/[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 723 by ph10, Sat Oct 8 15:55:23 2011 UTC revision 1284 by ph10, Fri Mar 15 11:54:58 2013 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) 1997-2011 University of Cambridge             Copyright (c) 1997-2013 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 37  POSSIBILITY OF SUCH DAMAGE. Line 37  POSSIBILITY OF SUCH DAMAGE.
37  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
38  */  */
39    
   
40  /* This module contains pcre_exec(), the externally visible function that does  /* This module contains pcre_exec(), the externally visible function that does
41  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
42  possible. There are also some static supporting functions. */  possible. There are also some static supporting functions. */
# Line 57  possible. There are also some static sup Line 56  possible. There are also some static sup
56  #undef min  #undef min
57  #undef max  #undef max
58    
59    /* The md->capture_last field uses the lower 16 bits for the last captured
60    substring (which can never be greater than 65535) and a bit in the top half
61    to mean "capture vector overflowed". This odd way of doing things was
62    implemented when it was realized that preserving and restoring the overflow bit
63    whenever the last capture number was saved/restored made for a neater
64    interface, and doing it this way saved on (a) another variable, which would
65    have increased the stack frame size (a big NO-NO in PCRE) and (b) another
66    separate set of save/restore instructions. The following defines are used in
67    implementing this. */
68    
69    #define CAPLMASK    0x0000ffff    /* The bits used for last_capture */
70    #define OVFLMASK    0xffff0000    /* The bits used for the overflow flag */
71    #define OVFLBIT     0x00010000    /* The bit that is set for overflow */
72    
73  /* Values for setting in md->match_function_type to indicate two special types  /* Values for setting in md->match_function_type to indicate two special types
74  of call to match(). We do it this way to save on using another stack variable,  of call to match(). We do it this way to save on using another stack variable,
75  as stack usage is to be discouraged. */  as stack usage is to be discouraged. */
# Line 74  defined PCRE_ERROR_xxx codes, which are Line 87  defined PCRE_ERROR_xxx codes, which are
87  negative to avoid the external error codes. */  negative to avoid the external error codes. */
88    
89  #define MATCH_ACCEPT       (-999)  #define MATCH_ACCEPT       (-999)
90  #define MATCH_COMMIT       (-998)  #define MATCH_KETRPOS      (-998)
91  #define MATCH_KETRPOS      (-997)  #define MATCH_ONCE         (-997)
92  #define MATCH_ONCE         (-996)  /* The next 5 must be kept together and in sequence so that a test that checks
93    for any one of them can use a range. */
94    #define MATCH_COMMIT       (-996)
95  #define MATCH_PRUNE        (-995)  #define MATCH_PRUNE        (-995)
96  #define MATCH_SKIP         (-994)  #define MATCH_SKIP         (-994)
97  #define MATCH_SKIP_ARG     (-993)  #define MATCH_SKIP_ARG     (-993)
98  #define MATCH_THEN         (-992)  #define MATCH_THEN         (-992)
99    #define MATCH_BACKTRACK_MAX MATCH_THEN
100  /* This is a convenience macro for code that occurs many times. */  #define MATCH_BACKTRACK_MIN MATCH_COMMIT
   
 #define MRRETURN(ra) \  
   { \  
   md->mark = markptr; \  
   RRETURN(ra); \  
   }  
101    
102  /* 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.
103  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,
# Line 101  because the offset vector is always a mu Line 110  because the offset vector is always a mu
110  static const char rep_min[] = { 0, 0, 1, 1, 0, 0 };  static const char rep_min[] = { 0, 0, 1, 1, 0, 0 };
111  static const char rep_max[] = { 0, 0, 0, 0, 1, 1 };  static const char rep_max[] = { 0, 0, 0, 0, 1, 1 };
112    
   
   
113  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
114  /*************************************************  /*************************************************
115  *        Debugging function to print chars       *  *        Debugging function to print chars       *
# Line 121  Returns:     nothing Line 128  Returns:     nothing
128  */  */
129    
130  static void  static void
131  pchars(const uschar *p, int length, BOOL is_subject, match_data *md)  pchars(const pcre_uchar *p, int length, BOOL is_subject, match_data *md)
132  {  {
133  unsigned int c;  pcre_uint32 c;
134    BOOL utf = md->utf;
135  if (is_subject && length > md->end_subject - p) length = md->end_subject - p;  if (is_subject && length > md->end_subject - p) length = md->end_subject - p;
136  while (length-- > 0)  while (length-- > 0)
137    if (isprint(c = *(p++))) printf("%c", c); else printf("\\x%02x", c);    if (isprint(c = RAWUCHARINCTEST(p))) printf("%c", (char)c); else printf("\\x{%02x}", c);
138  }  }
139  #endif  #endif
140    
# Line 148  Arguments: Line 156  Arguments:
156    md          points to match data block    md          points to match data block
157    caseless    TRUE if caseless    caseless    TRUE if caseless
158    
159  Returns:      < 0 if not matched, otherwise the number of subject bytes matched  Returns:      >= 0 the number of subject bytes matched
160                  -1 no match
161                  -2 partial match; always given if at end subject
162  */  */
163    
164  static int  static int
165  match_ref(int offset, register USPTR eptr, int length, match_data *md,  match_ref(int offset, register PCRE_PUCHAR eptr, int length, match_data *md,
166    BOOL caseless)    BOOL caseless)
167  {  {
168  USPTR eptr_start = eptr;  PCRE_PUCHAR eptr_start = eptr;
169  register USPTR p = md->start_subject + md->offset_vector[offset];  register PCRE_PUCHAR p = md->start_subject + md->offset_vector[offset];
170    #ifdef SUPPORT_UTF
171    BOOL utf = md->utf;
172    #endif
173    
174  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
175  if (eptr >= md->end_subject)  if (eptr >= md->end_subject)
# Line 171  pchars(p, length, FALSE, md); Line 184  pchars(p, length, FALSE, md);
184  printf("\n");  printf("\n");
185  #endif  #endif
186    
187  /* Always fail if reference not set (and not JavaScript compatible). */  /* Always fail if reference not set (and not JavaScript compatible - in that
188    case the length is passed as zero). */
189    
190  if (length < 0) return -1;  if (length < 0) return -1;
191    
# Line 181  ASCII characters. */ Line 195  ASCII characters. */
195    
196  if (caseless)  if (caseless)
197    {    {
198  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
199  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
200    if (md->utf8)    if (utf)
201      {      {
202      /* Match characters up to the end of the reference. NOTE: the number of      /* Match characters up to the end of the reference. NOTE: the number of
203      bytes matched may differ, because there are some characters whose upper and      data units matched may differ, because in UTF-8 there are some characters
204      lower case versions code as different numbers of bytes. For example, U+023A      whose upper and lower case versions code have different numbers of bytes.
205      (2 bytes in UTF-8) is the upper case version of U+2C65 (3 bytes in UTF-8);      For example, U+023A (2 bytes in UTF-8) is the upper case version of U+2C65
206      a sequence of 3 of the former uses 6 bytes, as does a sequence of two of      (3 bytes in UTF-8); a sequence of 3 of the former uses 6 bytes, as does a
207      the latter. It is important, therefore, to check the length along the      sequence of two of the latter. It is important, therefore, to check the
208      reference, not along the subject (earlier code did this wrong). */      length along the reference, not along the subject (earlier code did this
209        wrong). */
210    
211      USPTR endptr = p + length;      PCRE_PUCHAR endptr = p + length;
212      while (p < endptr)      while (p < endptr)
213        {        {
214        int c, d;        pcre_uint32 c, d;
215        if (eptr >= md->end_subject) return -1;        const ucd_record *ur;
216          if (eptr >= md->end_subject) return -2;   /* Partial match */
217        GETCHARINC(c, eptr);        GETCHARINC(c, eptr);
218        GETCHARINC(d, p);        GETCHARINC(d, p);
219        if (c != d && c != UCD_OTHERCASE(d)) return -1;        ur = GET_UCD(d);
220          if (c != d && c != d + ur->other_case)
221            {
222            const pcre_uint32 *pp = PRIV(ucd_caseless_sets) + ur->caseset;
223            for (;;)
224              {
225              if (c < *pp) return -1;
226              if (c == *pp++) break;
227              }
228            }
229        }        }
230      }      }
231    else    else
# Line 210  if (caseless) Line 235  if (caseless)
235    /* The same code works when not in UTF-8 mode and in UTF-8 mode when there    /* The same code works when not in UTF-8 mode and in UTF-8 mode when there
236    is no UCP support. */    is no UCP support. */
237      {      {
     if (eptr + length > md->end_subject) return -1;  
238      while (length-- > 0)      while (length-- > 0)
239        { if (md->lcc[*p++] != md->lcc[*eptr++]) return -1; }        {
240          pcre_uint32 cc, cp;
241          if (eptr >= md->end_subject) return -2;   /* Partial match */
242          cc = RAWUCHARTEST(eptr);
243          cp = RAWUCHARTEST(p);
244          if (TABLE_GET(cp, md->lcc, cp) != TABLE_GET(cc, md->lcc, cc)) return -1;
245          p++;
246          eptr++;
247          }
248      }      }
249    }    }
250    
# Line 221  are in UTF-8 mode. */ Line 253  are in UTF-8 mode. */
253    
254  else  else
255    {    {
256    if (eptr + length > md->end_subject) return -1;    while (length-- > 0)
257    while (length-- > 0) if (*p++ != *eptr++) return -1;      {
258        if (eptr >= md->end_subject) return -2;   /* Partial match */
259        if (RAWUCHARINCTEST(p) != RAWUCHARINCTEST(eptr)) return -1;
260        }
261    }    }
262    
263  return eptr - eptr_start;  return (int)(eptr - eptr_start);
264  }  }
265    
266    
# Line 277  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM Line 312  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM
312         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
313         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
314         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,
315         RM61,  RM62, RM63, RM64, RM65, RM66 };         RM61,  RM62, RM63, RM64, RM65, RM66, RM67, RM68 };
316    
317  /* 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
318  versions and production versions. Note that the "rw" argument of RMATCH isn't  versions and production versions. Note that the "rw" argument of RMATCH isn't
# Line 290  actually used in this definition. */ Line 325  actually used in this definition. */
325  #define RMATCH(ra,rb,rc,rd,re,rw) \  #define RMATCH(ra,rb,rc,rd,re,rw) \
326    { \    { \
327    printf("match() called in line %d\n", __LINE__); \    printf("match() called in line %d\n", __LINE__); \
328    rrc = match(ra,rb,mstart,markptr,rc,rd,re,rdepth+1); \    rrc = match(ra,rb,mstart,rc,rd,re,rdepth+1); \
329    printf("to line %d\n", __LINE__); \    printf("to line %d\n", __LINE__); \
330    }    }
