1386        {        {
1387        RMATCH(rrc, eptr, prev, offset_top, md, ims, eptrb, flags);        RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);
1388        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1389        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
1390        flags = match_tail_recursed;        flags = 0;
1391        goto TAIL_RECURSE;        goto TAIL_RECURSE;
1392        }        }
1393      /* Control never gets here */      /* Control never gets here */
# Line 1239  for (;;) Line 1420  for (;;)
1420      ecode++;      ecode++;
1421      break;      break;
1422    
1423        /* Reset the start of match point */
1424    
1425        case OP_SET_SOM:
1426        mstart = eptr;
1427        ecode++;
1428        break;
1429    
1430      /* Assert before internal newline if multiline, or before a terminating      /* Assert before internal newline if multiline, or before a terminating
1431      newline unless endonly is set, else end of subject unless noteol is set. */      newline unless endonly is set, else end of subject unless noteol is set. */
1432    
# Line 1290  for (;;) Line 1478  for (;;)
1478    
1479        /* Find out if the previous and current characters are "word" characters.        /* Find out if the previous and current characters are "word" characters.
1480        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
1481        be "non-word" characters. */        be "non-word" characters. Remember the earliest consulted character for
1482          partial matching. */
1483    
1484  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
1485        if (utf8)        if (utf8)
1486          {          {
1487          if (eptr == md->start_subject) prev_is_word = FALSE; else          if (eptr == md->start_subject) prev_is_word = FALSE; else
1488            {            {
1489            const uschar *lastptr = eptr - 1;            USPTR lastptr = eptr - 1;
1490            while((*lastptr & 0xc0) == 0x80) lastptr--;            while((*lastptr & 0xc0) == 0x80) lastptr--;
1491              if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;
1492            GETCHAR(c, lastptr);            GETCHAR(c, lastptr);
1493            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;            prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
1494            }            }
1495          if (eptr >= md->end_subject) cur_is_word = FALSE; else          if (eptr >= md->end_subject)
1496              {
1497              SCHECK_PARTIAL();
1498              cur_is_word = FALSE;
1499              }
1500            else
1501            {            {
1502            GETCHAR(c, eptr);            GETCHAR(c, eptr);
1503            cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;            cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
# Line 1311  for (;;) Line 1506  for (;;)
1506        else        else
1507  #endif  #endif
1508    
1509        /* More streamlined when not in UTF-8 mode */        /* Not in UTF-8 mode */
1510    
1511          {          {
1512          prev_is_word = (eptr != md->start_subject) &&          if (eptr == md->start_subject) prev_is_word = FALSE; else
1513            ((md->ctypes[eptr[-1]] & ctype_word) != 0);            {
1514          cur_is_word = (eptr < md->end_subject) &&            if (eptr <= md->start_used_ptr) md->start_used_ptr = eptr - 1;
1515            ((md->ctypes[*eptr] & ctype_word) != 0);            prev_is_word = ((md->ctypes[eptr[-1]] & ctype_word) != 0);
1516              }
1517            if (eptr >= md->end_subject)
1518              {
1519              SCHECK_PARTIAL();
1520              cur_is_word = FALSE;
1521              }
1522            else cur_is_word = ((md->ctypes[*eptr] & ctype_word) != 0);
1523          }          }
1524    
1525        /* Now see if the situation is what we want */        /* Now see if the situation is what we want */
# Line 1331  for (;;) Line 1533  for (;;)
1533      /* Match a single character type; inline for speed */      /* Match a single character type; inline for speed */
1534    
1535      case OP_ANY:      case OP_ANY:
1536      if ((ims & PCRE_DOTALL) == 0)      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
1537        /* Fall through */
1538    
1539        case OP_ALLANY:
1540        if (eptr++ >= md->end_subject)
1541        {        {
1542        if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);        SCHECK_PARTIAL();
1543          RRETURN(MATCH_NOMATCH);
1544        }        }
1545      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++;  
1546      ecode++;      ecode++;
1547      break;      break;
1548    
# Line 1345  for (;;) Line 1550  for (;;)
1550      any byte, even newline, independent of the setting of PCRE_DOTALL. */      any byte, even newline, independent of the setting of PCRE_DOTALL. */
1551    
1552      case OP_ANYBYTE:      case OP_ANYBYTE:
1553      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr++ >= md->end_subject)
1554          {
1555          SCHECK_PARTIAL();
1556          RRETURN(MATCH_NOMATCH);
1557          }
1558      ecode++;      ecode++;
1559      break;      break;
1560    
1561      case OP_NOT_DIGIT:      case OP_NOT_DIGIT:
1562      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1563          {
1564          SCHECK_PARTIAL();
1565          RRETURN(MATCH_NOMATCH);
1566          }
1567      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1568      if (      if (
1569  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1363  for (;;) Line 1576  for (;;)
1576      break;      break;
1577    
1578      case OP_DIGIT:      case OP_DIGIT:
1579      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1580          {
1581          SCHECK_PARTIAL();
1582          RRETURN(MATCH_NOMATCH);
1583          }
1584      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1585      if (      if (
1586  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1376  for (;;) Line 1593  for (;;)
1593      break;      break;
1594    
1595      case OP_NOT_WHITESPACE:      case OP_NOT_WHITESPACE:
1596      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1597          {
1598          SCHECK_PARTIAL();
1599          RRETURN(MATCH_NOMATCH);
1600          }
1601      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1602      if (      if (
1603  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1389  for (;;) Line 1610  for (;;)
1610      break;      break;
1611    
1612      case OP_WHITESPACE:      case OP_WHITESPACE:
1613      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1614          {
1615          SCHECK_PARTIAL();
1616          RRETURN(MATCH_NOMATCH);
1617          }
1618      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1619      if (      if (
1620  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1402  for (;;) Line 1627  for (;;)
1627      break;      break;
1628    
1629      case OP_NOT_WORDCHAR:      case OP_NOT_WORDCHAR:
1630      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1631          {
1632          SCHECK_PARTIAL();
1633          RRETURN(MATCH_NOMATCH);
1634          }
1635      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1636      if (      if (
1637  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1415  for (;;) Line 1644  for (;;)
1644      break;      break;
1645    
1646      case OP_WORDCHAR:      case OP_WORDCHAR:
1647      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1648          {
1649          SCHECK_PARTIAL();
1650          RRETURN(MATCH_NOMATCH);
1651          }
1652      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1653      if (      if (
1654  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 1428  for (;;) Line 1661  for (;;)
1661      break;      break;
1662    
1663      case OP_ANYNL:      case OP_ANYNL:
1664      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1665          {
1666          SCHECK_PARTIAL();
1667          RRETURN(MATCH_NOMATCH);
1668          }
1669      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1670      switch(c)      switch(c)
1671        {        {
# Line 1436  for (;;) Line 1673  for (;;)
1673        case 0x000d:        case 0x000d:
1674        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
1675        break;        break;
1676    
1677        case 0x000a:        case 0x000a:
1678          break;
1679    
1680        case 0x000b:        case 0x000b:
1681        case 0x000c:        case 0x000c:
1682        case 0x0085:        case 0x0085:
1683        case 0x2028:        case 0x2028:
1684        case 0x2029:        case 0x2029:
1685          if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
1686          break;
1687          }
1688        ecode++;
1689        break;
1690    
1691        case OP_NOT_HSPACE:
1692        if (eptr >= md->end_subject)
1693          {
1694          SCHECK_PARTIAL();
1695          RRETURN(MATCH_NOMATCH);
1696          }
1697        GETCHARINCTEST(c, eptr);
1698        switch(c)
1699          {
1700          default: break;
1701          case 0x09:      /* HT */
1702          case 0x20:      /* SPACE */
1703          case 0xa0:      /* NBSP */
1704          case 0x1680:    /* OGHAM SPACE MARK */
1705          case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
1706          case 0x2000:    /* EN QUAD */
1707          case 0x2001:    /* EM QUAD */
1708          case 0x2002:    /* EN SPACE */
1709          case 0x2003:    /* EM SPACE */
1710          case 0x2004:    /* THREE-PER-EM SPACE */
1711          case 0x2005:    /* FOUR-PER-EM SPACE */
1712          case 0x2006:    /* SIX-PER-EM SPACE */
1713          case 0x2007:    /* FIGURE SPACE */
1714          case 0x2008:    /* PUNCTUATION SPACE */
1715          case 0x2009:    /* THIN SPACE */
1716          case 0x200A:    /* HAIR SPACE */
1717          case 0x202f:    /* NARROW NO-BREAK SPACE */
1718          case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
1719          case 0x3000:    /* IDEOGRAPHIC SPACE */
1720          RRETURN(MATCH_NOMATCH);
1721          }
1722        ecode++;
1723        break;
1724    
1725        case OP_HSPACE:
1726        if (eptr >= md->end_subject)
1727          {
1728          SCHECK_PARTIAL();
1729          RRETURN(MATCH_NOMATCH);
1730          }
1731        GETCHARINCTEST(c, eptr);
1732        switch(c)
1733          {
1734          default: RRETURN(MATCH_NOMATCH);
1735          case 0x09:      /* HT */
1736          case 0x20:      /* SPACE */
1737          case 0xa0:      /* NBSP */
1738          case 0x1680:    /* OGHAM SPACE MARK */
1739          case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
1740          case 0x2000:    /* EN QUAD */
1741          case 0x2001:    /* EM QUAD */
1742          case 0x2002:    /* EN SPACE */
1743          case 0x2003:    /* EM SPACE */
1744          case 0x2004:    /* THREE-PER-EM SPACE */
1745          case 0x2005:    /* FOUR-PER-EM SPACE */
1746          case 0x2006:    /* SIX-PER-EM SPACE */
1747          case 0x2007:    /* FIGURE SPACE */
1748          case 0x2008:    /* PUNCTUATION SPACE */
1749          case 0x2009:    /* THIN SPACE */
1750          case 0x200A:    /* HAIR SPACE */
1751          case 0x202f:    /* NARROW NO-BREAK SPACE */
1752          case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
1753          case 0x3000:    /* IDEOGRAPHIC SPACE */
1754          break;
1755          }
1756        ecode++;
1757        break;
1758    
1759        case OP_NOT_VSPACE:
1760        if (eptr >= md->end_subject)
1761          {
1762          SCHECK_PARTIAL();
1763          RRETURN(MATCH_NOMATCH);
1764          }
1765        GETCHARINCTEST(c, eptr);
1766        switch(c)
1767          {
1768          default: break;
1769          case 0x0a:      /* LF */
1770          case 0x0b:      /* VT */
1771          case 0x0c:      /* FF */
1772          case 0x0d:      /* CR */
1773          case 0x85:      /* NEL */
1774          case 0x2028:    /* LINE SEPARATOR */
1775          case 0x2029:    /* PARAGRAPH SEPARATOR */
1776          RRETURN(MATCH_NOMATCH);
1777          }
1778        ecode++;
1779        break;
1780    
1781        case OP_VSPACE:
1782        if (eptr >= md->end_subject)
1783          {
1784          SCHECK_PARTIAL();
1785          RRETURN(MATCH_NOMATCH);
1786          }
1787        GETCHARINCTEST(c, eptr);
1788        switch(c)
1789          {
1790          default: RRETURN(MATCH_NOMATCH);
1791          case 0x0a:      /* LF */
1792          case 0x0b:      /* VT */
1793          case 0x0c:      /* FF */
1794          case 0x0d:      /* CR */
1795          case 0x85:      /* NEL */
1796          case 0x2028:    /* LINE SEPARATOR */
1797          case 0x2029:    /* PARAGRAPH SEPARATOR */
1798        break;        break;
1799        }        }
1800      ecode++;      ecode++;
# Line 1453  for (;;) Line 1806  for (;;)
1806    
1807      case OP_PROP:      case OP_PROP:
1808      case OP_NOTPROP:      case OP_NOTPROP:
1809      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1810          {
1811          SCHECK_PARTIAL();
1812          RRETURN(MATCH_NOMATCH);
1813          }
1814      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1815        {        {
1816        int chartype, script;        const ucd_record *prop = GET_UCD(c);
       int category = _pcre_ucp_findprop(c, &chartype, &script);  
1817    
1818        switch(ecode[1])        switch(ecode[1])
1819          {          {
# Line 1466  for (;;) Line 1822  for (;;)
1822          break;          break;
1823    
1824          case PT_LAMP:          case PT_LAMP:
1825          if ((chartype == ucp_Lu ||          if ((prop->chartype == ucp_Lu ||
1826               chartype == ucp_Ll ||               prop->chartype == ucp_Ll ||
1827               chartype == ucp_Lt) == (op == OP_NOTPROP))               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))
1828            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1829           break;           break;
1830    
1831          case PT_GC:          case PT_GC:
1832          if ((ecode[2] != category) == (op == OP_PROP))          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))
1833            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1834          break;          break;
1835    
1836          case PT_PC:          case PT_PC:
1837          if ((ecode[2] != chartype) == (op == OP_PROP))          if ((ecode[2] != prop->chartype) == (op == OP_PROP))