331  #define RRETURN(ra) \  #define RRETURN(ra) \
332    { \    { \
333    printf("match() returned %d from line %d ", ra, __LINE__); \    printf("match() returned %d from line %d\n", ra, __LINE__); \
334    return ra; \    return ra; \
335    }    }
336  #else  #else
337  #define RMATCH(ra,rb,rc,rd,re,rw) \  #define RMATCH(ra,rb,rc,rd,re,rw) \
338    rrc = match(ra,rb,mstart,markptr,rc,rd,re,rdepth+1)    rrc = match(ra,rb,mstart,rc,rd,re,rdepth+1)
339  #define RRETURN(ra) return ra  #define RRETURN(ra) return ra
340  #endif  #endif
341    
# Line 315  argument of match(), which never changes Line 350  argument of match(), which never changes
350    
351  #define RMATCH(ra,rb,rc,rd,re,rw)\  #define RMATCH(ra,rb,rc,rd,re,rw)\
352    {\    {\
353    heapframe *newframe = (heapframe *)(pcre_stack_malloc)(sizeof(heapframe));\    heapframe *newframe = frame->Xnextframe;\
354    if (newframe == NULL) RRETURN(PCRE_ERROR_NOMEMORY);\    if (newframe == NULL)\
355    frame->Xwhere = rw; \      {\
356        newframe = (heapframe *)(PUBL(stack_malloc))(sizeof(heapframe));\
357        if (newframe == NULL) RRETURN(PCRE_ERROR_NOMEMORY);\
358        newframe->Xnextframe = NULL;\
359        frame->Xnextframe = newframe;\
360        }\
361      frame->Xwhere = rw;\
362    newframe->Xeptr = ra;\    newframe->Xeptr = ra;\
363    newframe->Xecode = rb;\    newframe->Xecode = rb;\
364    newframe->Xmstart = mstart;\    newframe->Xmstart = mstart;\
   newframe->Xmarkptr = markptr;\  
365    newframe->Xoffset_top = rc;\    newframe->Xoffset_top = rc;\
366    newframe->Xeptrb = re;\    newframe->Xeptrb = re;\
367    newframe->Xrdepth = frame->Xrdepth + 1;\    newframe->Xrdepth = frame->Xrdepth + 1;\
# Line 337  argument of match(), which never changes Line 377  argument of match(), which never changes
377    {\    {\
378    heapframe *oldframe = frame;\    heapframe *oldframe = frame;\
379    frame = oldframe->Xprevframe;\    frame = oldframe->Xprevframe;\
   (pcre_stack_free)(oldframe);\  
380    if (frame != NULL)\    if (frame != NULL)\
381      {\      {\
382      rrc = ra;\      rrc = ra;\
# Line 351  argument of match(), which never changes Line 390  argument of match(), which never changes
390    
391  typedef struct heapframe {  typedef struct heapframe {
392    struct heapframe *Xprevframe;    struct heapframe *Xprevframe;
393      struct heapframe *Xnextframe;
394    
395    /* Function arguments that may change */    /* Function arguments that may change */
396    
397    USPTR Xeptr;    PCRE_PUCHAR Xeptr;
398    const uschar *Xecode;    const pcre_uchar *Xecode;
399    USPTR Xmstart;    PCRE_PUCHAR Xmstart;
   USPTR Xmarkptr;  
400    int Xoffset_top;    int Xoffset_top;
401    eptrblock *Xeptrb;    eptrblock *Xeptrb;
402    unsigned int Xrdepth;    unsigned int Xrdepth;
403    
404    /* Function local variables */    /* Function local variables */
405    
406    USPTR Xcallpat;    PCRE_PUCHAR Xcallpat;
407  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
408    USPTR Xcharptr;    PCRE_PUCHAR Xcharptr;
409  #endif  #endif
410    USPTR Xdata;    PCRE_PUCHAR Xdata;
411    USPTR Xnext;    PCRE_PUCHAR Xnext;
412    USPTR Xpp;    PCRE_PUCHAR Xpp;
413    USPTR Xprev;    PCRE_PUCHAR Xprev;
414    USPTR Xsaved_eptr;    PCRE_PUCHAR Xsaved_eptr;
415    
416    recursion_info Xnew_recursive;    recursion_info Xnew_recursive;
417    
# Line 382  typedef struct heapframe { Line 421  typedef struct heapframe {
421    
422  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
423    int Xprop_type;    int Xprop_type;
424    int Xprop_value;    unsigned int Xprop_value;
425    int Xprop_fail_result;    int Xprop_fail_result;
426    int Xoclength;    int Xoclength;
427    uschar Xocchars[8];    pcre_uchar Xocchars[6];
428  #endif  #endif
429    
430    int Xcodelink;    int Xcodelink;
# Line 395  typedef struct heapframe { Line 434  typedef struct heapframe {
434    int Xlength;    int Xlength;
435    int Xmax;    int Xmax;
436    int Xmin;    int Xmin;
437    int Xnumber;    unsigned int Xnumber;
438    int Xoffset;    int Xoffset;
439    int Xop;    unsigned int Xop;
440    int Xsave_capture_last;    pcre_int32 Xsave_capture_last;
441    int Xsave_offset1, Xsave_offset2, Xsave_offset3;    int Xsave_offset1, Xsave_offset2, Xsave_offset3;
442    int Xstacksave[REC_STACK_SAVE_MAX];    int Xstacksave[REC_STACK_SAVE_MAX];
443    
# Line 427  returns a negative (error) response, the Line 466  returns a negative (error) response, the
466  same response. */  same response. */
467    
468  /* These macros pack up tests that are used for partial matching, and which  /* These macros pack up tests that are used for partial matching, and which
469  appears several times in the code. We set the "hit end" flag if the pointer is  appear several times in the code. We set the "hit end" flag if the pointer is
470  at the end of the subject and also past the start of the subject (i.e.  at the end of the subject and also past the start of the subject (i.e.
471  something has been matched). For hard partial matching, we then return  something has been matched). For hard partial matching, we then return
472  immediately. The second one is used when we already know we are past the end of  immediately. The second one is used when we already know we are past the end of
# Line 438  the subject. */ Line 477  the subject. */
477        eptr > md->start_used_ptr) \        eptr > md->start_used_ptr) \
478      { \      { \
479      md->hitend = TRUE; \      md->hitend = TRUE; \
480      if (md->partial > 1) MRRETURN(PCRE_ERROR_PARTIAL); \      if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); \
481      }      }
482    
483  #define SCHECK_PARTIAL()\  #define SCHECK_PARTIAL()\
484    if (md->partial != 0 && eptr > md->start_used_ptr) \    if (md->partial != 0 && eptr > md->start_used_ptr) \
485      { \      { \
486      md->hitend = TRUE; \      md->hitend = TRUE; \
487      if (md->partial > 1) MRRETURN(PCRE_ERROR_PARTIAL); \      if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); \
488      }      }
489    
490    
491  /* Performance note: It might be tempting to extract commonly used fields from  /* Performance note: It might be tempting to extract commonly used fields from
492  the md structure (e.g. utf8, end_subject) into individual variables to improve  the md structure (e.g. utf, end_subject) into individual variables to improve
493  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
494  made performance worse.  made performance worse.
495    
# Line 459  Arguments: Line 498  Arguments:
498     ecode       pointer to current position in compiled code     ecode       pointer to current position in compiled code
499     mstart      pointer to the current match start position (can be modified     mstart      pointer to the current match start position (can be modified
500                   by encountering \K)                   by encountering \K)
    markptr     pointer to the most recent MARK name, or NULL  
501     offset_top  current top pointer     offset_top  current top pointer
502     md          pointer to "static" info for the match     md          pointer to "static" info for the match
503     eptrb       pointer to chain of blocks containing eptr at start of     eptrb       pointer to chain of blocks containing eptr at start of
# Line 474  Returns:       MATCH_MATCH if matched Line 512  Returns:       MATCH_MATCH if matched
512  */  */
513    
514  static int  static int
515  match(REGISTER USPTR eptr, REGISTER const uschar *ecode, USPTR mstart,  match(REGISTER PCRE_PUCHAR eptr, REGISTER const pcre_uchar *ecode,
516    const uschar *markptr, int offset_top, match_data *md, eptrblock *eptrb,    PCRE_PUCHAR mstart, int offset_top, match_data *md, eptrblock *eptrb,
517    unsigned int rdepth)    unsigned int rdepth)
518  {  {
519  /* These variables do not need to be preserved over recursion in this function,  /* These variables do not need to be preserved over recursion in this function,
# Line 484  so they can be ordinary variables in all Line 522  so they can be ordinary variables in all
522    
523  register int  rrc;         /* Returns from recursive calls */  register int  rrc;         /* Returns from recursive calls */
524  register int  i;           /* Used for loops not involving calls to RMATCH() */  register int  i;           /* Used for loops not involving calls to RMATCH() */
525  register unsigned int c;   /* Character values not kept over RMATCH() calls */  register pcre_uint32 c;    /* Character values not kept over RMATCH() calls */
526  register BOOL utf8;        /* Local copy of UTF-8 flag for speed */  register BOOL utf;         /* Local copy of UTF flag for speed */
527    
528  BOOL minimize, possessive; /* Quantifier options */  BOOL minimize, possessive; /* Quantifier options */
529  BOOL caseless;  BOOL caseless;
530  int condcode;  int condcode;
531    
532  /* 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
533  preserved over calls to RMATCH() are part of a "frame" which is obtained from  preserved over calls to RMATCH() are part of a "frame". We set up the top-level
534  heap storage. Set up the top-level frame here; others are obtained from the  frame on the stack here; subsequent instantiations are obtained from the heap
535  heap whenever RMATCH() does a "recursion". See the macro definitions above. */  whenever RMATCH() does a "recursion". See the macro definitions above. Putting
536    the top-level on the stack rather than malloc-ing them all gives a performance
537    boost in many cases where there is not much "recursion". */
538    
539  #ifdef NO_RECURSE  #ifdef NO_RECURSE
540  heapframe *frame = (heapframe *)(pcre_stack_malloc)(sizeof(heapframe));  heapframe *frame = (heapframe *)md->match_frames_base;
 if (frame == NULL) RRETURN(PCRE_ERROR_NOMEMORY);  
 frame->Xprevframe = NULL;            /* Marks the top level */  
541    
542  /* Copy in the original argument variables */  /* Copy in the original argument variables */
543    
544  frame->Xeptr = eptr;  frame->Xeptr = eptr;
545  frame->Xecode = ecode;  frame->Xecode = ecode;
546  frame->Xmstart = mstart;  frame->Xmstart = mstart;
 frame->Xmarkptr = markptr;  
547  frame->Xoffset_top = offset_top;  frame->Xoffset_top = offset_top;
548  frame->Xeptrb = eptrb;  frame->Xeptrb = eptrb;
549  frame->Xrdepth = rdepth;  frame->Xrdepth = rdepth;
# Line 520  HEAP_RECURSE: Line 557  HEAP_RECURSE:
557  #define eptr               frame->Xeptr  #define eptr               frame->Xeptr
558  #define ecode              frame->Xecode  #define ecode              frame->Xecode
559  #define mstart             frame->Xmstart  #define mstart             frame->Xmstart
 #define markptr            frame->Xmarkptr  
560  #define offset_top         frame->Xoffset_top  #define offset_top         frame->Xoffset_top
561  #define eptrb              frame->Xeptrb  #define eptrb              frame->Xeptrb
562  #define rdepth             frame->Xrdepth  #define rdepth             frame->Xrdepth
563    
564  /* Ditto for the local variables */  /* Ditto for the local variables */
565    
566  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
567  #define charptr            frame->Xcharptr  #define charptr            frame->Xcharptr
568  #endif  #endif
569  #define callpat            frame->Xcallpat  #define callpat            frame->Xcallpat
# Line 585  declarations can be cut out in a block. Line 621  declarations can be cut out in a block.
621  below are for variables that do not have to be preserved over a recursive call  below are for variables that do not have to be preserved over a recursive call
622  to RMATCH(). */  to RMATCH(). */
623    
624  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
625  const uschar *charptr;  const pcre_uchar *charptr;
626  #endif  #endif
627  const uschar *callpat;  const pcre_uchar *callpat;
628  const uschar *data;  const pcre_uchar *data;
629  const uschar *next;  const pcre_uchar *next;
630  USPTR         pp;  PCRE_PUCHAR       pp;
631  const uschar *prev;  const pcre_uchar *prev;
632  USPTR         saved_eptr;  PCRE_PUCHAR       saved_eptr;
633    
634  recursion_info new_recursive;  recursion_info new_recursive;
635    
# Line 603  BOOL prev_is_word; Line 639  BOOL prev_is_word;
639    
640  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
641  int prop_type;  int prop_type;
642  int prop_value;  unsigned int prop_value;
643  int prop_fail_result;  int prop_fail_result;
644  int oclength;  int oclength;
645  uschar occhars[8];  pcre_uchar occhars[6];
646  #endif  #endif
647    
648  int codelink;  int codelink;
# Line 614  int ctype; Line 650  int ctype;
650  int length;  int length;
651  int max;  int max;
652  int min;  int min;
653  int number;  unsigned int number;
654  int offset;  int offset;
655  int op;  unsigned int op;
656  int save_capture_last;  pcre_int32 save_capture_last;
657  int save_offset1, save_offset2, save_offset3;  int save_offset1, save_offset2, save_offset3;
658  int stacksave[REC_STACK_SAVE_MAX];  int stacksave[REC_STACK_SAVE_MAX];
659    
660  eptrblock newptrb;  eptrblock newptrb;
661    
662    /* There is a special fudge for calling match() in a way that causes it to
663    measure the size of its basic stack frame when the stack is being used for
664    recursion. The second argument (ecode) being NULL triggers this behaviour. It
665    cannot normally ever be NULL. The return is the negated value of the frame
666    size. */
667    
668    if (ecode == NULL)
669      {
670      if (rdepth == 0)
671        return match((PCRE_PUCHAR)&rdepth, NULL, NULL, 0, NULL, NULL, 1);
672      else
673        {
674        int len = (char *)&rdepth - (char *)eptr;
675        return (len > 0)? -len : len;
676        }
677      }
678  #endif     /* NO_RECURSE */  #endif     /* NO_RECURSE */
679    
680  /* To save space on the stack and in the heap frame, I have doubled up on some  /* To save space on the stack and in the heap frame, I have doubled up on some
# Line 634  the alternative names that are used. */ Line 687  the alternative names that are used. */
687  #define code_offset   codelink  #define code_offset   codelink
688  #define condassert    condition  #define condassert    condition
689  #define matched_once  prev_is_word  #define matched_once  prev_is_word
690    #define foc           number
691    #define save_mark     data
692    
693  /* These statements are here to stop the compiler complaining about unitialized  /* These statements are here to stop the compiler complaining about unitialized
694  variables. */  variables. */
# Line 659  defined). However, RMATCH isn't like a f Line 714  defined). However, RMATCH isn't like a f
714  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,
715  however, impact performance when true recursion is being used. */  however, impact performance when true recursion is being used. */
716    
717  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
718  utf8 = md->utf8;       /* Local copy of the flag */  utf = md->utf;       /* Local copy of the flag */
719  #else  #else
720  utf8 = FALSE;  utf = FALSE;
721  #endif  #endif
722    
723  /* 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
# Line 701  for (;;) Line 756  for (;;)
756    switch(op)    switch(op)
757      {      {
758      case OP_MARK:      case OP_MARK:
759      markptr = ecode + 2;      md->nomatch_mark = ecode + 2;
760      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      md->mark = NULL;    /* In case previously set by assertion */
761        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
762        eptrb, RM55);        eptrb, RM55);
763        if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
764             md->mark == NULL) md->mark = ecode + 2;
765    
766      /* A return of MATCH_SKIP_ARG means that matching failed at SKIP with an      /* A return of MATCH_SKIP_ARG means that matching failed at SKIP with an
767      argument, and we must check whether that argument matches this MARK's      argument, and we must check whether that argument matches this MARK's
# Line 712  for (;;) Line 770  for (;;)
770      position and return MATCH_SKIP. Otherwise, pass back the return code      position and return MATCH_SKIP. Otherwise, pass back the return code
771      unaltered. */      unaltered. */
772    
773      if (rrc == MATCH_SKIP_ARG &&      else if (rrc == MATCH_SKIP_ARG &&
774          strcmp((char *)markptr, (char *)(md->start_match_ptr)) == 0)          STRCMP_UC_UC_TEST(ecode + 2, md->start_match_ptr) == 0)
775        {        {
776        md->start_match_ptr = eptr;        md->start_match_ptr = eptr;
777        RRETURN(MATCH_SKIP);        RRETURN(MATCH_SKIP);
778        }        }
   
     if (md->mark == NULL) md->mark = markptr;  
779      RRETURN(rrc);      RRETURN(rrc);
780    
781      case OP_FAIL:      case OP_FAIL:
782      MRRETURN(MATCH_NOMATCH);      RRETURN(MATCH_NOMATCH);
   
     /* COMMIT overrides PRUNE, SKIP, and THEN */  
783    
784      case OP_COMMIT:      case OP_COMMIT:
785      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
786        eptrb, RM52);        eptrb, RM52);
787      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE &&      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
788          rrc != MATCH_SKIP && rrc != MATCH_SKIP_ARG &&      RRETURN(MATCH_COMMIT);
         rrc != MATCH_THEN)  
       RRETURN(rrc);  
     MRRETURN(MATCH_COMMIT);  
   
     /* PRUNE overrides THEN */  
789    
790      case OP_PRUNE:      case OP_PRUNE:
791      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
792        eptrb, RM51);        eptrb, RM51);
793      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
794      MRRETURN(MATCH_PRUNE);      RRETURN(MATCH_PRUNE);
795    
796      case OP_PRUNE_ARG:      case OP_PRUNE_ARG:
797      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      md->nomatch_mark = ecode + 2;
798        md->mark = NULL;    /* In case previously set by assertion */
799        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
800        eptrb, RM56);        eptrb, RM56);
801      if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);      if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
802      md->mark = ecode + 2;           md->mark == NULL) md->mark = ecode + 2;
803        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
804      RRETURN(MATCH_PRUNE);      RRETURN(MATCH_PRUNE);
805    
     /* SKIP overrides PRUNE and THEN */  
   
806      case OP_SKIP:      case OP_SKIP:
807      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
808        eptrb, RM53);        eptrb, RM53);
809      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
       RRETURN(rrc);  
810      md->start_match_ptr = eptr;   /* Pass back current position */      md->start_match_ptr = eptr;   /* Pass back current position */
811      MRRETURN(MATCH_SKIP);      RRETURN(MATCH_SKIP);
812    
813        /* Note that, for Perl compatibility, SKIP with an argument does NOT set
814        nomatch_mark. When a pattern match ends with a SKIP_ARG for which there was
815        not a matching mark, we have to re-run the match, ignoring the SKIP_ARG
816        that failed and any that preceed it (either they also failed, or were not
817        triggered). To do this, we maintain a count of executed SKIP_ARGs. If a
818        SKIP_ARG gets to top level, the match is re-run with md->ignore_skip_arg
819        set to the count of the one that failed. */
820    
821      case OP_SKIP_ARG:      case OP_SKIP_ARG:
822      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top, md,      md->skip_arg_count++;
823        if (md->skip_arg_count <= md->ignore_skip_arg)
824          {
825          ecode += PRIV(OP_lengths)[*ecode] + ecode[1];
826          break;
827          }
828        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
829        eptrb, RM57);        eptrb, RM57);
830      if (rrc != MATCH_NOMATCH && rrc != MATCH_PRUNE && rrc != MATCH_THEN)      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
831        RRETURN(rrc);  
   
832      /* Pass back the current skip name by overloading md->start_match_ptr and      /* Pass back the current skip name by overloading md->start_match_ptr and
833      returning the special MATCH_SKIP_ARG return code. This will either be      returning the special MATCH_SKIP_ARG return code. This will either be
834      caught by a matching MARK, or get to the top, where it is treated the same      caught by a matching MARK, or get to the top, where it causes a rematch
835      as PRUNE. */      with md->ignore_skip_arg set to the value of md->skip_arg_count. */
836    
837      md->start_match_ptr = ecode + 2;      md->start_match_ptr = ecode + 2;
838      RRETURN(MATCH_SKIP_ARG);      RRETURN(MATCH_SKIP_ARG);
# Line 780  for (;;) Line 842  for (;;)
842      match pointer to do this. */      match pointer to do this. */
843    
844      case OP_THEN:      case OP_THEN:
845      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
846        eptrb, RM54);        eptrb, RM54);
847      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
848      md->start_match_ptr = ecode;      md->start_match_ptr = ecode;
849      MRRETURN(MATCH_THEN);      RRETURN(MATCH_THEN);
850    
851      case OP_THEN_ARG:      case OP_THEN_ARG:
852      RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode] + ecode[1], offset_top,      md->nomatch_mark = ecode + 2;
853        md->mark = NULL;    /* In case previously set by assertion */
854        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top,
855        md, eptrb, RM58);        md, eptrb, RM58);
856        if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
857             md->mark == NULL) md->mark = ecode + 2;
858      if (rrc != MATCH_NOMATCH) RRETURN(rrc);      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
859      md->start_match_ptr = ecode;      md->start_match_ptr = ecode;
     md->mark = ecode + 2;  
860      RRETURN(MATCH_THEN);      RRETURN(MATCH_THEN);
861    
862      /* Handle an atomic group that does not contain any capturing parentheses.      /* Handle an atomic group that does not contain any capturing parentheses.
863      This can be handled like an assertion. Prior to 8.13, all atomic groups      This can be handled like an assertion. Prior to 8.13, all atomic groups
864      were handled this way. In 8.13, the code was changed as below for ONCE, so      were handled this way. In 8.13, the code was changed as below for ONCE, so
865      that backups pass through the group and thereby reset captured values.      that backups pass through the group and thereby reset captured values.
866      However, this uses a lot more stack, so in 8.20, atomic groups that do not      However, this uses a lot more stack, so in 8.20, atomic groups that do not
867      contain any captures generate OP_ONCE_NC, which can be handled in the old,      contain any captures generate OP_ONCE_NC, which can be handled in the old,
868      less stack intensive way.      less stack intensive way.