1838            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1839          break;          break;
1840    
1841          case PT_SC:          case PT_SC:
1842          if ((ecode[2] != script) == (op == OP_PROP))          if ((ecode[2] != prop->script) == (op == OP_PROP))
1843            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
1844          break;          break;
1845    
# Line 1499  for (;;) Line 1855  for (;;)
1855      is in the binary; otherwise a compile-time error occurs. */      is in the binary; otherwise a compile-time error occurs. */
1856    
1857      case OP_EXTUNI:      case OP_EXTUNI:
1858      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
1859          {
1860          SCHECK_PARTIAL();
1861          RRETURN(MATCH_NOMATCH);
1862          }
1863      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
1864        {        {
1865        int chartype, script;        int category = UCD_CATEGORY(c);
       int category = _pcre_ucp_findprop(c, &chartype, &script);  
1866        if (category == ucp_M) RRETURN(MATCH_NOMATCH);        if (category == ucp_M) RRETURN(MATCH_NOMATCH);
1867        while (eptr < md->end_subject)        while (eptr < md->end_subject)
1868          {          {
# Line 1512  for (;;) Line 1871  for (;;)
1871            {            {
1872            GETCHARLEN(c, eptr, len);            GETCHARLEN(c, eptr, len);
1873            }            }
1874          category = _pcre_ucp_findprop(c, &chartype, &script);          category = UCD_CATEGORY(c);
1875          if (category != ucp_M) break;          if (category != ucp_M) break;
1876          eptr += len;          eptr += len;
1877          }          }
# Line 1533  for (;;) Line 1892  for (;;)
1892      case OP_REF:      case OP_REF:
1893        {        {
1894        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
1895        ecode += 3;                                 /* Advance past item */        ecode += 3;
1896    
1897          /* If the reference is unset, there are two possibilities:
1898    
1899        /* If the reference is unset, set the length to be longer than the amount        (a) In the default, Perl-compatible state, set the length to be longer
1900        of subject left; this ensures that every attempt at a match fails. We        than the amount of subject left; this ensures that every attempt at a
1901        can't just fail here, because of the possibility of quantifiers with zero        match fails. We can't just fail here, because of the possibility of
1902        minima. */        quantifiers with zero minima.
1903    
1904        length = (offset >= offset_top || md->offset_vector[offset] < 0)?        (b) If the JavaScript compatibility flag is set, set the length to zero
1905          md->end_subject - eptr + 1 :        so that the back reference matches an empty string.
1906          md->offset_vector[offset+1] - md->offset_vector[offset];  
1907          Otherwise, set the length to the length of what was matched by the
1908          referenced subpattern. */
1909    
1910          if (offset >= offset_top || md->offset_vector[offset] < 0)
1911            length = (md->jscript_compat)? 0 : md->end_subject - eptr + 1;
1912          else
1913            length = md->offset_vector[offset+1] - md->offset_vector[offset];
1914    
1915        /* Set up for repetition, or handle the non-repeated case */        /* Set up for repetition, or handle the non-repeated case */
1916    
# Line 1571  for (;;) Line 1939  for (;;)
1939          break;          break;
1940    
1941          default:               /* No repeat follows */          default:               /* No repeat follows */
1942          if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH);          if (!match_ref(offset, eptr, length, md, ims))
1943              {
1944              CHECK_PARTIAL();
1945              RRETURN(MATCH_NOMATCH);
1946              }
1947          eptr += length;          eptr += length;
1948          continue;              /* With the main loop */          continue;              /* With the main loop */
1949          }          }
# Line 1587  for (;;) Line 1959  for (;;)
1959    
1960        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
1961          {          {
1962          if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH);          if (!match_ref(offset, eptr, length, md, ims))
1963              {
1964              CHECK_PARTIAL();
1965              RRETURN(MATCH_NOMATCH);
1966              }
1967          eptr += length;          eptr += length;
1968          }          }
1969    
# Line 1602  for (;;) Line 1978  for (;;)
1978          {          {
1979          for (fi = min;; fi++)          for (fi = min;; fi++)
1980            {            {
1981            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);
1982            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1983            if (fi >= max || !match_ref(offset, eptr, length, md, ims))            if (fi >= max) RRETURN(MATCH_NOMATCH);
1984              if (!match_ref(offset, eptr, length, md, ims))
1985                {
1986                CHECK_PARTIAL();
1987              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
1988                }
1989            eptr += length;            eptr += length;
1990            }            }
1991          /* Control never gets here */          /* Control never gets here */
# Line 1623  for (;;) Line 2003  for (;;)
2003            }            }
2004          while (eptr >= pp)          while (eptr >= pp)
2005            {            {
2006            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM15);
2007            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2008            eptr -= length;            eptr -= length;
2009            }            }
# Line 1632  for (;;) Line 2012  for (;;)
2012        }        }
2013      /* Control never gets here */      /* Control never gets here */
2014    
   
   
2015      /* Match a bit-mapped character class, possibly repeatedly. This op code is      /* Match a bit-mapped character class, possibly repeatedly. This op code is
2016      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,
2017      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 1688  for (;;) Line 2066  for (;;)
2066          {          {
2067          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2068            {            {
2069            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
2070                {
2071                SCHECK_PARTIAL();
2072                RRETURN(MATCH_NOMATCH);
2073                }
2074            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
2075            if (c > 255)            if (c > 255)
2076              {              {
# Line 1706  for (;;) Line 2088  for (;;)
2088          {          {
2089          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2090            {            {
2091            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
2092                {
2093                SCHECK_PARTIAL();
2094                RRETURN(MATCH_NOMATCH);
2095                }
2096            c = *eptr++;            c = *eptr++;
2097            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);            if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2098            }            }
# Line 1728  for (;;) Line 2114  for (;;)
2114            {            {
2115            for (fi = min;; fi++)            for (fi = min;; fi++)
2116              {              {
2117              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);
2118              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2119              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2120                if (eptr >= md->end_subject)
2121                  {
2122                  SCHECK_PARTIAL();
2123                  RRETURN(MATCH_NOMATCH);
2124                  }
2125              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
2126              if (c > 255)              if (c > 255)
2127                {                {
# Line 1748  for (;;) Line 2139  for (;;)
2139            {            {
2140            for (fi = min;; fi++)            for (fi = min;; fi++)
2141              {              {
2142              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);
2143              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2144              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2145                if (eptr >= md->end_subject)
2146                  {
2147                  SCHECK_PARTIAL();
2148                  RRETURN(MATCH_NOMATCH);
2149                  }
2150              c = *eptr++;              c = *eptr++;
2151              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);              if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2152              }              }
# Line 1785  for (;;) Line 2181  for (;;)
2181              }              }
2182            for (;;)            for (;;)
2183              {              {
2184              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM18);
2185              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2186              if (eptr-- == pp) break;        /* Stop if tried at original pos */              if (eptr-- == pp) break;        /* Stop if tried at original pos */
2187              BACKCHAR(eptr);              BACKCHAR(eptr);
# Line 1804  for (;;) Line 2200  for (;;)
2200              }              }
2201            while (eptr >= pp)            while (eptr >= pp)
2202              {              {
2203              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM19);
2204              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2205              eptr--;              eptr--;
2206              }              }
# Line 1817  for (;;) Line 2213  for (;;)
2213    
2214    
2215      /* Match an extended character class. This opcode is encountered only      /* Match an extended character class. This opcode is encountered only
2216      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
2217        mode, because Unicode properties are supported in non-UTF-8 mode. */
2218    
2219  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
2220      case OP_XCLASS:      case OP_XCLASS:
# Line 1858  for (;;) Line 2255  for (;;)
2255    
2256        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2257          {          {
2258          if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);          if (eptr >= md->end_subject)
2259          GETCHARINC(c, eptr);            {
2260              SCHECK_PARTIAL();
2261              RRETURN(MATCH_NOMATCH);
2262              }
2263            GETCHARINCTEST(c, eptr);
2264          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);          if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);
2265          }          }
2266    
# Line 1875  for (;;) Line 2276  for (;;)
2276          {          {
2277          for (fi = min;; fi++)          for (fi = min;; fi++)
2278            {            {
2279            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);
2280            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2281            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
2282            GETCHARINC(c, eptr);            if (eptr >= md->end_subject)
2283                {
2284                SCHECK_PARTIAL();
2285                RRETURN(MATCH_NOMATCH);
2286                }
2287              GETCHARINCTEST(c, eptr);
2288            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);            if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH);
2289            }            }
2290          /* Control never gets here */          /* Control never gets here */
# Line 1893  for (;;) Line 2299  for (;;)
2299            {            {
2300            int len = 1;            int len = 1;
2301            if (eptr >= md->end_subject) break;            if (eptr >= md->end_subject) break;
2302            GETCHARLEN(c, eptr, len);            GETCHARLENTEST(c, eptr, len);
2303            if (!_pcre_xclass(c, data)) break;            if (!_pcre_xclass(c, data)) break;
2304            eptr += len;            eptr += len;
2305            }            }
2306          for(;;)          for(;;)
2307            {            {
2308            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);
2309            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2310            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
2311            BACKCHAR(eptr)            if (utf8) BACKCHAR(eptr);
2312            }            }
2313          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2314          }          }
# Line 1920  for (;;) Line 2326  for (;;)
2326        length = 1;        length = 1;
2327        ecode++;        ecode++;
2328        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
2329        if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);        if (length > md->end_subject - eptr)
2330            {
2331            CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
2332            RRETURN(MATCH_NOMATCH);
2333            }
2334        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);
2335        }        }
2336      else      else
# Line 1928  for (;;) Line 2338  for (;;)
2338    
2339      /* Non-UTF-8 mode */      /* Non-UTF-8 mode */
2340        {        {
2341        if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH);        if (md->end_subject - eptr < 1)
2342            {
2343            SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
2344            RRETURN(MATCH_NOMATCH);
2345            }
2346        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);
2347        ecode += 2;        ecode += 2;
2348        }        }
# Line 1944  for (;;) Line 2358  for (;;)
2358        ecode++;        ecode++;
2359        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
2360    
2361        if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);        if (length > md->end_subject - eptr)
2362            {
2363            CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
2364            RRETURN(MATCH_NOMATCH);
2365            }
2366    
2367        /* 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
2368        can use the fast lookup table. */        can use the fast lookup table. */
# Line 1968  for (;;) Line 2386  for (;;)
2386          if (fc != dc)          if (fc != dc)
2387            {            {
2388  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2389            if (dc != _pcre_ucp_othercase(fc))            if (dc != UCD_OTHERCASE(fc))
2390  #endif  #endif
2391              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2392            }            }
# Line 1979  for (;;) Line 2397  for (;;)
2397    
2398      /* Non-UTF-8 mode */      /* Non-UTF-8 mode */
2399        {        {
2400        if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH);        if (md->end_subject - eptr < 1)
2401            {
2402            SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
2403            RRETURN(MATCH_NOMATCH);
2404            }
2405        if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);        if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2406        ecode += 2;        ecode += 2;
2407        }        }
# Line 2033  for (;;) Line 2455  for (;;)
2455      case OP_MINQUERY:      case OP_MINQUERY:
2456      c = *ecode++ - OP_STAR;      c = *ecode++ - OP_STAR;
2457      minimize = (c & 1) != 0;      minimize = (c & 1) != 0;
2458    
2459      min = rep_min[c];                 /* Pick up values from tables; */      min = rep_min[c];                 /* Pick up values from tables; */
2460      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
2461      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
2462    
2463      /* 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. */  
2464    
2465      REPEATCHAR:      REPEATCHAR:
2466  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
# Line 2048  for (;;) Line 2469  for (;;)
2469        length = 1;        length = 1;
2470        charptr = ecode;        charptr = ecode;
2471        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
       if (min * length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
2472        ecode += length;        ecode += length;
2473    
2474        /* Handle multibyte character matching specially here. There is        /* Handle multibyte character matching specially here. There is
# Line 2059  for (;;) Line 2479  for (;;)
2479  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2480          unsigned int othercase;          unsigned int othercase;
2481          if ((ims & PCRE_CASELESS) != 0 &&          if ((ims & PCRE_CASELESS) != 0 &&
2482              (othercase = _pcre_ucp_othercase(fc)) != NOTACHAR)              (othercase = UCD_OTHERCASE(fc)) != fc)
2483            oclength = _pcre_ord2utf8(othercase, occhars);            oclength = _pcre_ord2utf8(othercase, occhars);
2484          else oclength = 0;          else oclength = 0;
2485  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2486    
2487          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2488            {            {
2489            if (memcmp(eptr, charptr, length) == 0) eptr += length;            if (eptr <= md->end_subject - length &&
2490                memcmp(eptr, charptr, length) == 0) eptr += length;
2491  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2492            /* Need braces because of following else */            else if (oclength > 0 &&
2493            else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                     eptr <= md->end_subject - oclength &&
2494                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2495    #endif  /* SUPPORT_UCP */
2496            else            else
2497              {              {
2498              if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);              CHECK_PARTIAL();
2499              eptr += oclength;              RRETURN(MATCH_NOMATCH);
2500              }              }
 #else   /* without SUPPORT_UCP */  
           else { RRETURN(MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2501            }            }
2502    
2503          if (min == max) continue;          if (min == max) continue;
# Line 2086  for (;;) Line 2506  for (;;)
2506            {            {
2507            for (fi = min;; fi++)            for (fi = min;; fi++)
2508              {              {
2509              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);
2510              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2511              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2512              if (memcmp(eptr, charptr, length) == 0) eptr += length;              if (eptr <= md->end_subject - length &&
2513                  memcmp(eptr, charptr, length) == 0) eptr += length;
2514  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2515              /* Need braces because of following else */              else if (oclength > 0 &&
2516              else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }                       eptr <= md->end_subject - oclength &&
2517                         memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
2518    #endif  /* SUPPORT_UCP */
2519              else              else
2520                {                {
2521                if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);                CHECK_PARTIAL();
2522                eptr += oclength;                RRETURN(MATCH_NOMATCH);
2523                }                }
 #else   /* without SUPPORT_UCP */  
             else { RRETURN (MATCH_NOMATCH); }  
 #endif  /* SUPPORT_UCP */  
2524              }              }
2525            /* Control never gets here */            /* Control never gets here */
2526            }            }
# Line 2110  for (;;) Line 2530  for (;;)
2530            pp = eptr;            pp = eptr;
2531            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2532              {              {
2533              if (eptr > md->end_subject - length) break;              if (eptr <= md->end_subject - length &&
2534              if (memcmp(eptr, charptr, length) == 0) eptr += length;                  memcmp(eptr, charptr, length) == 0) eptr += length;
2535  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2536              else if (oclength == 0) break;              else if (oclength > 0 &&
2537              else                       eptr <= md->end_subject - oclength &&
2538                {                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;
               if (memcmp(eptr, occhars, oclength) != 0) break;  
               eptr += oclength;  
               }  
 #else   /* without SUPPORT_UCP */  
             else break;  
2539  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2540                else break;
2541              }              }
2542    
2543            if (possessive) continue;            if (possessive) continue;
2544    
2545            for(;;)            for(;;)
2546             {              {
2547             RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);
2548             if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2549             if (eptr == pp) RRETURN(MATCH_NOMATCH);              if (eptr == pp) { RRETURN(MATCH_NOMATCH); }
2550  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2551             eptr--;              eptr--;
2552             BACKCHAR(eptr);              BACKCHAR(eptr);
2553  #else   /* without SUPPORT_UCP */  #else   /* without SUPPORT_UCP */
2554             eptr -= length;              eptr -= length;
2555  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2556             }              }
2557            }            }
2558          /* Control never gets here */          /* Control never gets here */
2559          }          }
# Line 2149  for (;;) Line 2566  for (;;)
2566  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF8 */
2567    
2568      /* When not in UTF-8 mode, load a single-byte character. */      /* When not in UTF-8 mode, load a single-byte character. */
2569        {  
2570        if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);      fc = *ecode++;
       fc = *ecode++;  
       }  
2571    
2572      /* 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
2573      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 2170  for (;;) Line 2585  for (;;)
2585        {        {
2586        fc = md->lcc[fc];        fc = md->lcc[fc];
2587        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
2588            {
2589            if (eptr >= md->end_subject)
2590              {
2591              SCHECK_PARTIAL();
2592              RRETURN(MATCH_NOMATCH);
2593              }
2594          if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);          if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2595            }
2596        if (min == max) continue;        if (min == max) continue;
2597        if (minimize)        if (minimize)
2598          {          {
2599          for (fi = min;; fi++)          for (fi = min;; fi++)
2600            {            {
2601            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);
2602            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2603            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) RRETURN(MATCH_NOMATCH);
2604                fc != md->lcc[*eptr++])            if (eptr >= md->end_subject)
2605                {
2606                SCHECK_PARTIAL();
2607              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2608                }
2609              if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2610            }            }
2611          /* Control never gets here */          /* Control never gets here */
2612          }          }
# Line 2192  for (;;) Line 2618  for (;;)
2618            if (eptr >= md->end_subject || fc != md->lcc[*eptr]) break;            if (eptr >= md->end_subject || fc != md->lcc[*eptr]) break;
2619            eptr++;            eptr++;
2620            }            }
2621    
2622          if (possessive) continue;          if (possessive) continue;
2623    
2624          while (eptr >= pp)          while (eptr >= pp)
2625            {            {
2626            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);
2627            eptr--;            eptr--;
2628            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2629            }            }
# Line 2208  for (;;) Line 2636  for (;;)
2636    
2637      else      else
2638        {        {
2639        for (i = 1; i <= min; i++) if (fc != *eptr++) RRETURN(MATCH_NOMATCH);        for (i = 1; i <= min; i++)
2640            {
2641            if (eptr >= md->end_subject)
2642              {
2643              SCHECK_PARTIAL();
2644              RRETURN(MATCH_NOMATCH);
2645              }
2646            if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
2647            }
2648    
2649        if (min == max) continue;        if (min == max) continue;
2650    
2651        if (minimize)        if (minimize)
2652          {          {
2653          for (fi = min;; fi++)          for (fi = min;; fi++)
2654            {            {
2655            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);
2656            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2657            if (fi >= max || eptr >= md->end_subject || fc != *eptr++)            if (fi >= max) RRETURN(MATCH_NOMATCH);
2658              if (eptr >= md->end_subject)
2659                {
2660                SCHECK_PARTIAL();
2661              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2662                }
2663              if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
2664            }            }
2665          /* Control never gets here */          /* Control never gets here */
2666          }          }
# Line 2230  for (;;) Line 2673  for (;;)
2673            eptr++;            eptr++;
2674            }            }
2675          if (possessive) continue;          if (possessive) continue;
2676    
2677          while (eptr >= pp)          while (eptr >= pp)
2678            {            {
2679            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27);
2680            eptr--;            eptr--;
2681            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2682            }            }
# Line 2245  for (;;) Line 2689  for (;;)
2689      checking can be multibyte. */      checking can be multibyte. */
2690    
2691      case OP_NOT:      case OP_NOT:
2692      if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr >= md->end_subject)
2693          {
2694          SCHECK_PARTIAL();
2695          RRETURN(MATCH_NOMATCH);
2696          }
2697      ecode++;      ecode++;
2698      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2699      if ((ims & PCRE_CASELESS) != 0)      if ((ims & PCRE_CASELESS) != 0)
# Line 2322  for (;;) Line 2770  for (;;)
2770      max = rep_max[c];                 /* zero for max => infinity */      max = rep_max[c];                 /* zero for max => infinity */
2771      if (max == 0) max = INT_MAX;      if (max == 0) max = INT_MAX;
2772    
2773      /* 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. */  
2774    
2775      REPEATNOTCHAR:      REPEATNOTCHAR:
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
2776      fc = *ecode++;      fc = *ecode++;
2777    
2778      /* 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 2352  for (;;) Line 2797  for (;;)
2797          register unsigned int d;          register unsigned int d;
2798          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2799            {            {
2800              if (eptr >= md->end_subject)
2801                {
2802                SCHECK_PARTIAL();
2803                RRETURN(MATCH_NOMATCH);
2804                }
2805            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
2806            if (d < 256) d = md->lcc[d];            if (d < 256) d = md->lcc[d];
2807            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
# Line 2363  for (;;) Line 2813  for (;;)
2813        /* Not UTF-8 mode */        /* Not UTF-8 mode */
2814          {          {
2815          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2816              {
2817              if (eptr >= md->end_subject)
2818                {
2819                SCHECK_PARTIAL();
2820                RRETURN(MATCH_NOMATCH);
2821                }
2822            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);            if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2823              }
2824          }          }