869    
870      Check the alternative branches in turn - the matching won't pass the KET      Check the alternative branches in turn - the matching won't pass the KET
# Line 810  for (;;) Line 875  for (;;)
875      case OP_ONCE_NC:      case OP_ONCE_NC:
876      prev = ecode;      prev = ecode;
877      saved_eptr = eptr;      saved_eptr = eptr;
878        save_mark = md->mark;
879      do      do
880        {        {
881        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM64);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM64);
# Line 821  for (;;) Line 887  for (;;)
887        if (rrc == MATCH_THEN)        if (rrc == MATCH_THEN)
888          {          {
889          next = ecode + GET(ecode,1);          next = ecode + GET(ecode,1);
890          if (md->start_match_ptr < next &&          if (md->start_match_ptr < next &&
891              (*ecode == OP_ALT || *next == OP_ALT))              (*ecode == OP_ALT || *next == OP_ALT))
892            rrc = MATCH_NOMATCH;            rrc = MATCH_NOMATCH;
893          }          }
894    
895        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
896        ecode += GET(ecode,1);        ecode += GET(ecode,1);
897          md->mark = save_mark;
898        }        }
899      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
900    
# Line 867  for (;;) Line 934  for (;;)
934        }        }
935      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
936        {        {
       md->match_function_type = MATCH_CBEGROUP;  
937        RMATCH(eptr, prev, offset_top, md, eptrb, RM66);        RMATCH(eptr, prev, offset_top, md, eptrb, RM66);
938        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
939        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
# Line 907  for (;;) Line 973  for (;;)
973        save_offset2 = md->offset_vector[offset+1];        save_offset2 = md->offset_vector[offset+1];
974        save_offset3 = md->offset_vector[md->offset_end - number];        save_offset3 = md->offset_vector[md->offset_end - number];
975        save_capture_last = md->capture_last;        save_capture_last = md->capture_last;
976          save_mark = md->mark;
977    
978        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));
979        md->offset_vector[md->offset_end - number] =        md->offset_vector[md->offset_end - number] =
# Line 915  for (;;) Line 982  for (;;)
982        for (;;)        for (;;)
983          {          {
984          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
985          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
986            eptrb, RM1);            eptrb, RM1);
987          if (rrc == MATCH_ONCE) break;  /* Backing up through an atomic group */          if (rrc == MATCH_ONCE) break;  /* Backing up through an atomic group */
988    
989          /* If we backed up to a THEN, check whether it is within the current          /* If we backed up to a THEN, check whether it is within the current
990          branch by comparing the address of the THEN that is passed back with          branch by comparing the address of the THEN that is passed back with
991          the end of the branch. If it is within the current branch, and the          the end of the branch. If it is within the current branch, and the
992          branch is one of two or more alternatives (it either starts or ends          branch is one of two or more alternatives (it either starts or ends
993          with OP_ALT), we have reached the limit of THEN's action, so convert          with OP_ALT), we have reached the limit of THEN's action, so convert
994          the return code to NOMATCH, which will cause normal backtracking to          the return code to NOMATCH, which will cause normal backtracking to
995          happen from now on. Otherwise, THEN is passed back to an outer          happen from now on. Otherwise, THEN is passed back to an outer
996          alternative. This implements Perl's treatment of parenthesized groups,          alternative. This implements Perl's treatment of parenthesized groups,
997          where a group not containing | does not affect the current alternative,          where a group not containing | does not affect the current alternative,
998          that is, (X) is NOT the same as (X|(*F)). */          that is, (X) is NOT the same as (X|(*F)). */
999    
1000          if (rrc == MATCH_THEN)          if (rrc == MATCH_THEN)
1001            {            {
1002            next = ecode + GET(ecode,1);            next = ecode + GET(ecode,1);
1003            if (md->start_match_ptr < next &&            if (md->start_match_ptr < next &&
1004                (*ecode == OP_ALT || *next == OP_ALT))                (*ecode == OP_ALT || *next == OP_ALT))
1005              rrc = MATCH_NOMATCH;              rrc = MATCH_NOMATCH;
1006            }            }
1007    
1008          /* Anything other than NOMATCH is passed back. */          /* Anything other than NOMATCH is passed back. */
1009    
1010          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1011          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
1012          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
1013            md->mark = save_mark;
1014          if (*ecode != OP_ALT) break;          if (*ecode != OP_ALT) break;
1015          }          }
1016    
# Line 953  for (;;) Line 1021  for (;;)
1021    
1022        /* At this point, rrc will be one of MATCH_ONCE or MATCH_NOMATCH. */        /* At this point, rrc will be one of MATCH_ONCE or MATCH_NOMATCH. */
1023    
       if (md->mark == NULL) md->mark = markptr;  
1024        RRETURN(rrc);        RRETURN(rrc);
1025        }        }
1026    
# Line 995  for (;;) Line 1062  for (;;)
1062    
1063      for (;;)      for (;;)
1064        {        {
1065        if (op >= OP_SBRA || op == OP_ONCE) md->match_function_type = MATCH_CBEGROUP;        if (op >= OP_SBRA || op == OP_ONCE)
1066            md->match_function_type = MATCH_CBEGROUP;
1067    
1068        /* If this is not a possibly empty group, and there are no (*THEN)s in        /* If this is not a possibly empty group, and there are no (*THEN)s in
1069        the pattern, and this is the final alternative, optimize as described        the pattern, and this is the final alternative, optimize as described
# Line 1003  for (;;) Line 1071  for (;;)
1071    
1072        else if (!md->hasthen && ecode[GET(ecode, 1)] != OP_ALT)        else if (!md->hasthen && ecode[GET(ecode, 1)] != OP_ALT)
1073          {          {
1074          ecode += _pcre_OP_lengths[*ecode];          ecode += PRIV(OP_lengths)[*ecode];
1075          goto TAIL_RECURSE;          goto TAIL_RECURSE;
1076          }          }
1077    
1078        /* In all other cases, we have to make another call to match(). */        /* In all other cases, we have to make another call to match(). */
1079    
1080        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, eptrb,        save_mark = md->mark;
1081          save_capture_last = md->capture_last;
1082          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb,
1083          RM2);          RM2);
1084    
1085        /* See comment in the code for capturing groups above about handling        /* See comment in the code for capturing groups above about handling
1086        THEN. */        THEN. */
1087    
1088        if (rrc == MATCH_THEN)        if (rrc == MATCH_THEN)
1089          {          {
1090          next = ecode + GET(ecode,1);          next = ecode + GET(ecode,1);
1091          if (md->start_match_ptr < next &&          if (md->start_match_ptr < next &&
1092              (*ecode == OP_ALT || *next == OP_ALT))              (*ecode == OP_ALT || *next == OP_ALT))
1093            rrc = MATCH_NOMATCH;            rrc = MATCH_NOMATCH;
1094          }          }
1095    
1096        if (rrc != MATCH_NOMATCH)        if (rrc != MATCH_NOMATCH)
1097          {          {
1098          if (rrc == MATCH_ONCE)          if (rrc == MATCH_ONCE)
1099            {            {
1100            const uschar *scode = ecode;            const pcre_uchar *scode = ecode;
1101            if (*scode != OP_ONCE)           /* If not at start, find it */            if (*scode != OP_ONCE)           /* If not at start, find it */
1102              {              {
1103              while (*scode == OP_ALT) scode += GET(scode, 1);              while (*scode == OP_ALT) scode += GET(scode, 1);
# Line 1038  for (;;) Line 1108  for (;;)
1108          RRETURN(rrc);          RRETURN(rrc);
1109          }          }
1110        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1111          md->mark = save_mark;
1112        if (*ecode != OP_ALT) break;        if (*ecode != OP_ALT) break;
1113          md->capture_last = save_capture_last;
1114        }        }
1115    
     if (md->mark == NULL) md->mark = markptr;  
1116      RRETURN(MATCH_NOMATCH);      RRETURN(MATCH_NOMATCH);
1117    
1118      /* Handle possessive capturing brackets with an unlimited repeat. We come      /* Handle possessive capturing brackets with an unlimited repeat. We come
# Line 1070  for (;;) Line 1141  for (;;)
1141      if (offset < md->offset_max)      if (offset < md->offset_max)
1142        {        {
1143        matched_once = FALSE;        matched_once = FALSE;
1144        code_offset = ecode - md->start_code;        code_offset = (int)(ecode - md->start_code);
1145    
1146        save_offset1 = md->offset_vector[offset];        save_offset1 = md->offset_vector[offset];
1147        save_offset2 = md->offset_vector[offset+1];        save_offset2 = md->offset_vector[offset+1];
# Line 1093  for (;;) Line 1164  for (;;)
1164          md->offset_vector[md->offset_end - number] =          md->offset_vector[md->offset_end - number] =
1165            (int)(eptr - md->start_subject);            (int)(eptr - md->start_subject);
1166          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;          if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
1167          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
1168            eptrb, RM63);            eptrb, RM63);
1169          if (rrc == MATCH_KETRPOS)          if (rrc == MATCH_KETRPOS)
1170            {            {
# Line 1104  for (;;) Line 1175  for (;;)
1175            matched_once = TRUE;            matched_once = TRUE;
1176            continue;            continue;
1177            }            }
1178    
1179          /* See comment in the code for capturing groups above about handling          /* See comment in the code for capturing groups above about handling
1180          THEN. */          THEN. */
1181    
1182          if (rrc == MATCH_THEN)          if (rrc == MATCH_THEN)
1183            {            {
1184            next = ecode + GET(ecode,1);            next = ecode + GET(ecode,1);
1185            if (md->start_match_ptr < next &&            if (md->start_match_ptr < next &&
1186                (*ecode == OP_ALT || *next == OP_ALT))                (*ecode == OP_ALT || *next == OP_ALT))
1187              rrc = MATCH_NOMATCH;              rrc = MATCH_NOMATCH;
1188            }            }
1189    
1190          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1191          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
# Line 1129  for (;;) Line 1200  for (;;)
1200          md->offset_vector[md->offset_end - number] = save_offset3;          md->offset_vector[md->offset_end - number] = save_offset3;
1201          }          }
1202    
       if (md->mark == NULL) md->mark = markptr;  
1203        if (allow_zero || matched_once)        if (allow_zero || matched_once)
1204          {          {
1205          ecode += 1 + LINK_SIZE;          ecode += 1 + LINK_SIZE;