2825    
2826        if (min == max) continue;        if (min == max) continue;
# Line 2377  for (;;) Line 2834  for (;;)
2834            register unsigned int d;            register unsigned int d;
2835            for (fi = min;; fi++)            for (fi = min;; fi++)
2836              {              {
2837              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);
2838              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2839                if (fi >= max) RRETURN(MATCH_NOMATCH);
2840                if (eptr >= md->end_subject)
2841                  {
2842                  SCHECK_PARTIAL();
2843                  RRETURN(MATCH_NOMATCH);
2844                  }
2845              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
2846              if (d < 256) d = md->lcc[d];              if (d < 256) d = md->lcc[d];
2847              if (fi >= max || eptr >= md->end_subject || fc == d)              if (fc == d) RRETURN(MATCH_NOMATCH);
               RRETURN(MATCH_NOMATCH);  
2848              }              }
2849            }            }
2850          else          else
# Line 2391  for (;;) Line 2853  for (;;)
2853            {            {
2854            for (fi = min;; fi++)            for (fi = min;; fi++)
2855              {              {
2856              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);
2857              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2858              if (fi >= max || eptr >= md->end_subject || fc == md->lcc[*eptr++])              if (fi >= max) RRETURN(MATCH_NOMATCH);
2859                if (eptr >= md->end_subject)
2860                  {
2861                  SCHECK_PARTIAL();
2862                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2863                  }
2864                if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH);
2865              }              }
2866            }            }
2867          /* Control never gets here */          /* Control never gets here */
# Line 2423  for (;;) Line 2890  for (;;)
2890          if (possessive) continue;          if (possessive) continue;
2891          for(;;)          for(;;)
2892              {              {
2893              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM30);
2894              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2895              if (eptr-- == pp) break;        /* Stop if tried at original pos */              if (eptr-- == pp) break;        /* Stop if tried at original pos */
2896              BACKCHAR(eptr);              BACKCHAR(eptr);
# Line 2441  for (;;) Line 2908  for (;;)
2908            if (possessive) continue;            if (possessive) continue;
2909            while (eptr >= pp)            while (eptr >= pp)
2910              {              {
2911              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM31);
2912              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2913              eptr--;              eptr--;
2914              }              }
# Line 2463  for (;;) Line 2930  for (;;)
2930          register unsigned int d;          register unsigned int d;
2931          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2932            {            {
2933              if (eptr >= md->end_subject)
2934                {
2935                SCHECK_PARTIAL();
2936                RRETURN(MATCH_NOMATCH);
2937                }
2938            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
2939            if (fc == d) RRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
2940            }            }
# Line 2472  for (;;) Line 2944  for (;;)
2944        /* Not UTF-8 mode */        /* Not UTF-8 mode */
2945          {          {
2946          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2947              {
2948              if (eptr >= md->end_subject)
2949                {
2950                SCHECK_PARTIAL();
2951                RRETURN(MATCH_NOMATCH);
2952                }
2953            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
2954              }
2955          }          }
2956    
2957        if (min == max) continue;        if (min == max) continue;
# Line 2486  for (;;) Line 2965  for (;;)
2965            register unsigned int d;            register unsigned int d;
2966            for (fi = min;; fi++)            for (fi = min;; fi++)
2967              {              {
2968              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);
2969              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2970              GETCHARINC(d, eptr);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2971              if (fi >= max || eptr >= md->end_subject || fc == d)              if (eptr >= md->end_subject)
2972                  {
2973                  SCHECK_PARTIAL();
2974                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2975                  }
2976                GETCHARINC(d, eptr);
2977                if (fc == d) RRETURN(MATCH_NOMATCH);
2978              }              }
2979            }            }
2980          else          else
# Line 2499  for (;;) Line 2983  for (;;)
2983            {            {
2984            for (fi = min;; fi++)            for (fi = min;; fi++)
2985              {              {
2986              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);
2987              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2988              if (fi >= max || eptr >= md->end_subject || fc == *eptr++)              if (fi >= max) RRETURN(MATCH_NOMATCH);
2989                if (eptr >= md->end_subject)
2990                  {
2991                  SCHECK_PARTIAL();
2992                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2993                  }
2994                if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
2995              }              }
2996            }            }
2997          /* Control never gets here */          /* Control never gets here */
# Line 2530  for (;;) Line 3019  for (;;)
3019            if (possessive) continue;            if (possessive) continue;
3020            for(;;)            for(;;)
3021              {              {
3022              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM34);
3023              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3024              if (eptr-- == pp) break;        /* Stop if tried at original pos */              if (eptr-- == pp) break;        /* Stop if tried at original pos */
3025              BACKCHAR(eptr);              BACKCHAR(eptr);
# Line 2548  for (;;) Line 3037  for (;;)
3037            if (possessive) continue;            if (possessive) continue;
3038            while (eptr >= pp)            while (eptr >= pp)
3039              {              {
3040              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM35);
3041              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3042              eptr--;              eptr--;
3043              }              }
# Line 2636  for (;;) Line 3125  for (;;)
3125    
3126      /* First, ensure the minimum number of matches are present. Use inline      /* First, ensure the minimum number of matches are present. Use inline
3127      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
3128      (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  
3129      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
3130      and single-bytes. */      and single-bytes. */
3131    
     if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH);  
3132      if (min > 0)      if (min > 0)
3133        {        {
3134  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 2654  for (;;) Line 3140  for (;;)
3140            if (prop_fail_result) RRETURN(MATCH_NOMATCH);            if (prop_fail_result) RRETURN(MATCH_NOMATCH);
3141            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3142              {              {
3143              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3144              GETCHARINC(c, eptr);                {
3145                  SCHECK_PARTIAL();
3146                  RRETURN(MATCH_NOMATCH);
3147                  }
3148                GETCHARINCTEST(c, eptr);
3149              }              }
3150            break;            break;
3151    
3152            case PT_LAMP:            case PT_LAMP:
3153            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3154              {              {
3155              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3156              GETCHARINC(c, eptr);                {
3157              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3158                  RRETURN(MATCH_NOMATCH);
3159                  }
3160                GETCHARINCTEST(c, eptr);
3161                prop_chartype = UCD_CHARTYPE(c);
3162              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
3163                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
3164                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 2675  for (;;) Line 3169  for (;;)
3169            case PT_GC:            case PT_GC:
3170            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3171              {              {
3172              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3173              GETCHARINC(c, eptr);                {
3174              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3175                  RRETURN(MATCH_NOMATCH);
3176                  }
3177                GETCHARINCTEST(c, eptr);
3178                prop_category = UCD_CATEGORY(c);
3179              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
3180                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3181              }              }
# Line 2686  for (;;) Line 3184  for (;;)
3184            case PT_PC:            case PT_PC:
3185            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3186              {              {
3187              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3188              GETCHARINC(c, eptr);                {
3189              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3190                  RRETURN(MATCH_NOMATCH);
3191                  }
3192                GETCHARINCTEST(c, eptr);
3193                prop_chartype = UCD_CHARTYPE(c);
3194              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
3195                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3196              }              }
# Line 2697  for (;;) Line 3199  for (;;)
3199            case PT_SC:            case PT_SC:
3200            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
3201              {              {
3202              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (eptr >= md->end_subject)
3203              GETCHARINC(c, eptr);                {
3204              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);                SCHECK_PARTIAL();
3205                  RRETURN(MATCH_NOMATCH);
3206                  }
3207                GETCHARINCTEST(c, eptr);
3208                prop_script = UCD_SCRIPT(c);
3209              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
3210                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3211              }              }
# Line 2717  for (;;) Line 3223  for (;;)
3223          {          {
3224          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3225            {            {
3226              if (eptr >= md->end_subject)
3227                {
3228                SCHECK_PARTIAL();
3229                RRETURN(MATCH_NOMATCH);
3230                }
3231            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3232            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
3233            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);
3234            while (eptr < md->end_subject)            while (eptr < md->end_subject)
3235              {              {
3236              int len = 1;              int len = 1;
3237              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
3238                {                else { GETCHARLEN(c, eptr, len); }
3239                GETCHARLEN(c, eptr, len);              prop_category = UCD_CATEGORY(c);
               }  
             prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);  
3240              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
3241              eptr += len;              eptr += len;
3242              }              }
# Line 2745  for (;;) Line 3254  for (;;)
3254          case OP_ANY:          case OP_ANY:
3255          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3256            {            {
3257            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3258                 ((ims & PCRE_DOTALL) == 0 && IS_NEWLINE(eptr)))              {
3259                SCHECK_PARTIAL();
3260                RRETURN(MATCH_NOMATCH);
3261                }
3262              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
3263              eptr++;
3264              while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3265              }
3266            break;
3267    
3268            case OP_ALLANY:
3269            for (i = 1; i <= min; i++)
3270              {
3271              if (eptr >= md->end_subject)
3272                {
3273                SCHECK_PARTIAL();
3274              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3275                }
3276            eptr++;            eptr++;
3277            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3278            }            }
3279          break;          break;
3280    
3281          case OP_ANYBYTE:          case OP_ANYBYTE:
3282            if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
3283          eptr += min;          eptr += min;
3284          break;          break;
3285    
3286          case OP_ANYNL:          case OP_ANYNL:
3287          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3288            {            {
3289            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3290                {
3291                SCHECK_PARTIAL();
3292                RRETURN(MATCH_NOMATCH);
3293                }
3294            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3295            switch(c)            switch(c)
3296              {              {
# Line 2768  for (;;) Line 3298  for (;;)
3298              case 0x000d:              case 0x000d:
3299              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3300              break;              break;
3301    
3302              case 0x000a:              case 0x000a:
3303                break;
3304    
3305              case 0x000b:              case 0x000b:
3306              case 0x000c:              case 0x000c:
3307              case 0x0085:              case 0x0085:
3308              case 0x2028:              case 0x2028:
3309              case 0x2029:              case 0x2029:
3310                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3311                break;
3312                }
3313              }
3314            break;
3315    
3316            case OP_NOT_HSPACE:
3317            for (i = 1; i <= min; i++)
3318              {
3319              if (eptr >= md->end_subject)
3320                {
3321                SCHECK_PARTIAL();
3322                RRETURN(MATCH_NOMATCH);
3323                }
3324              GETCHARINC(c, eptr);
3325              switch(c)
3326                {
3327                default: break;
3328                case 0x09:      /* HT */
3329                case 0x20:      /* SPACE */
3330                case 0xa0:      /* NBSP */
3331                case 0x1680:    /* OGHAM SPACE MARK */
3332                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3333                case 0x2000:    /* EN QUAD */
3334                case 0x2001:    /* EM QUAD */
3335                case 0x2002:    /* EN SPACE */
3336                case 0x2003:    /* EM SPACE */
3337                case 0x2004:    /* THREE-PER-EM SPACE */
3338                case 0x2005:    /* FOUR-PER-EM SPACE */
3339                case 0x2006:    /* SIX-PER-EM SPACE */
3340                case 0x2007:    /* FIGURE SPACE */
3341                case 0x2008:    /* PUNCTUATION SPACE */
3342                case 0x2009:    /* THIN SPACE */
3343                case 0x200A:    /* HAIR SPACE */
3344                case 0x202f:    /* NARROW NO-BREAK SPACE */
3345                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3346                case 0x3000:    /* IDEOGRAPHIC SPACE */
3347                RRETURN(MATCH_NOMATCH);
3348                }
3349              }
3350            break;
3351    
3352            case OP_HSPACE:
3353            for (i = 1; i <= min; i++)
3354              {
3355              if (eptr >= md->end_subject)
3356                {
3357                SCHECK_PARTIAL();
3358                RRETURN(MATCH_NOMATCH);
3359                }
3360              GETCHARINC(c, eptr);
3361              switch(c)
3362                {
3363                default: RRETURN(MATCH_NOMATCH);
3364                case 0x09:      /* HT */
3365                case 0x20:      /* SPACE */
3366                case 0xa0:      /* NBSP */
3367                case 0x1680:    /* OGHAM SPACE MARK */
3368                case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3369                case 0x2000:    /* EN QUAD */
3370                case 0x2001:    /* EM QUAD */
3371                case 0x2002:    /* EN SPACE */
3372                case 0x2003:    /* EM SPACE */
3373                case 0x2004:    /* THREE-PER-EM SPACE */
3374                case 0x2005:    /* FOUR-PER-EM SPACE */
3375                case 0x2006:    /* SIX-PER-EM SPACE */
3376                case 0x2007:    /* FIGURE SPACE */
3377                case 0x2008:    /* PUNCTUATION SPACE */
3378                case 0x2009:    /* THIN SPACE */
3379                case 0x200A:    /* HAIR SPACE */
3380                case 0x202f:    /* NARROW NO-BREAK SPACE */
3381                case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3382                case 0x3000:    /* IDEOGRAPHIC SPACE */
3383                break;
3384                }
3385              }
3386            break;
3387    
3388            case OP_NOT_VSPACE:
3389            for (i = 1; i <= min; i++)
3390              {
3391              if (eptr >= md->end_subject)
3392                {
3393                SCHECK_PARTIAL();
3394                RRETURN(MATCH_NOMATCH);
3395                }
3396              GETCHARINC(c, eptr);
3397              switch(c)
3398                {
3399                default: break;
3400                case 0x0a:      /* LF */
3401                case 0x0b:      /* VT */
3402                case 0x0c:      /* FF */
3403                case 0x0d:      /* CR */
3404                case 0x85:      /* NEL */
3405                case 0x2028:    /* LINE SEPARATOR */
3406                case 0x2029:    /* PARAGRAPH SEPARATOR */
3407                RRETURN(MATCH_NOMATCH);
3408                }
3409              }
3410            break;
3411    
3412            case OP_VSPACE:
3413            for (i = 1; i <= min; i++)
3414              {
3415              if (eptr >= md->end_subject)
3416                {
3417                SCHECK_PARTIAL();
3418                RRETURN(MATCH_NOMATCH);
3419                }
3420              GETCHARINC(c, eptr);
3421              switch(c)
3422                {
3423                default: RRETURN(MATCH_NOMATCH);
3424                case 0x0a:      /* LF */
3425                case 0x0b:      /* VT */
3426                case 0x0c:      /* FF */
3427                case 0x0d:      /* CR */
3428                case 0x85:      /* NEL */
3429                case 0x2028:    /* LINE SEPARATOR */
3430                case 0x2029:    /* PARAGRAPH SEPARATOR */
3431              break;              break;
3432              }              }
3433            }            }
# Line 2782  for (;;) Line 3436  for (;;)
3436          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
3437          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3438            {            {
3439            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3440                {
3441                SCHECK_PARTIAL();
3442                RRETURN(MATCH_NOMATCH);
3443                }
3444            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3445            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)
3446              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
# Line 2792  for (;;) Line 3450  for (;;)
3450          case OP_DIGIT:          case OP_DIGIT:
3451          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3452            {            {
3453            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3454               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)              {
3455                SCHECK_PARTIAL();
3456                RRETURN(MATCH_NOMATCH);
3457                }
3458              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)
3459              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3460            /* 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 */
3461            }            }
# Line 2802  for (;;) Line 3464  for (;;)
3464          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
3465          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3466            {            {
3467            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3468               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_space) != 0))              {
3469                SCHECK_PARTIAL();
3470              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3471            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;              }
3472              if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)
3473                RRETURN(MATCH_NOMATCH);
3474              while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3475            }            }
3476          break;          break;
3477    
3478          case OP_WHITESPACE:          case OP_WHITESPACE:
3479          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3480            {            {
3481            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3482               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)              {
3483                SCHECK_PARTIAL();
3484                RRETURN(MATCH_NOMATCH);
3485                }
3486              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)
3487              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3488            /* 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 */
3489            }            }
# Line 2823  for (;;) Line 3493  for (;;)
3493          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3494            {            {
3495            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3496               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_word) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0))
3497              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3498            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3499            }            }
3500          break;          break;
3501    
3502          case OP_WORDCHAR:          case OP_WORDCHAR:
3503          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3504            {            {
3505            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject)
3506               *eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)              {
3507                SCHECK_PARTIAL();
3508                RRETURN(MATCH_NOMATCH);
3509                }
3510              if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)
3511              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3512            /* 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 */
3513            }            }
# Line 2847  for (;;) Line 3521  for (;;)
3521  #endif     /* SUPPORT_UTF8 */  #endif     /* SUPPORT_UTF8 */
3522    
3523        /* 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
3524        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. */  
3525    
3526        switch(ctype)        switch(ctype)
3527          {          {
3528          case OP_ANY:          case OP_ANY:
3529          if ((ims & PCRE_DOTALL) == 0)          for (i = 1; i <= min; i++)
3530            {            {
3531            for (i = 1; i <= min; i++)            if (eptr >= md->end_subject)
3532              {              {
3533              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);              SCHECK_PARTIAL();
3534              eptr++;              RRETURN(MATCH_NOMATCH);
3535              }              }
3536              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
3537              eptr++;
3538            }            }
         else eptr += min;  
3539          break;          break;
3540    
3541          case OP_ANYBYTE:          case OP_ALLANY:
3542            if (eptr > md->end_subject - min)
3543              {
3544              SCHECK_PARTIAL();
3545              RRETURN(MATCH_NOMATCH);
3546              }
3547          eptr += min;          eptr += min;
3548          break;          break;
3549    
3550          /* Because of the CRLF case, we can't assume the minimum number of          case OP_ANYBYTE:
3551          bytes are present in this case. */          if (eptr > md->end_subject - min)
3552              {
3553              SCHECK_PARTIAL();
3554              RRETURN(MATCH_NOMATCH);
3555              }
3556            eptr += min;
3557            break;
3558    
3559          case OP_ANYNL:          case OP_ANYNL:
3560          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3561            {            {
3562            if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (eptr >= md->end_subject)
3563                {
3564                SCHECK_PARTIAL();
3565                RRETURN(MATCH_NOMATCH);
3566                }
3567            switch(*eptr++)            switch(*eptr++)
3568              {              {
3569              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
# Line 2882  for (;;) Line 3571  for (;;)
3571              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3572              break;              break;
3573              case 0x000a:              case 0x000a:
3574                break;
3575    
3576              case 0x000b:              case 0x000b:
3577              case 0x000c:              case 0x000c:
3578              case 0x0085:              case 0x0085:
3579                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3580                break;
3581                }
3582              }
3583            break;
3584    
3585            case OP_NOT_HSPACE:
3586            for (i = 1; i <= min; i++)
3587              {
3588              if (eptr >= md->end_subject)
3589                {
3590                SCHECK_PARTIAL();
3591                RRETURN(MATCH_NOMATCH);
3592                }
3593              switch(*eptr++)
3594                {
3595                default: break;
3596                case 0x09:      /* HT */
3597                case 0x20:      /* SPACE */
3598                case 0xa0:      /* NBSP */
3599                RRETURN(MATCH_NOMATCH);
3600                }
3601              }
3602            break;
3603    
3604            case OP_HSPACE:
3605            for (i = 1; i <= min; i++)
3606              {
3607              if (eptr >= md->end_subject)
3608                {
3609                SCHECK_PARTIAL();
3610                RRETURN(MATCH_NOMATCH);
3611                }
3612              switch(*eptr++)
3613                {
3614                default: RRETURN(MATCH_NOMATCH);
3615                case 0x09:      /* HT */
3616                case 0x20:      /* SPACE */
3617                case 0xa0:      /* NBSP */
3618                break;
3619                }
3620              }
3621            break;
3622    
3623            case OP_NOT_VSPACE:
3624            for (i = 1; i <= min; i++)
3625              {
3626              if (eptr >= md->end_subject)
3627                {
3628                SCHECK_PARTIAL();
3629                RRETURN(MATCH_NOMATCH);
3630                }
3631              switch(*eptr++)
3632                {
3633                default: break;
3634                case 0x0a:      /* LF */
3635                case 0x0b:      /* VT */
3636                case 0x0c:      /* FF */
3637                case 0x0d:      /* CR */
3638                case 0x85:      /* NEL */
3639                RRETURN(MATCH_NOMATCH);
3640                }
3641              }
3642            break;