# Line 1161  for (;;) Line 1231  for (;;)
1231    
1232      POSSESSIVE_NON_CAPTURE:      POSSESSIVE_NON_CAPTURE:
1233      matched_once = FALSE;      matched_once = FALSE;
1234      code_offset = ecode - md->start_code;      code_offset = (int)(ecode - md->start_code);
1235        save_capture_last = md->capture_last;
1236    
1237      for (;;)      for (;;)
1238        {        {
1239        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
1240        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
1241          eptrb, RM48);          eptrb, RM48);
1242        if (rrc == MATCH_KETRPOS)        if (rrc == MATCH_KETRPOS)
1243          {          {
# Line 1176  for (;;) Line 1247  for (;;)
1247          matched_once = TRUE;          matched_once = TRUE;
1248          continue;          continue;
1249          }          }
1250    
1251        /* See comment in the code for capturing groups above about handling        /* See comment in the code for capturing groups above about handling
1252        THEN. */        THEN. */
1253    
1254        if (rrc == MATCH_THEN)        if (rrc == MATCH_THEN)
1255          {          {
1256          next = ecode + GET(ecode,1);          next = ecode + GET(ecode,1);
1257          if (md->start_match_ptr < next &&          if (md->start_match_ptr < next &&
1258              (*ecode == OP_ALT || *next == OP_ALT))              (*ecode == OP_ALT || *next == OP_ALT))
1259            rrc = MATCH_NOMATCH;            rrc = MATCH_NOMATCH;
1260          }          }
1261    
1262        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1263        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1264        if (*ecode != OP_ALT) break;        if (*ecode != OP_ALT) break;
1265          md->capture_last = save_capture_last;
1266        }        }
1267    
1268      if (matched_once || allow_zero)      if (matched_once || allow_zero)
# Line 1216  for (;;) Line 1288  for (;;)
1288    
1289      if (ecode[LINK_SIZE+1] == OP_CALLOUT)      if (ecode[LINK_SIZE+1] == OP_CALLOUT)
1290        {        {
1291        if (pcre_callout != NULL)        if (PUBL(callout) != NULL)
1292          {          {
1293          pcre_callout_block cb;          PUBL(callout_block) cb;
1294          cb.version          = 2;   /* Version 1 of the callout block */          cb.version          = 2;   /* Version 1 of the callout block */
1295          cb.callout_number   = ecode[LINK_SIZE+2];          cb.callout_number   = ecode[LINK_SIZE+2];
1296          cb.offset_vector    = md->offset_vector;          cb.offset_vector    = md->offset_vector;
1297    #if defined COMPILE_PCRE8
1298          cb.subject          = (PCRE_SPTR)md->start_subject;          cb.subject          = (PCRE_SPTR)md->start_subject;
1299    #elif defined COMPILE_PCRE16
1300            cb.subject          = (PCRE_SPTR16)md->start_subject;
1301    #elif defined COMPILE_PCRE32
1302            cb.subject          = (PCRE_SPTR32)md->start_subject;
1303    #endif
1304          cb.subject_length   = (int)(md->end_subject - md->start_subject);          cb.subject_length   = (int)(md->end_subject - md->start_subject);
1305          cb.start_match      = (int)(mstart - md->start_subject);          cb.start_match      = (int)(mstart - md->start_subject);
1306          cb.current_position = (int)(eptr - md->start_subject);          cb.current_position = (int)(eptr - md->start_subject);
1307          cb.pattern_position = GET(ecode, LINK_SIZE + 3);          cb.pattern_position = GET(ecode, LINK_SIZE + 3);
1308          cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);          cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);
1309          cb.capture_top      = offset_top/2;          cb.capture_top      = offset_top/2;
1310          cb.capture_last     = md->capture_last;          cb.capture_last     = md->capture_last & CAPLMASK;
1311            /* Internal change requires this for API compatibility. */
1312            if (cb.capture_last == 0) cb.capture_last = -1;
1313          cb.callout_data     = md->callout_data;          cb.callout_data     = md->callout_data;
1314          cb.mark             = markptr;          cb.mark             = md->nomatch_mark;
1315          if ((rrc = (*pcre_callout)(&cb)) > 0) MRRETURN(MATCH_NOMATCH);          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1316          if (rrc < 0) RRETURN(rrc);          if (rrc < 0) RRETURN(rrc);
1317          }          }
1318        ecode += _pcre_OP_lengths[OP_CALLOUT];        ecode += PRIV(OP_lengths)[OP_CALLOUT];
1319          codelink -= PRIV(OP_lengths)[OP_CALLOUT];
1320        }        }
1321    
1322      condcode = ecode[LINK_SIZE+1];      condcode = ecode[LINK_SIZE+1];
# Line 1251  for (;;) Line 1332  for (;;)
1332          }          }
1333        else        else
1334          {          {
1335          int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/          unsigned int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/
1336          condition =  (recno == RREF_ANY || recno == md->recursive->group_num);          condition = (recno == RREF_ANY || recno == md->recursive->group_num);
1337    
1338          /* If the test is for recursion into a specific subpattern, and it is          /* If the test is for recursion into a specific subpattern, and it is
1339          false, but the test was set up by name, scan the table to see if the          false, but the test was set up by name, scan the table to see if the
1340          name refers to any other numbers, and test them. The condition is true          name refers to any other numbers, and test them. The condition is true
1341          if any one is set. */          if any one is set. */
1342    
1343          if (!condition && condcode == OP_NRREF && recno != RREF_ANY)          if (!condition && condcode == OP_NRREF)
1344            {            {
1345            uschar *slotA = md->name_table;            pcre_uchar *slotA = md->name_table;
1346            for (i = 0; i < md->name_count; i++)            for (i = 0; i < md->name_count; i++)
1347              {              {
1348              if (GET2(slotA, 0) == recno) break;              if (GET2(slotA, 0) == recno) break;
# Line 1274  for (;;) Line 1355  for (;;)
1355    
1356            if (i < md->name_count)            if (i < md->name_count)
1357              {              {
1358              uschar *slotB = slotA;              pcre_uchar *slotB = slotA;
1359              while (slotB > md->name_table)              while (slotB > md->name_table)
1360                {                {
1361                slotB -= md->name_entry_size;                slotB -= md->name_entry_size;
1362                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1363                  {                  {
1364                  condition = GET2(slotB, 0) == md->recursive->group_num;                  condition = GET2(slotB, 0) == md->recursive->group_num;
1365                  if (condition) break;                  if (condition) break;
# Line 1294  for (;;) Line 1375  for (;;)
1375                for (i++; i < md->name_count; i++)                for (i++; i < md->name_count; i++)
1376                  {                  {
1377                  slotB += md->name_entry_size;                  slotB += md->name_entry_size;
1378                  if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                  if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1379                    {                    {
1380                    condition = GET2(slotB, 0) == md->recursive->group_num;                    condition = GET2(slotB, 0) == md->recursive->group_num;
1381                    if (condition) break;                    if (condition) break;
# Line 1307  for (;;) Line 1388  for (;;)
1388    
1389          /* Chose branch according to the condition */          /* Chose branch according to the condition */
1390    
1391          ecode += condition? 3 : GET(ecode, 1);          ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
1392          }          }
1393        }        }
1394    
# Line 1323  for (;;) Line 1404  for (;;)
1404    
1405        if (!condition && condcode == OP_NCREF)        if (!condition && condcode == OP_NCREF)
1406          {          {
1407          int refno = offset >> 1;          unsigned int refno = offset >> 1;
1408          uschar *slotA = md->name_table;          pcre_uchar *slotA = md->name_table;
1409    
1410          for (i = 0; i < md->name_count; i++)          for (i = 0; i < md->name_count; i++)
1411            {            {
# Line 1338  for (;;) Line 1419  for (;;)
1419    
1420          if (i < md->name_count)          if (i < md->name_count)
1421            {            {
1422            uschar *slotB = slotA;            pcre_uchar *slotB = slotA;
1423            while (slotB > md->name_table)            while (slotB > md->name_table)
1424              {              {
1425              slotB -= md->name_entry_size;              slotB -= md->name_entry_size;
1426              if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)              if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1427                {                {
1428                offset = GET2(slotB, 0) << 1;                offset = GET2(slotB, 0) << 1;
1429                condition = offset < offset_top &&                condition = offset < offset_top &&
# Line 1360  for (;;) Line 1441  for (;;)
1441              for (i++; i < md->name_count; i++)              for (i++; i < md->name_count; i++)
1442                {                {
1443                slotB += md->name_entry_size;                slotB += md->name_entry_size;
1444                if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0)                if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
1445                  {                  {
1446                  offset = GET2(slotB, 0) << 1;                  offset = GET2(slotB, 0) << 1;
1447                  condition = offset < offset_top &&                  condition = offset < offset_top &&
# Line 1375  for (;;) Line 1456  for (;;)
1456    
1457        /* Chose branch according to the condition */        /* Chose branch according to the condition */
1458    
1459        ecode += condition? 3 : GET(ecode, 1);        ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
1460        }        }
1461    
1462      else if (condcode == OP_DEF)     /* DEFINE - always false */      else if (condcode == OP_DEF)     /* DEFINE - always false */
# Line 1400  for (;;) Line 1481  for (;;)
1481          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
1482          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
1483          }          }
1484    
1485        /* PCRE doesn't allow the effect of (*THEN) to escape beyond an        /* PCRE doesn't allow the effect of (*THEN) to escape beyond an
1486        assertion; it is therefore treated as NOMATCH. */        assertion; it is therefore treated as NOMATCH. */
1487    
1488        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1489          {          {
1490          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
1491          }          }
# Line 1432  for (;;) Line 1513  for (;;)
1513          ecode += 1 + LINK_SIZE;          ecode += 1 + LINK_SIZE;
1514          goto TAIL_RECURSE;          goto TAIL_RECURSE;
1515          }          }
1516    
1517        md->match_function_type = MATCH_CBEGROUP;        md->match_function_type = MATCH_CBEGROUP;
1518        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM49);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM49);