3643    
3644            case OP_VSPACE:
3645            for (i = 1; i <= min; i++)
3646              {
3647              if (eptr >= md->end_subject)
3648                {
3649                SCHECK_PARTIAL();
3650                RRETURN(MATCH_NOMATCH);
3651                }
3652              switch(*eptr++)
3653                {
3654                default: RRETURN(MATCH_NOMATCH);
3655                case 0x0a:      /* LF */
3656                case 0x0b:      /* VT */
3657                case 0x0c:      /* FF */
3658                case 0x0d:      /* CR */
3659                case 0x85:      /* NEL */
3660              break;              break;
3661              }              }
3662            }            }
# Line 2892  for (;;) Line 3664  for (;;)
3664    
3665          case OP_NOT_DIGIT:          case OP_NOT_DIGIT:
3666          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3667              {
3668              if (eptr >= md->end_subject)
3669                {
3670                SCHECK_PARTIAL();
3671                RRETURN(MATCH_NOMATCH);
3672                }
3673            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);
3674              }
3675          break;          break;
3676    
3677          case OP_DIGIT:          case OP_DIGIT:
3678          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3679              {
3680              if (eptr >= md->end_subject)
3681                {
3682                SCHECK_PARTIAL();
3683                RRETURN(MATCH_NOMATCH);
3684                }
3685            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);
3686              }
3687          break;          break;
3688    
3689          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
3690          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3691              {
3692              if (eptr >= md->end_subject)
3693                {
3694                SCHECK_PARTIAL();
3695                RRETURN(MATCH_NOMATCH);
3696                }
3697            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);
3698              }
3699          break;          break;
3700    
3701          case OP_WHITESPACE:          case OP_WHITESPACE:
3702          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3703              {
3704              if (eptr >= md->end_subject)
3705                {
3706                SCHECK_PARTIAL();
3707                RRETURN(MATCH_NOMATCH);
3708                }
3709            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);            if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);
3710              }
3711          break;          break;
3712    
3713          case OP_NOT_WORDCHAR:          case OP_NOT_WORDCHAR:
3714          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3715              {
3716              if (eptr >= md->end_subject)
3717                {
3718                SCHECK_PARTIAL();
3719                RRETURN(MATCH_NOMATCH);
3720                }
3721            if ((md->ctypes[*eptr++] & ctype_word) != 0)            if ((md->ctypes[*eptr++] & ctype_word) != 0)
3722              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3723              }
3724          break;          break;
3725    
3726          case OP_WORDCHAR:          case OP_WORDCHAR:
3727          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3728              {
3729              if (eptr >= md->end_subject)
3730                {
3731                SCHECK_PARTIAL();
3732                RRETURN(MATCH_NOMATCH);
3733                }
3734            if ((md->ctypes[*eptr++] & ctype_word) == 0)            if ((md->ctypes[*eptr++] & ctype_word) == 0)
3735              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3736              }
3737          break;          break;
3738    
3739          default:          default:
# Line 2945  for (;;) Line 3759  for (;;)
3759            case PT_ANY:            case PT_ANY:
3760            for (fi = min;; fi++)            for (fi = min;; fi++)
3761              {              {
3762              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);
3763              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3764              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3765                if (eptr >= md->end_subject)
3766                  {
3767                  SCHECK_PARTIAL();
3768                  RRETURN(MATCH_NOMATCH);
3769                  }
3770              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3771              if (prop_fail_result) RRETURN(MATCH_NOMATCH);              if (prop_fail_result) RRETURN(MATCH_NOMATCH);
3772              }              }
# Line 2956  for (;;) Line 3775  for (;;)
3775            case PT_LAMP:            case PT_LAMP:
3776            for (fi = min;; fi++)            for (fi = min;; fi++)
3777              {              {
3778              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);
3779              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3780              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3781                if (eptr >= md->end_subject)
3782                  {
3783                  SCHECK_PARTIAL();
3784                  RRETURN(MATCH_NOMATCH);
3785                  }
3786              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3787              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
3788              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
3789                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
3790                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 2971  for (;;) Line 3795  for (;;)
3795            case PT_GC:            case PT_GC:
3796            for (fi = min;; fi++)            for (fi = min;; fi++)
3797              {              {
3798              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);
3799              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3800              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3801                if (eptr >= md->end_subject)
3802                  {
3803                  SCHECK_PARTIAL();
3804                  RRETURN(MATCH_NOMATCH);
3805                  }
3806              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3807              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
3808              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
3809                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3810              }              }
# Line 2984  for (;;) Line 3813  for (;;)
3813            case PT_PC:            case PT_PC:
3814            for (fi = min;; fi++)            for (fi = min;; fi++)
3815              {              {
3816              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);
3817              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3818              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3819                if (eptr >= md->end_subject)
3820                  {
3821                  SCHECK_PARTIAL();
3822                  RRETURN(MATCH_NOMATCH);
3823                  }
3824              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3825              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
3826              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
3827                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3828              }              }
# Line 2997  for (;;) Line 3831  for (;;)
3831            case PT_SC:            case PT_SC:
3832            for (fi = min;; fi++)            for (fi = min;; fi++)
3833              {              {
3834              RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);              RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);
3835              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3836              if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3837                if (eptr >= md->end_subject)
3838                  {
3839                  SCHECK_PARTIAL();
3840                  RRETURN(MATCH_NOMATCH);
3841                  }
3842              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
3843              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_script = UCD_SCRIPT(c);
3844              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
3845                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3846              }              }
# Line 3019  for (;;) Line 3858  for (;;)
3858          {          {
3859          for (fi = min;; fi++)          for (fi = min;; fi++)
3860            {            {
3861            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);
3862            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3863            if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3864              if (eptr >= md->end_subject)
3865                {
3866                SCHECK_PARTIAL();
3867                RRETURN(MATCH_NOMATCH);
3868                }
3869            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3870            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
3871            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);            if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH);
3872            while (eptr < md->end_subject)            while (eptr < md->end_subject)
3873              {              {
3874              int len = 1;              int len = 1;
3875              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr;
3876                {                else { GETCHARLEN(c, eptr, len); }
3877                GETCHARLEN(c, eptr, len);              prop_category = UCD_CATEGORY(c);
               }  
             prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);  
3878              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
3879              eptr += len;              eptr += len;
3880              }              }
# Line 3048  for (;;) Line 3890  for (;;)
3890          {          {
3891          for (fi = min;; fi++)          for (fi = min;; fi++)
3892            {            {
3893            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);
3894            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3895            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) RRETURN(MATCH_NOMATCH);
3896                 (ctype == OP_ANY && (ims & PCRE_DOTALL) == 0 &&            if (eptr >= md->end_subject)
3897                  IS_NEWLINE(eptr)))              {
3898                SCHECK_PARTIAL();
3899                RRETURN(MATCH_NOMATCH);
3900                }
3901              if (ctype == OP_ANY && IS_NEWLINE(eptr))
3902              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
   
3903            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
3904            switch(ctype)            switch(ctype)
3905              {              {
3906              case OP_ANY:        /* This is the DOTALL case */              case OP_ANY:        /* This is the non-NL case */
3907              break;              case OP_ALLANY:
   
3908              case OP_ANYBYTE:              case OP_ANYBYTE:
3909              break;              break;
3910    
# Line 3072  for (;;) Line 3916  for (;;)
3916                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3917                break;                break;
3918                case 0x000a:                case 0x000a:
3919                  break;
3920    
3921                case 0x000b:                case 0x000b:
3922                case 0x000c:                case 0x000c:
3923                case 0x0085:                case 0x0085:
3924                case 0x2028:                case 0x2028:
3925                case 0x2029:                case 0x2029:
3926                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3927                  break;
3928                  }
3929                break;
3930    
3931                case OP_NOT_HSPACE:
3932                switch(c)
3933                  {
3934                  default: break;
3935                  case 0x09:      /* HT */
3936                  case 0x20:      /* SPACE */
3937                  case 0xa0:      /* NBSP */
3938                  case 0x1680:    /* OGHAM SPACE MARK */
3939                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3940                  case 0x2000:    /* EN QUAD */
3941                  case 0x2001:    /* EM QUAD */
3942                  case 0x2002:    /* EN SPACE */
3943                  case 0x2003:    /* EM SPACE */
3944                  case 0x2004:    /* THREE-PER-EM SPACE */
3945                  case 0x2005:    /* FOUR-PER-EM SPACE */
3946                  case 0x2006:    /* SIX-PER-EM SPACE */
3947                  case 0x2007:    /* FIGURE SPACE */
3948                  case 0x2008:    /* PUNCTUATION SPACE */
3949                  case 0x2009:    /* THIN SPACE */
3950                  case 0x200A:    /* HAIR SPACE */
3951                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3952                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3953                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3954                  RRETURN(MATCH_NOMATCH);
3955                  }
3956                break;
3957    
3958                case OP_HSPACE:
3959                switch(c)
3960                  {
3961                  default: RRETURN(MATCH_NOMATCH);
3962                  case 0x09:      /* HT */
3963                  case 0x20:      /* SPACE */
3964                  case 0xa0:      /* NBSP */
3965                  case 0x1680:    /* OGHAM SPACE MARK */
3966                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
3967                  case 0x2000:    /* EN QUAD */
3968                  case 0x2001:    /* EM QUAD */
3969                  case 0x2002:    /* EN SPACE */
3970                  case 0x2003:    /* EM SPACE */
3971                  case 0x2004:    /* THREE-PER-EM SPACE */
3972                  case 0x2005:    /* FOUR-PER-EM SPACE */
3973                  case 0x2006:    /* SIX-PER-EM SPACE */
3974                  case 0x2007:    /* FIGURE SPACE */
3975                  case 0x2008:    /* PUNCTUATION SPACE */
3976                  case 0x2009:    /* THIN SPACE */
3977                  case 0x200A:    /* HAIR SPACE */
3978                  case 0x202f:    /* NARROW NO-BREAK SPACE */
3979                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
3980                  case 0x3000:    /* IDEOGRAPHIC SPACE */
3981                  break;
3982                  }
3983                break;
3984    
3985                case OP_NOT_VSPACE:
3986                switch(c)
3987                  {
3988                  default: break;
3989                  case 0x0a:      /* LF */
3990                  case 0x0b:      /* VT */
3991                  case 0x0c:      /* FF */
3992                  case 0x0d:      /* CR */
3993                  case 0x85:      /* NEL */
3994                  case 0x2028:    /* LINE SEPARATOR */
3995                  case 0x2029:    /* PARAGRAPH SEPARATOR */
3996                  RRETURN(MATCH_NOMATCH);
3997                  }
3998                break;
3999    
4000                case OP_VSPACE:
4001                switch(c)
4002                  {
4003                  default: RRETURN(MATCH_NOMATCH);
4004                  case 0x0a:      /* LF */
4005                  case 0x0b:      /* VT */
4006                  case 0x0c:      /* FF */
4007                  case 0x0d:      /* CR */
4008                  case 0x85:      /* NEL */
4009                  case 0x2028:    /* LINE SEPARATOR */
4010                  case 0x2029:    /* PARAGRAPH SEPARATOR */
4011                break;                break;
4012                }                }
4013              break;              break;
# Line 3122  for (;;) Line 4053  for (;;)
4053          {          {
4054          for (fi = min;; fi++)          for (fi = min;; fi++)
4055            {            {
4056            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);
4057            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4058            if (fi >= max || eptr >= md->end_subject ||            if (fi >= max) RRETURN(MATCH_NOMATCH);
4059                 ((ims & PCRE_DOTALL) == 0 && IS_NEWLINE(eptr)))            if (eptr >= md->end_subject)
4060                {
4061                SCHECK_PARTIAL();
4062                RRETURN(MATCH_NOMATCH);
4063                }
4064              if (ctype == OP_ANY && IS_NEWLINE(eptr))
4065              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
   
4066            c = *eptr++;            c = *eptr++;
4067            switch(ctype)            switch(ctype)
4068              {              {
4069              case OP_ANY:   /* This is the DOTALL case */              case OP_ANY:     /* This is the non-NL case */
4070              break;              case OP_ALLANY:
   
4071              case OP_ANYBYTE:              case OP_ANYBYTE:
4072              break;              break;
4073    
# Line 3144  for (;;) Line 4078  for (;;)
4078                case 0x000d:                case 0x000d:
4079                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
4080                break;                break;
4081    
4082                case 0x000a:                case 0x000a:
4083                  break;
4084    
4085                case 0x000b:                case 0x000b:
4086                case 0x000c:                case 0x000c:
4087                case 0x0085:                case 0x0085:
4088                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4089                  break;
4090                  }
4091                break;
4092    
4093                case OP_NOT_HSPACE:
4094                switch(c)
4095                  {
4096                  default: break;
4097                  case 0x09:      /* HT */
4098                  case 0x20:      /* SPACE */
4099                  case 0xa0:      /* NBSP */
4100                  RRETURN(MATCH_NOMATCH);
4101                  }
4102                break;
4103    
4104                case OP_HSPACE:
4105                switch(c)
4106                  {
4107                  default: RRETURN(MATCH_NOMATCH);
4108                  case 0x09:      /* HT */
4109                  case 0x20:      /* SPACE */
4110                  case 0xa0:      /* NBSP */
4111                  break;
4112                  }
4113                break;
4114    
4115                case OP_NOT_VSPACE:
4116                switch(c)
4117                  {
4118                  default: break;
4119                  case 0x0a:      /* LF */
4120                  case 0x0b:      /* VT */
4121                  case 0x0c:      /* FF */
4122                  case 0x0d:      /* CR */
4123                  case 0x85:      /* NEL */
4124                  RRETURN(MATCH_NOMATCH);
4125                  }
4126                break;
4127    
4128                case OP_VSPACE:
4129                switch(c)
4130                  {
4131                  default: RRETURN(MATCH_NOMATCH);
4132                  case 0x0a:      /* LF */
4133                  case 0x0b:      /* VT */
4134                  case 0x0c:      /* FF */
4135                  case 0x0d:      /* CR */
4136                  case 0x85:      /* NEL */
4137                break;                break;
4138                }                }
4139              break;              break;
# Line 3214  for (;;) Line 4200  for (;;)
4200              int len = 1;              int len = 1;
4201              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4202              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4203              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
4204              if ((prop_chartype == ucp_Lu ||              if ((prop_chartype == ucp_Lu ||
4205                   prop_chartype == ucp_Ll ||                   prop_chartype == ucp_Ll ||
4206                   prop_chartype == ucp_Lt) == prop_fail_result)                   prop_chartype == ucp_Lt) == prop_fail_result)
# Line 3229  for (;;) Line 4215  for (;;)
4215              int len = 1;              int len = 1;
4216              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4217              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4218              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4219              if ((prop_category == prop_value) == prop_fail_result)              if ((prop_category == prop_value) == prop_fail_result)
4220                break;                break;
4221              eptr+= len;              eptr+= len;
# Line 3242  for (;;) Line 4228  for (;;)
4228              int len = 1;              int len = 1;
4229              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4230              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4231              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_chartype = UCD_CHARTYPE(c);
4232              if ((prop_chartype == prop_value) == prop_fail_result)              if ((prop_chartype == prop_value) == prop_fail_result)
4233                break;                break;
4234              eptr+= len;              eptr+= len;
# Line 3255  for (;;) Line 4241  for (;;)
4241              int len = 1;              int len = 1;
4242              if (eptr >= md->end_subject) break;              if (eptr >= md->end_subject) break;
4243              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
4244              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_script = UCD_SCRIPT(c);
4245              if ((prop_script == prop_value) == prop_fail_result)              if ((prop_script == prop_value) == prop_fail_result)
4246                break;                break;
4247              eptr+= len;              eptr+= len;
# Line 3268  for (;;) Line 4254  for (;;)
4254          if (possessive) continue;          if (possessive) continue;
4255          for(;;)          for(;;)
4256            {            {
4257            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);
4258            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4259            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
4260            BACKCHAR(eptr);            if (utf8) BACKCHAR(eptr);
4261            }            }
4262          }          }
4263    
# Line 3284  for (;;) Line 4270  for (;;)
4270            {            {
4271            if (eptr >= md->end_subject) break;            if (eptr >= md->end_subject) break;
4272            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
4273            prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);            prop_category = UCD_CATEGORY(c);
4274            if (prop_category == ucp_M) break;            if (prop_category == ucp_M) break;
4275            while (eptr < md->end_subject)            while (eptr < md->end_subject)
4276              {              {
# Line 3293  for (;;) Line 4279  for (;;)
4279                {                {
4280                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
4281                }                }
4282              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4283              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
4284              eptr += len;              eptr += len;
4285              }              }
# Line 3304  for (;;) Line 4290  for (;;)
4290          if (possessive) continue;          if (possessive) continue;
4291          for(;;)          for(;;)
4292            {            {
4293            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM45);
4294            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4295            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
4296            for (;;)                        /* Move back over one extended */            for (;;)                        /* Move back over one extended */
4297              {              {
4298              int len = 1;              int len = 1;
             BACKCHAR(eptr);  
4299              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr; else
4300                {                {
4301                  BACKCHAR(eptr);
4302                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
4303                }                }
4304              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = UCD_CATEGORY(c);
4305              if (prop_category != ucp_M) break;              if (prop_category != ucp_M) break;
4306              eptr--;              eptr--;
4307              }              }
# Line 3333  for (;;) Line 4319  for (;;)
4319          switch(ctype)          switch(ctype)
4320            {            {
4321            case OP_ANY:            case OP_ANY:
   
           /* Special code is required for UTF8, but when the maximum is  
           unlimited we don't need it, so we repeat the non-UTF8 code. This is  
           probably worth it, because .* is quite a common idiom. */  
   
4322            if (max < INT_MAX)            if (max < INT_MAX)
4323              {              {
4324              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  
4325                {                {
4326                for (i = min; i < max; i++)                if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
4327                  {                eptr++;
4328                  if (eptr >= md->end_subject) break;                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
                 eptr++;  
                 while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  
                 }  
4329                }                }
4330              }              }
4331    
# Line 3364  for (;;) Line 4333  for (;;)
4333    
4334            else            else
4335              {              {
4336              if ((ims & PCRE_DOTALL) == 0)              for (i = min; i < max; i++)
4337                {                {
4338                for (i = min; i < max; i++)                if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
4339                  {                eptr++;
4340                  if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
                 eptr++;  
                 }  
               break;  
4341                }                }
4342              else              }
4343              break;
4344    
4345              case OP_ALLANY:
4346              if (max < INT_MAX)
4347                {
4348                for (i = min; i < max; i++)
4349                {                {
4350                c = max - min;                if (eptr >= md->end_subject) break;
4351                if (c > (unsigned int)(md->end_subject - eptr))                eptr++;
4352                  c = md->end_subject - eptr;                while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
               eptr += c;  
4353                }                }
4354              }              }
4355              else eptr = md->end_subject;   /* Unlimited UTF-8 repeat */
4356            break;            break;
4357    
4358            /* The byte case is the same as non-UTF8 */            /* The byte case is the same as non-UTF8 */
# Line 3405  for (;;) Line 4377  for (;;)
4377                }                }
4378              else              else
4379                {                {
4380                if (c != 0x000a && c != 0x000b && c != 0x000c &&                if (c != 0x000a &&
4381                    c != 0x0085 && c != 0x2028 && c != 0x2029)                    (md->bsr_anycrlf ||
4382                       (c != 0x000b && c != 0x000c &&
4383                        c != 0x0085 && c != 0x2028 && c != 0x2029)))
4384                  break;                  break;
4385                eptr += len;                eptr += len;
4386                }                }
4387              }              }
4388            break;            break;
4389    
4390              case OP_NOT_HSPACE:
4391              case OP_HSPACE:
4392              for (i = min; i < max; i++)
4393                {
4394                BOOL gotspace;
4395                int len = 1;
4396                if (eptr >= md->end_subject) break;
4397                GETCHARLEN(c, eptr, len);
4398                switch(c)
4399                  {
4400                  default: gotspace = FALSE; break;
4401                  case 0x09:      /* HT */
4402                  case 0x20:      /* SPACE */
4403                  case 0xa0:      /* NBSP */
4404                  case 0x1680:    /* OGHAM SPACE MARK */
4405                  case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */
4406                  case 0x2000:    /* EN QUAD */
4407                  case 0x2001:    /* EM QUAD */
4408                  case 0x2002:    /* EN SPACE */