1519        RRETURN(rrc);        RRETURN(rrc);
# Line 1451  for (;;) Line 1532  for (;;)
1532      to close any currently open capturing brackets. */      to close any currently open capturing brackets. */
1533    
1534      case OP_CLOSE:      case OP_CLOSE:
1535      number = GET2(ecode, 1);      number = GET2(ecode, 1);   /* Must be less than 65536 */
1536      offset = number << 1;      offset = number << 1;
1537    
1538  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
# Line 1459  for (;;) Line 1540  for (;;)
1540        printf("\n");        printf("\n");
1541  #endif  #endif
1542    
1543      md->capture_last = number;      md->capture_last = (md->capture_last & OVFLMASK) | number;
1544      if (offset >= md->offset_max) md->offset_overflow = TRUE; else      if (offset >= md->offset_max) md->capture_last |= OVFLBIT; else
1545        {        {
1546        md->offset_vector[offset] =        md->offset_vector[offset] =
1547          md->offset_vector[md->offset_end - number];          md->offset_vector[md->offset_end - number];
1548        md->offset_vector[offset+1] = (int)(eptr - md->start_subject);        md->offset_vector[offset+1] = (int)(eptr - md->start_subject);
1549        if (offset_top <= offset) offset_top = offset + 2;        if (offset_top <= offset) offset_top = offset + 2;
1550        }        }
1551      ecode += 3;      ecode += 1 + IMM2_SIZE;
1552      break;      break;
1553    
1554    
# Line 1487  for (;;) Line 1568  for (;;)
1568           (md->notempty ||           (md->notempty ||
1569             (md->notempty_atstart &&             (md->notempty_atstart &&
1570               mstart == md->start_subject + md->start_offset)))               mstart == md->start_subject + md->start_offset)))
1571        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
1572    
1573      /* Otherwise, we have a match. */      /* Otherwise, we have a match. */
1574    
# Line 1496  for (;;) Line 1577  for (;;)
1577      md->start_match_ptr = mstart;       /* and the start (\K can modify) */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
1578    
1579      /* For some reason, the macros don't work properly if an expression is      /* For some reason, the macros don't work properly if an expression is
1580      given as the argument to MRRETURN when the heap is in use. */      given as the argument to RRETURN when the heap is in use. */
1581    
1582      rrc = (op == OP_END)? MATCH_MATCH : MATCH_ACCEPT;      rrc = (op == OP_END)? MATCH_MATCH : MATCH_ACCEPT;
1583      MRRETURN(rrc);      RRETURN(rrc);
1584    
1585      /* Assertion brackets. Check the alternative branches in turn - the      /* Assertion brackets. Check the alternative branches in turn - the
1586      matching won't pass the KET for an assertion. If any one branch matches,      matching won't pass the KET for an assertion. If any one branch matches,
# Line 1514  for (;;) Line 1595  for (;;)
1595    
1596      case OP_ASSERT:      case OP_ASSERT:
1597      case OP_ASSERTBACK:      case OP_ASSERTBACK:
1598        save_mark = md->mark;
1599      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1600        {        {
1601        condassert = TRUE;        condassert = TRUE;
# Line 1527  for (;;) Line 1609  for (;;)
1609        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
1610          {          {
1611          mstart = md->start_match_ptr;   /* In case \K reset it */          mstart = md->start_match_ptr;   /* In case \K reset it */
         markptr = md->mark;  
1612          break;          break;
1613          }          }
1614          md->mark = save_mark;
1615        /* PCRE does not allow THEN to escape beyond an assertion; it is treated  
1616        as NOMATCH. */        /* A COMMIT failure must fail the entire assertion, without trying any
1617          subsequent branches. */
1618    
1619          if (rrc == MATCH_COMMIT) RRETURN(MATCH_NOMATCH);
1620    
1621          /* PCRE does not allow THEN to escape beyond an assertion; it
1622          is treated as NOMATCH. */
1623    
1624        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1625        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1626        }        }
1627      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
1628    
1629      if (*ecode == OP_KET) MRRETURN(MATCH_NOMATCH);      if (*ecode == OP_KET) RRETURN(MATCH_NOMATCH);
1630    
1631      /* If checking an assertion for a condition, return MATCH_MATCH. */      /* If checking an assertion for a condition, return MATCH_MATCH. */
1632    
# Line 1559  for (;;) Line 1646  for (;;)
1646    
1647      case OP_ASSERT_NOT:      case OP_ASSERT_NOT:
1648      case OP_ASSERTBACK_NOT:      case OP_ASSERTBACK_NOT:
1649        save_mark = md->mark;
1650      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1651        {        {
1652        condassert = TRUE;        condassert = TRUE;
# Line 1569  for (;;) Line 1657  for (;;)
1657      do      do
1658        {        {
1659        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);
1660        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) MRRETURN(MATCH_NOMATCH);        md->mark = save_mark;
1661          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) RRETURN(MATCH_NOMATCH);
1662        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)
1663          {          {
1664          do ecode += GET(ecode,1); while (*ecode == OP_ALT);          do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1665          break;          break;
1666          }          }
1667    
1668        /* PCRE does not allow THEN to escape beyond an assertion; it is treated        /* PCRE does not allow THEN to escape beyond an assertion; it is treated
1669        as NOMATCH. */        as NOMATCH. */
1670    
1671        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
# Line 1595  for (;;) Line 1684  for (;;)
1684      back a number of characters, not bytes. */      back a number of characters, not bytes. */
1685    
1686      case OP_REVERSE:      case OP_REVERSE:
1687  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
1688      if (utf8)      if (utf)
1689        {        {
1690        i = GET(ecode, 1);        i = GET(ecode, 1);
1691        while (i-- > 0)        while (i-- > 0)
1692          {          {
1693          eptr--;          eptr--;
1694          if (eptr < md->start_subject) MRRETURN(MATCH_NOMATCH);          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1695          BACKCHAR(eptr);          BACKCHAR(eptr);
1696          }          }
1697        }        }
# Line 1613  for (;;) Line 1702  for (;;)
1702    
1703        {        {
1704        eptr -= GET(ecode, 1);        eptr -= GET(ecode, 1);
1705        if (eptr < md->start_subject) MRRETURN(MATCH_NOMATCH);        if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
1706        }        }
1707    
1708      /* Save the earliest consulted character, then skip to next op code */      /* Save the earliest consulted character, then skip to next op code */
# Line 1627  for (;;) Line 1716  for (;;)
1716      function is able to force a failure. */      function is able to force a failure. */
1717    
1718      case OP_CALLOUT:      case OP_CALLOUT:
1719      if (pcre_callout != NULL)      if (PUBL(callout) != NULL)
1720        {        {
1721        pcre_callout_block cb;        PUBL(callout_block) cb;
1722        cb.version          = 2;   /* Version 1 of the callout block */        cb.version          = 2;   /* Version 1 of the callout block */
1723        cb.callout_number   = ecode[1];        cb.callout_number   = ecode[1];
1724        cb.offset_vector    = md->offset_vector;        cb.offset_vector    = md->offset_vector;
1725    #if defined COMPILE_PCRE8
1726        cb.subject          = (PCRE_SPTR)md->start_subject;        cb.subject          = (PCRE_SPTR)md->start_subject;
1727    #elif defined COMPILE_PCRE16
1728          cb.subject          = (PCRE_SPTR16)md->start_subject;
1729    #elif defined COMPILE_PCRE32
1730          cb.subject          = (PCRE_SPTR32)md->start_subject;
1731    #endif
1732        cb.subject_length   = (int)(md->end_subject - md->start_subject);        cb.subject_length   = (int)(md->end_subject - md->start_subject);
1733        cb.start_match      = (int)(mstart - md->start_subject);        cb.start_match      = (int)(mstart - md->start_subject);
1734        cb.current_position = (int)(eptr - md->start_subject);        cb.current_position = (int)(eptr - md->start_subject);
1735        cb.pattern_position = GET(ecode, 2);        cb.pattern_position = GET(ecode, 2);
1736        cb.next_item_length = GET(ecode, 2 + LINK_SIZE);        cb.next_item_length = GET(ecode, 2 + LINK_SIZE);
1737        cb.capture_top      = offset_top/2;        cb.capture_top      = offset_top/2;
1738        cb.capture_last     = md->capture_last;        cb.capture_last     = md->capture_last & CAPLMASK;
1739          /* Internal change requires this for API compatibility. */
1740          if (cb.capture_last == 0) cb.capture_last = -1;
1741        cb.callout_data     = md->callout_data;        cb.callout_data     = md->callout_data;
1742        cb.mark             = markptr;        cb.mark             = md->nomatch_mark;
1743        if ((rrc = (*pcre_callout)(&cb)) > 0) MRRETURN(MATCH_NOMATCH);        if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1744        if (rrc < 0) RRETURN(rrc);        if (rrc < 0) RRETURN(rrc);
1745        }        }
1746      ecode += 2 + 2*LINK_SIZE;      ecode += 2 + 2*LINK_SIZE;
# Line 1669  for (;;) Line 1766  for (;;)
1766      case OP_RECURSE:      case OP_RECURSE:
1767        {        {
1768        recursion_info *ri;        recursion_info *ri;
1769        int recno;        unsigned int recno;
1770    
1771        callpat = md->start_code + GET(ecode, 1);        callpat = md->start_code + GET(ecode, 1);
1772        recno = (callpat == md->start_code)? 0 :        recno = (callpat == md->start_code)? 0 :
# Line 1686  for (;;) Line 1783  for (;;)
1783        /* Add to "recursing stack" */        /* Add to "recursing stack" */
1784    
1785        new_recursive.group_num = recno;        new_recursive.group_num = recno;
1786          new_recursive.saved_capture_last = md->capture_last;
1787        new_recursive.subject_position = eptr;        new_recursive.subject_position = eptr;
1788        new_recursive.prevrec = md->recursive;        new_recursive.prevrec = md->recursive;
1789        md->recursive = &new_recursive;        md->recursive = &new_recursive;
# Line 1702  for (;;) Line 1800  for (;;)
1800        else        else
1801          {          {
1802          new_recursive.offset_save =          new_recursive.offset_save =
1803            (int *)(pcre_malloc)(new_recursive.saved_max * sizeof(int));            (int *)(PUBL(malloc))(new_recursive.saved_max * sizeof(int));
1804          if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY);          if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY);
1805          }          }
1806        memcpy(new_recursive.offset_save, md->offset_vector,        memcpy(new_recursive.offset_save, md->offset_vector,
1807              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
1808    
1809        /* OK, now we can do the recursion. After processing each alternative,        /* OK, now we can do the recursion. After processing each alternative,
1810        restore the offset data. If there were nested recursions, md->recursive        restore the offset data and the last captured value. If there were nested
1811        might be changed, so reset it before looping. */        recursions, md->recursive might be changed, so reset it before looping.