4409                  case 0x2003:    /* EM SPACE */
4410                  case 0x2004:    /* THREE-PER-EM SPACE */
4411                  case 0x2005:    /* FOUR-PER-EM SPACE */
4412                  case 0x2006:    /* SIX-PER-EM SPACE */
4413                  case 0x2007:    /* FIGURE SPACE */
4414                  case 0x2008:    /* PUNCTUATION SPACE */
4415                  case 0x2009:    /* THIN SPACE */
4416                  case 0x200A:    /* HAIR SPACE */
4417                  case 0x202f:    /* NARROW NO-BREAK SPACE */
4418                  case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */
4419                  case 0x3000:    /* IDEOGRAPHIC SPACE */
4420                  gotspace = TRUE;
4421                  break;
4422                  }
4423                if (gotspace == (ctype == OP_NOT_HSPACE)) break;
4424                eptr += len;
4425                }
4426              break;
4427    
4428              case OP_NOT_VSPACE:
4429              case OP_VSPACE:
4430              for (i = min; i < max; i++)
4431                {
4432                BOOL gotspace;
4433                int len = 1;
4434                if (eptr >= md->end_subject) break;
4435                GETCHARLEN(c, eptr, len);
4436                switch(c)
4437                  {
4438                  default: gotspace = FALSE; break;
4439                  case 0x0a:      /* LF */
4440                  case 0x0b:      /* VT */
4441                  case 0x0c:      /* FF */
4442                  case 0x0d:      /* CR */
4443                  case 0x85:      /* NEL */
4444                  case 0x2028:    /* LINE SEPARATOR */
4445                  case 0x2029:    /* PARAGRAPH SEPARATOR */
4446                  gotspace = TRUE;
4447                  break;
4448                  }
4449                if (gotspace == (ctype == OP_NOT_VSPACE)) break;
4450                eptr += len;
4451                }
4452              break;
4453    
4454            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
4455            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4456              {              {
# Line 3488  for (;;) Line 4526  for (;;)
4526          if (possessive) continue;          if (possessive) continue;
4527          for(;;)          for(;;)
4528            {            {
4529            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM46);
4530            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4531            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
4532            BACKCHAR(eptr);            BACKCHAR(eptr);
4533            }            }
4534          }          }
4535        else        else
4536  #endif  #endif  /* SUPPORT_UTF8 */
4537    
4538        /* Not UTF-8 mode */        /* Not UTF-8 mode */
4539          {          {
4540          switch(ctype)          switch(ctype)
4541            {            {
4542            case OP_ANY:            case OP_ANY:
4543            if ((ims & PCRE_DOTALL) == 0)            for (i = min; i < max; i++)
4544              {              {
4545              for (i = min; i < max; i++)              if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;
4546                {              eptr++;
               if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break;  
               eptr++;  
               }  
             break;  
4547              }              }
4548            /* For DOTALL case, fall through and treat as \C */            break;
4549    
4550              case OP_ALLANY:
4551            case OP_ANYBYTE:            case OP_ANYBYTE:
4552            c = max - min;            c = max - min;
4553            if (c > (unsigned int)(md->end_subject - eptr))            if (c > (unsigned int)(md->end_subject - eptr))
# Line 3532  for (;;) Line 4567  for (;;)
4567                }                }
4568              else              else
4569                {                {
4570                if (c != 0x000a && c != 0x000b && c != 0x000c && c != 0x0085)                if (c != 0x000a &&
4571                      (md->bsr_anycrlf ||
4572                        (c != 0x000b && c != 0x000c && c != 0x0085)))
4573                  break;                  break;
4574                eptr++;                eptr++;
4575                }                }
4576              }              }
4577            break;            break;
4578    
4579              case OP_NOT_HSPACE:
4580              for (i = min; i < max; i++)
4581                {
4582                if (eptr >= md->end_subject) break;
4583                c = *eptr;
4584                if (c == 0x09 || c == 0x20 || c == 0xa0) break;
4585                eptr++;
4586                }
4587              break;
4588    
4589              case OP_HSPACE:
4590              for (i = min; i < max; i++)
4591                {
4592                if (eptr >= md->end_subject) break;
4593                c = *eptr;
4594                if (c != 0x09 && c != 0x20 && c != 0xa0) break;
4595                eptr++;
4596                }
4597              break;
4598    
4599              case OP_NOT_VSPACE:
4600              for (i = min; i < max; i++)
4601                {
4602                if (eptr >= md->end_subject) break;
4603                c = *eptr;
4604                if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)
4605                  break;
4606                eptr++;
4607                }
4608              break;
4609    
4610              case OP_VSPACE:
4611              for (i = min; i < max; i++)
4612                {
4613                if (eptr >= md->end_subject) break;
4614                c = *eptr;
4615                if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85)
4616                  break;
4617                eptr++;
4618                }
4619              break;
4620    
4621            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
4622            for (i = min; i < max; i++)            for (i = min; i < max; i++)
4623              {              {
# Line 3602  for (;;) Line 4681  for (;;)
4681          if (possessive) continue;          if (possessive) continue;
4682          while (eptr >= pp)          while (eptr >= pp)
4683            {            {
4684            RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM47);
4685            eptr--;            eptr--;
4686            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
4687            }            }
# Line 3628  for (;;) Line 4707  for (;;)
4707    
4708    }             /* End of main loop */    }             /* End of main loop */
4709  /* Control never reaches here */  /* Control never reaches here */
4710    
4711    
4712    /* When compiling to use the heap rather than the stack for recursive calls to
4713    match(), the RRETURN() macro jumps here. The number that is saved in
4714    frame->Xwhere indicates which label we actually want to return to. */
4715    
4716    #ifdef NO_RECURSE
4717    #define LBL(val) case val: goto L_RM##val;
4718    HEAP_RETURN:
4719    switch (frame->Xwhere)
4720      {
4721      LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8)
4722      LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(17)
4723      LBL(19) LBL(24) LBL(25) LBL(26) LBL(27) LBL(29) LBL(31) LBL(33)
4724      LBL(35) LBL(43) LBL(47) LBL(48) LBL(49) LBL(50) LBL(51) LBL(52)
4725      LBL(53) LBL(54)
4726    #ifdef SUPPORT_UTF8
4727      LBL(16) LBL(18) LBL(20) LBL(21) LBL(22) LBL(23) LBL(28) LBL(30)
4728      LBL(32) LBL(34) LBL(42) LBL(46)
4729    #ifdef SUPPORT_UCP
4730      LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
4731    #endif  /* SUPPORT_UCP */
4732    #endif  /* SUPPORT_UTF8 */
4733      default:
4734      DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));
4735      return PCRE_ERROR_INTERNAL;
4736      }
4737    #undef LBL
4738    #endif  /* NO_RECURSE */
4739  }  }
4740    
4741    
# Line 3640  Undefine all the macros that were define Line 4748  Undefine all the macros that were define
4748  #ifdef NO_RECURSE  #ifdef NO_RECURSE
4749  #undef eptr  #undef eptr
4750  #undef ecode  #undef ecode
4751    #undef mstart
4752  #undef offset_top  #undef offset_top
4753  #undef ims  #undef ims
4754  #undef eptrb  #undef eptrb
# Line 3712  Returns:          > 0 => success; value Line 4821  Returns:          > 0 => success; value
4821                   < -1 => some kind of unexpected problem                   < -1 => some kind of unexpected problem
4822  */  */
4823    
4824  PCRE_DATA_SCOPE int  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
4825  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,
4826    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,
4827    int offsetcount)    int offsetcount)
# Line 3736  const uschar *tables; Line 4845  const uschar *tables;
4845  const uschar *start_bits = NULL;  const uschar *start_bits = NULL;
4846  USPTR start_match = (USPTR)subject + start_offset;  USPTR start_match = (USPTR)subject + start_offset;
4847  USPTR end_subject;  USPTR end_subject;
4848    USPTR start_partial = NULL;
4849  USPTR req_byte_ptr = start_match - 1;  USPTR req_byte_ptr = start_match - 1;
 eptrblock eptrchain[EPTR_WORK_SIZE];  
4850    
4851  pcre_study_data internal_study;  pcre_study_data internal_study;
4852  const pcre_study_data *study;  const pcre_study_data *study;
# Line 3800  if (re->magic_number != MAGIC_NUMBER) Line 4909  if (re->magic_number != MAGIC_NUMBER)
4909  /* Set up other data */  /* Set up other data */
4910    
4911  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
4912  startline = (re->options & PCRE_STARTLINE) != 0;  startline = (re->flags & PCRE_STARTLINE) != 0;
4913  firstline = (re->options & PCRE_FIRSTLINE) != 0;  firstline = (re->options & PCRE_FIRSTLINE) != 0;
4914    
4915  /* The code starts after the real_pcre block and the capture name table. */  /* The code starts after the real_pcre block and the capture name table. */
# Line 3815  end_subject = md->end_subject; Line 4924  end_subject = md->end_subject;
4924    
4925  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;
4926  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;
4927    md->jscript_compat = (re->options & PCRE_JAVASCRIPT_COMPAT) != 0;
4928    
4929  md->notbol = (options & PCRE_NOTBOL) != 0;  md->notbol = (options & PCRE_NOTBOL) != 0;
4930  md->noteol = (options & PCRE_NOTEOL) != 0;  md->noteol = (options & PCRE_NOTEOL) != 0;
4931  md->notempty = (options & PCRE_NOTEMPTY) != 0;  md->notempty = (options & PCRE_NOTEMPTY) != 0;
4932  md->partial = (options & PCRE_PARTIAL) != 0;  md->notempty_atstart = (options & PCRE_NOTEMPTY_ATSTART) != 0;
4933    md->partial = ((options & PCRE_PARTIAL_HARD) != 0)? 2 :
4934                  ((options & PCRE_PARTIAL_SOFT) != 0)? 1 : 0;
4935  md->hitend = FALSE;  md->hitend = FALSE;
4936    
4937  md->recursive = NULL;                   /* No recursion at top level */  md->recursive = NULL;                   /* No recursion at top level */
 md->eptrchain = eptrchain;              /* Make workspace generally available */  
4938    
4939  md->lcc = tables + lcc_offset;  md->lcc = tables + lcc_offset;
4940  md->ctypes = tables + ctypes_offset;  md->ctypes = tables + ctypes_offset;
4941    
4942    /* Handle different \R options. */
4943    
4944    switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
4945      {
4946      case 0:
4947      if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
4948        md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0;
4949      else
4950    #ifdef BSR_ANYCRLF
4951      md->bsr_anycrlf = TRUE;
4952    #else
4953      md->bsr_anycrlf = FALSE;
4954    #endif
4955      break;
4956    
4957      case PCRE_BSR_ANYCRLF:
4958      md->bsr_anycrlf = TRUE;
4959      break;
4960    
4961      case PCRE_BSR_UNICODE:
4962      md->bsr_anycrlf = FALSE;
4963      break;
4964    
4965      default: return PCRE_ERROR_BADNEWLINE;
4966      }
4967    
4968  /* Handle different types of newline. The three bits give eight cases. If  /* Handle different types of newline. The three bits give eight cases. If
4969  nothing is set at run time, whatever was used at compile time applies. */  nothing is set at run time, whatever was used at compile time applies. */
4970    
4971  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
4972         PCRE_NEWLINE_BITS)          (pcre_uint32)options) & PCRE_NEWLINE_BITS)
4973    {    {
4974    case 0: newline = NEWLINE; break;   /* Compile-time default */    case 0: newline = NEWLINE; break;   /* Compile-time default */
4975    case PCRE_NEWLINE_CR: newline = '\r'; break;    case PCRE_NEWLINE_CR: newline = CHAR_CR; break;
4976    case PCRE_NEWLINE_LF: newline = '\n'; break;    case PCRE_NEWLINE_LF: newline = CHAR_NL; break;
4977    case PCRE_NEWLINE_CR+    case PCRE_NEWLINE_CR+
4978         PCRE_NEWLINE_LF: newline = ('\r' << 8) | '\n'; break;         PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break;
4979    case PCRE_NEWLINE_ANY: newline = -1; break;    case PCRE_NEWLINE_ANY: newline = -1; break;
4980      case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
4981    default: return PCRE_ERROR_BADNEWLINE;    default: return PCRE_ERROR_BADNEWLINE;
4982    }    }
4983    
4984  if (newline < 0)  if (newline == -2)
4985      {
4986      md->nltype = NLTYPE_ANYCRLF;
4987      }
4988    else if (newline < 0)
4989    {    {
4990    md->nltype = NLTYPE_ANY;    md->nltype = NLTYPE_ANY;
4991    }    }
# Line 3863  else Line 5005  else
5005      }      }
5006    }    }
5007    
5008  /* Partial matching is supported only for a restricted set of regexes at the  /* Partial matching was originally supported only for a restricted set of
5009  moment. */  regexes; from release 8.00 there are no restrictions, but the bits are still
5010    defined (though never set). So there's no harm in leaving this code. */
5011    
5012  if (md->partial && (re->options & PCRE_NOPARTIAL) != 0)  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
5013    return PCRE_ERROR_BADPARTIAL;    return PCRE_ERROR_BADPARTIAL;
5014    
5015  /* Check a UTF-8 string if required. Unfortunately there's no way of passing  /* Check a UTF-8 string if required. Unfortunately there's no way of passing
# Line 3875  back the character offset. */ Line 5018  back the character offset. */
5018  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
5019  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)
5020    {    {
5021