1812          */
1813    
1814        DPRINTF(("Recursing into group %d\n", new_recursive.group_num));        DPRINTF(("Recursing into group %d\n", new_recursive.group_num));
1815        cbegroup = (*callpat >= OP_SBRA);        cbegroup = (*callpat >= OP_SBRA);
1816        do        do
1817          {          {
1818          if (cbegroup) md->match_function_type = MATCH_CBEGROUP;          if (cbegroup) md->match_function_type = MATCH_CBEGROUP;
1819          RMATCH(eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,          RMATCH(eptr, callpat + PRIV(OP_lengths)[*callpat], offset_top,
1820            md, eptrb, RM6);            md, eptrb, RM6);
1821          memcpy(md->offset_vector, new_recursive.offset_save,          memcpy(md->offset_vector, new_recursive.offset_save,
1822              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
1823            md->capture_last = new_recursive.saved_capture_last;
1824          md->recursive = new_recursive.prevrec;          md->recursive = new_recursive.prevrec;
1825          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
1826            {            {
1827            DPRINTF(("Recursion matched\n"));            DPRINTF(("Recursion matched\n"));
1828            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
1829              (pcre_free)(new_recursive.offset_save);              (PUBL(free))(new_recursive.offset_save);
1830    
1831            /* Set where we got to in the subject, and reset the start in case            /* Set where we got to in the subject, and reset the start in case
1832            it was changed by \K. This *is* propagated back out of a recursion,            it was changed by \K. This *is* propagated back out of a recursion,
# Line 1737  for (;;) Line 1837  for (;;)
1837            goto RECURSION_MATCHED;        /* Exit loop; end processing */            goto RECURSION_MATCHED;        /* Exit loop; end processing */
1838            }            }
1839    
1840          /* PCRE does not allow THEN to escape beyond a recursion; it is treated          /* PCRE does not allow THEN, SKIP, PRUNE or COMMIT to escape beyond a
1841          as NOMATCH. */          recursion; they are treated as NOMATCH. These codes are defined in a
1842            range that can be tested for. Any other return code is an error. */
1843    
1844          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)          else if (rrc != MATCH_NOMATCH &&
1845                     (rrc < MATCH_BACKTRACK_MIN || rrc > MATCH_BACKTRACK_MAX))
1846            {            {
1847            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1848            if (new_recursive.offset_save != stacksave)            if (new_recursive.offset_save != stacksave)
1849              (pcre_free)(new_recursive.offset_save);              (PUBL(free))(new_recursive.offset_save);
1850            RRETURN(rrc);            RRETURN(rrc);
1851            }            }
1852    
# Line 1756  for (;;) Line 1858  for (;;)
1858        DPRINTF(("Recursion didn't match\n"));        DPRINTF(("Recursion didn't match\n"));
1859        md->recursive = new_recursive.prevrec;        md->recursive = new_recursive.prevrec;
1860        if (new_recursive.offset_save != stacksave)        if (new_recursive.offset_save != stacksave)
1861          (pcre_free)(new_recursive.offset_save);          (PUBL(free))(new_recursive.offset_save);
1862        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
1863        }        }
1864    
1865      RECURSION_MATCHED:      RECURSION_MATCHED:
# Line 1826  for (;;) Line 1928  for (;;)
1928        }        }
1929      else saved_eptr = NULL;      else saved_eptr = NULL;
1930    
1931      /* If we are at the end of an assertion group or a non-capturing atomic      /* If we are at the end of an assertion group or a non-capturing atomic
1932      group, stop matching and return MATCH_MATCH, but record the current high      group, stop matching and return MATCH_MATCH, but record the current high
1933      water mark for use by positive assertions. We also need to record the match      water mark for use by positive assertions. We also need to record the match
1934      start in case it was changed by \K. */      start in case it was changed by \K. */
1935    
1936      if ((*prev >= OP_ASSERT && *prev <= OP_ASSERTBACK_NOT) ||      if ((*prev >= OP_ASSERT && *prev <= OP_ASSERTBACK_NOT) ||
1937           *prev == OP_ONCE_NC)           *prev == OP_ONCE_NC)
1938        {        {
1939        md->end_match_ptr = eptr;      /* For ONCE_NC */        md->end_match_ptr = eptr;      /* For ONCE_NC */
1940        md->end_offset_top = offset_top;        md->end_offset_top = offset_top;
1941        md->start_match_ptr = mstart;        md->start_match_ptr = mstart;
1942        MRRETURN(MATCH_MATCH);         /* Sets md->mark */        RRETURN(MATCH_MATCH);         /* Sets md->mark */
1943        }        }
1944    
1945      /* For capturing groups we have to check the group number back at the start      /* For capturing groups we have to check the group number back at the start
# Line 1870  for (;;) Line 1972  for (;;)
1972    
1973        /* Deal with capturing */        /* Deal with capturing */
1974    
1975        md->capture_last = number;        md->capture_last = (md->capture_last & OVFLMASK) | number;
1976        if (offset >= md->offset_max) md->offset_overflow = TRUE; else        if (offset >= md->offset_max) md->capture_last |= OVFLBIT; else
1977          {          {
1978          /* If offset is greater than offset_top, it means that we are          /* If offset is greater than offset_top, it means that we are
1979          "skipping" a capturing group, and that group's offsets must be marked          "skipping" a capturing group, and that group's offsets must be marked
# Line 1951  for (;;) Line 2053  for (;;)
2053          }          }
2054        if (*prev >= OP_SBRA)    /* Could match an empty string */        if (*prev >= OP_SBRA)    /* Could match an empty string */
2055          {          {
         md->match_function_type = MATCH_CBEGROUP;  
2056          RMATCH(eptr, prev, offset_top, md, eptrb, RM50);          RMATCH(eptr, prev, offset_top, md, eptrb, RM50);
2057          RRETURN(rrc);          RRETURN(rrc);
2058          }          }
# Line 1960  for (;;) Line 2061  for (;;)
2061        }        }
2062      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
2063        {        {
       if (*prev >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;  
2064        RMATCH(eptr, prev, offset_top, md, eptrb, RM13);        RMATCH(eptr, prev, offset_top, md, eptrb, RM13);
2065        if (rrc == MATCH_ONCE && md->once_target == prev) rrc = MATCH_NOMATCH;        if (rrc == MATCH_ONCE && md->once_target == prev) rrc = MATCH_NOMATCH;
2066        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
# Line 1979  for (;;) Line 2079  for (;;)
2079      /* Not multiline mode: start of subject assertion, unless notbol. */      /* Not multiline mode: start of subject assertion, unless notbol. */
2080    
2081      case OP_CIRC:      case OP_CIRC:
2082      if (md->notbol && eptr == md->start_subject) MRRETURN(MATCH_NOMATCH);      if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH);
2083    
2084      /* Start of subject assertion */      /* Start of subject assertion */
2085    
2086      case OP_SOD:      case OP_SOD:
2087      if (eptr != md->start_subject) MRRETURN(MATCH_NOMATCH);      if (eptr != md->start_subject) RRETURN(MATCH_NOMATCH);
2088      ecode++;      ecode++;
2089      break;      break;
2090    
2091      /* Multiline mode: start of subject unless notbol, or after any newline. */      /* Multiline mode: start of subject unless notbol, or after any newline. */
2092    
2093      case OP_CIRCM:      case OP_CIRCM:
2094      if (md->notbol && eptr == md->start_subject) MRRETURN(MATCH_NOMATCH);      if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH);
2095      if (eptr != md->start_subject &&      if (eptr != md->start_subject &&
2096          (eptr == md->end_subject || !WAS_NEWLINE(eptr)))          (eptr == md->end_subject || !WAS_NEWLINE(eptr)))
2097        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2098      ecode++;      ecode++;
2099      break;      break;
2100    
2101      /* Start of match assertion */      /* Start of match assertion */
2102    
2103      case OP_SOM:      case OP_SOM:
2104      if (eptr != md->start_subject + md->start_offset) MRRETURN(MATCH_NOMATCH);      if (eptr != md->start_subject + md->start_offset) RRETURN(MATCH_NOMATCH);
2105      ecode++;      ecode++;
2106      break;      break;
2107    
# Line 2017  for (;;) Line 2117  for (;;)
2117    
2118      case OP_DOLLM:      case OP_DOLLM:
2119      if (eptr < md->end_subject)      if (eptr < md->end_subject)
2120        { if (!IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH); }        {
2121          if (!IS_NEWLINE(eptr))
2122            {
2123            if (md->partial != 0 &&
2124                eptr + 1 >= md->end_subject &&
2125                NLBLOCK->nltype == NLTYPE_FIXED &&
2126                NLBLOCK->nllen == 2 &&
2127                RAWUCHARTEST(eptr) == NLBLOCK->nl[0])
2128              {
2129              md->hitend = TRUE;
2130              if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
2131              }
2132            RRETURN(MATCH_NOMATCH);
2133            }
2134          }
2135      else      else
2136        {        {
2137        if (md->noteol) MRRETURN(MATCH_NOMATCH);        if (md->noteol) RRETURN(MATCH_NOMATCH);
2138        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2139        }        }
2140      ecode++;      ecode++;
# Line 2030  for (;;) Line 2144  for (;;)
2144      subject unless noteol is set. */      subject unless noteol is set. */
2145    
2146      case OP_DOLL:      case OP_DOLL:
2147      if (md->noteol) MRRETURN(MATCH_NOMATCH);      if (md->noteol) RRETURN(MATCH_NOMATCH);
2148      if (!md->endonly) goto ASSERT_NL_OR_EOS;      if (!md->endonly) goto ASSERT_NL_OR_EOS;
2149    
2150      /* ... else fall through for endonly */      /* ... else fall through for endonly */
# Line 2038  for (;;) Line 2152  for (;;)
2152      /* End of subject assertion (\z) */      /* End of subject assertion (\z) */
2153    
2154      case OP_EOD:      case OP_EOD:
2155      if (eptr < md->end_subject) MRRETURN(MATCH_NOMATCH);      if (eptr < md->end_subject) RRETURN(MATCH_NOMATCH);
2156      SCHECK_PARTIAL();      SCHECK_PARTIAL();
2157      ecode++;      ecode++;
2158      break;      break;
# Line 2049  for (;;) Line 2163  for (;;)
2163      ASSERT_NL_OR_EOS:      ASSERT_NL_OR_EOS:
2164      if (eptr < md->end_subject &&      if (eptr < md->end_subject &&
2165          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))          (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))
2166        MRRETURN(MATCH_NOMATCH);        {
2167          if (md->partial != 0 &&
2168              eptr + 1 >= md->end_subject &&
2169              NLBLOCK->nltype == NLTYPE_FIXED &&
2170              NLBLOCK->nllen == 2 &&
2171              RAWUCHARTEST(eptr) == NLBLOCK->nl[0])
2172            {
2173            md->hitend = TRUE;
2174            if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
2175            }
2176          RRETURN(MATCH_NOMATCH);
2177          }
2178    
2179      /* Either at end of string or \n before end. */      /* Either at end of string or \n before end. */
2180    
# Line 2068  for (;;) Line 2193  for (;;)
2193        be "non-word" characters. Remember the earliest consulted character for        be "non-word" characters. Remember the earliest consulted character for
2194        partial matching. */        partial matching. */
2195    
2196  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2197        if (utf8)        if (utf)
2198          {          {
2199          /* Get status of previous character */          /* Get status of previous character */
2200    
2201          if (eptr == md->start_subject) prev_is_word = FALSE; else          if (eptr == md->start_subject) prev_is_word = FALSE; else
2202            {            {
2203            USPTR lastptr = eptr - 1;            PCRE_PUCHAR lastptr = eptr - 1;
2204            while((*lastptr & 0xc0) == 0x80) lastptr--;            BACKCHAR(lastptr);
2205            if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;            if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;
2206            GETCHAR(c, lastptr);            GETCHAR(c, lastptr);
2207  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 2141  for (;;) Line 2266  for (;;)
2266              }              }
2267            else            else
2268  #endif  #endif
2269            prev_is_word = ((md->ctypes[eptr[-1]] & ctype_word) != 0);            prev_is_word = MAX_255(eptr[-1])
2270                && ((md->ctypes[eptr[-1]] & ctype_word) != 0);
2271            }            }
2272    
2273          /* Get status of next character */          /* Get status of next character */
# Line 2164  for (;;) Line 2290  for (;;)
2290            }            }
2291          else          else
2292  #endif  #endif
2293          cur_is_word = ((md->ctypes[*eptr] & ctype_word) != 0);          cur_is_word = MAX_255(*eptr)
2294              && ((md->ctypes[*eptr] & ctype_word) != 0);
2295          }          }
2296    
2297        /* Now see if the situation is what we want */        /* Now see if the situation is what we want */
2298    
2299        if ((*ecode++ == OP_WORD_BOUNDARY)?        if ((*ecode++ == OP_WORD_BOUNDARY)?
2300             cur_is_word == prev_is_word : cur_is_word != prev_is_word)             cur_is_word == prev_is_word : cur_is_word != prev_is_word)
2301          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2302        }        }
2303      break;      break;
2304    
2305      /* Match a single character type; inline for speed */      /* Match any single character type except newline; have to take care with
2306        CRLF newlines and partial matching. */
2307    
2308      case OP_ANY:      case OP_ANY:
2309      if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);      if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
2310        if (md->partial != 0 &&
2311            eptr + 1 >= md->end_subject &&
2312            NLBLOCK->nltype == NLTYPE_FIXED &&
2313            NLBLOCK->nllen == 2 &&
2314            RAWUCHARTEST(eptr) == NLBLOCK->nl[0])
2315          {
2316          md->hitend = TRUE;
2317          if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
2318          }
2319    
2320      /* Fall through */      /* Fall through */
2321    
2322        /* Match any single character whatsoever. */
2323    
2324      case OP_ALLANY:      case OP_ALLANY:
2325      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */
2326        {                            /* not be updated before SCHECK_PARTIAL. */        {                            /* not be updated before SCHECK_PARTIAL. */
2327        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2328        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2329        }        }
2330      eptr++;      eptr++;
2331      if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;  #ifdef SUPPORT_UTF
2332        if (utf) ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
2333    #endif
2334      ecode++;      ecode++;
2335      break;      break;
2336    
# Line 2199  for (;;) Line 2341  for (;;)
2341      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */      if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */
2342        {                            /* not be updated before SCHECK_PARTIAL. */        {                            /* not be updated before SCHECK_PARTIAL. */
2343        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2344        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2345        }        }
2346      eptr++;      eptr++;
2347      ecode++;      ecode++;
# Line 2209  for (;;) Line 2351  for (;;)
2351      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2352        {        {
2353        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2354        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2355        }        }
2356      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2357      if (      if (
2358  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2359         c < 256 &&         c < 256 &&
2360  #endif  #endif
2361         (md->ctypes[c] & ctype_digit) != 0         (md->ctypes[c] & ctype_digit) != 0
2362         )         )
2363        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2364      ecode++;      ecode++;
2365      break;      break;
2366    
# Line 2226  for (;;) Line 2368  for (;;)
2368      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2369        {        {
2370        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2371        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2372        }        }
2373      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2374      if (      if (
2375  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2376         c >= 256 ||         c > 255 ||
2377  #endif  #endif
2378         (md->ctypes[c] & ctype_digit) == 0         (md->ctypes[c] & ctype_digit) == 0
2379         )         )
2380        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2381      ecode++;      ecode++;
2382      break;      break;
2383    
# Line 2243  for (;;) Line 2385  for (;;)
2385      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2386        {        {
2387        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2388        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2389        }        }
2390      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2391      if (      if (
2392  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2393         c < 256 &&         c < 256 &&
2394  #endif  #endif
2395         (md->ctypes[c] & ctype_space) != 0         (md->ctypes[c] & ctype_space) != 0
2396         )         )
2397        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2398      ecode++;      ecode++;
2399      break;      break;
2400    
# Line 2260  for (;;) Line 2402  for (;;)
2402      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2403        {        {
2404        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2405        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2406        }        }
2407      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2408      if (      if (
2409  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2410         c >= 256 ||         c > 255 ||
2411  #endif  #endif
2412         (md->ctypes[c] & ctype_space) == 0         (md->ctypes[c] & ctype_space) == 0
2413         )         )
2414        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2415      ecode++;      ecode++;
2416      break;      break;
2417    
# Line 2277  for (;;) Line 2419  for (;;)
2419      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2420        {        {
2421        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2422        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2423        }        }
2424      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2425      if (      if (
2426  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2427         c < 256 &&         c < 256 &&
2428  #endif  #endif
2429         (md->ctypes[c] & ctype_word) != 0         (md->ctypes[c] & ctype_word) != 0
2430         )         )
2431        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2432      ecode++;      ecode++;
2433      break;      break;
2434    
# Line 2294  for (;;) Line 2436  for (;;)
2436      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2437        {        {
2438        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2439        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2440        }        }
2441      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2442      if (      if (
2443  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
2444         c >= 256 ||         c > 255 ||
2445  #endif  #endif
2446         (md->ctypes[c] & ctype_word) == 0         (md->ctypes[c] & ctype_word) == 0
2447         )         )
2448        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2449      ecode++;      ecode++;
2450      break;      break;
2451    
# Line 2311  for (;;) Line 2453  for (;;)
2453      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2454        {        {
2455        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2456        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2457        }        }
2458      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2459      switch(c)      switch(c)
2460        {        {
2461        default: MRRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
2462    
2463        case 0x000d:        case CHAR_CR:
2464        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr >= md->end_subject)
2465            {
2466            SCHECK_PARTIAL();
2467            }
2468          else if (RAWUCHARTEST(eptr) == CHAR_LF) eptr++;
2469        break;        break;
2470    
2471        case 0x000a:        case CHAR_LF:
2472        break;        break;
2473    
2474        case 0x000b:        case CHAR_VT:
2475        case 0x000c:        case CHAR_FF:
2476        case 0x0085:        case CHAR_NEL:
2477    #ifndef EBCDIC
2478        case 0x2028:        case 0x2028:
2479        case 0x2029:        case 0x2029:
2480        if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);  #endif  /* Not EBCDIC */
2481          if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
2482        break;        break;
2483        }        }
2484      ecode++;      ecode++;
# Line 2340  for (;;) Line 2488  for (;;)
2488      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2489        {        {
2490        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2491        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2492        }        }
2493      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2494      switch(c)      switch(c)
2495        {        {
2496          HSPACE_CASES: RRETURN(MATCH_NOMATCH);  /* Byte and multibyte cases */
2497        default: break;        default: break;
       case 0x09:      /* HT */  
       case 0x20:      /* SPACE */  
       case 0xa0:      /* NBSP */  
       case 0x1680:    /* OGHAM SPACE MARK */  
       case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */  
       case 0x2000:    /* EN QUAD */  
       case 0x2001:    /* EM QUAD */  
       case 0x2002:    /* EN SPACE */  
       case 0x2003:    /* EM SPACE */  
       case 0x2004:    /* THREE-PER-EM SPACE */  
       case 0x2005:    /* FOUR-PER-EM SPACE */  
       case 0x2006:    /* SIX-PER-EM SPACE */  
       case 0x2007:    /* FIGURE SPACE */  
       case 0x2008:    /* PUNCTUATION SPACE */  
       case 0x2009:    /* THIN SPACE */  
       case 0x200A:    /* HAIR SPACE */  
       case 0x202f:    /* NARROW NO-BREAK SPACE */  
       case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */  
       case 0x3000:    /* IDEOGRAPHIC SPACE */  
       MRRETURN(MATCH_NOMATCH);  
2498        }        }
2499      ecode++;      ecode++;
2500      break;      break;
# Line 2374  for (;;) Line 2503  for (;;)
2503      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2504        {        {
2505        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2506        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2507        }        }
2508      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2509      switch(c)      switch(c)
2510        {        {
2511        default: MRRETURN(MATCH_NOMATCH);        HSPACE_CASES: break;  /* Byte and multibyte cases */
2512        case 0x09:      /* HT */        default: RRETURN(MATCH_NOMATCH);
       case 0x20:      /* SPACE */  
       case 0xa0:      /* NBSP */  
       case 0x1680:    /* OGHAM SPACE MARK */  
       case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */  
       case 0x2000:    /* EN QUAD */  
       case 0x2001:    /* EM QUAD */  
       case 0x2002:    /* EN SPACE */  
       case 0x2003:    /* EM SPACE */  
       case 0x2004:    /* THREE-PER-EM SPACE */  
       case 0x2005:    /* FOUR-PER-EM SPACE */  
       case 0x2006:    /* SIX-PER-EM SPACE */  
       case 0x2007:    /* FIGURE SPACE */  
       case 0x2008:    /* PUNCTUATION SPACE */  
       case 0x2009:    /* THIN SPACE */  
       case 0x200A:    /* HAIR SPACE */  
       case 0x202f:    /* NARROW NO-BREAK SPACE */  
       case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */  
       case 0x3000:    /* IDEOGRAPHIC SPACE */  
       break;  
2513        }        }
2514      ecode++;      ecode++;
2515      break;      break;
# Line 2408  for (;;) Line 2518  for (;;)
2518      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2519        {        {
2520        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2521        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2522        }        }
2523      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2524      switch(c)      switch(c)
2525        {        {
2526          VSPACE_CASES: RRETURN(MATCH_NOMATCH);
2527        default: break;        default: break;
       case 0x0a:      /* LF */  
       case 0x0b:      /* VT */  
       case 0x0c:      /* FF */  
       case 0x0d:      /* CR */  
       case 0x85:      /* NEL */  
       case 0x2028:    /* LINE SEPARATOR */  
       case 0x2029:    /* PARAGRAPH SEPARATOR */  
       MRRETURN(MATCH_NOMATCH);  
2528        }        }
2529      ecode++;      ecode++;
2530      break;      break;
# Line 2430  for (;;) Line 2533  for (;;)
2533      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2534        {        {
2535        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2536        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2537        }        }
2538      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2539      switch(c)      switch(c)
2540        {        {
2541        default: MRRETURN(MATCH_NOMATCH);        VSPACE_CASES: break;
2542        case 0x0a:      /* LF */        default: RRETURN(MATCH_NOMATCH);
       case 0x0b:      /* VT */  
       case 0x0c:      /* FF */  
       case 0x0d:      /* CR */  
       case 0x85:      /* NEL */  
       case 0x2028:    /* LINE SEPARATOR */  
       case 0x2029:    /* PARAGRAPH SEPARATOR */  
       break;  
2543        }        }
2544      ecode++;      ecode++;
2545      break;      break;
# Line 2457  for (;;) Line 2553  for (;;)
2553      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2554        {        {
2555        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2556        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2557        }        }
2558      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2559        {        {
2560          const pcre_uint32 *cp;
2561        const ucd_record *prop = GET_UCD(c);        const ucd_record *prop = GET_UCD(c);
2562    
2563        switch(ecode[1])        switch(ecode[1])
2564          {          {
2565          case PT_ANY:          case PT_ANY:
2566          if (op == OP_NOTPROP) MRRETURN(MATCH_NOMATCH);          if (op == OP_NOTPROP) RRETURN(MATCH_NOMATCH);
2567          break;          break;
2568    
2569          case PT_LAMP:          case PT_LAMP:
2570          if ((prop->chartype == ucp_Lu ||          if ((prop->chartype == ucp_Lu ||
2571               prop->chartype == ucp_Ll ||               prop->chartype == ucp_Ll ||
2572               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))               prop->chartype == ucp_Lt) == (op == OP_NOTPROP))
2573            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2574          break;          break;
2575    
2576          case PT_GC:          case PT_GC:
2577          if ((ecode[2] != _pcre_ucp_gentype[prop->chartype]) == (op == OP_PROP))          if ((ecode[2] != PRIV(ucp_gentype)[prop->chartype]) == (op == OP_PROP))
2578            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2579          break;          break;
2580    
2581          case PT_PC:          case PT_PC:
2582          if ((ecode[2] != prop->chartype) == (op == OP_PROP))          if ((ecode[2] != prop->chartype) == (op == OP_PROP))
2583            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2584          break;          break;
2585    
2586          case PT_SC:          case PT_SC:
2587          if ((ecode[2] != prop->script) == (op == OP_PROP))          if ((ecode[2] != prop->script) == (op == OP_PROP))
2588            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2589          break;          break;
2590    
2591          /* These are specials */          /* These are specials */
2592    
2593          case PT_ALNUM:          case PT_ALNUM:
2594          if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
2595               _pcre_ucp_gentype[prop->chartype] == ucp_N) == (op == OP_NOTPROP))               PRIV(ucp_gentype)[prop->chartype] == ucp_N) == (op == OP_NOTPROP))
2596            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2597          break;          break;
2598    
2599          case PT_SPACE:    /* Perl space */          case PT_SPACE:    /* Perl space */
2600          if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
2601               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
2602                 == (op == OP_NOTPROP))                 == (op == OP_NOTPROP))
2603            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2604          break;          break;
2605    
2606          case PT_PXSPACE:  /* POSIX space */          case PT_PXSPACE:  /* POSIX space */
2607          if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
2608               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
2609               c == CHAR_FF || c == CHAR_CR)               c == CHAR_FF || c == CHAR_CR)
2610                 == (op == OP_NOTPROP))                 == (op == OP_NOTPROP))
2611            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2612          break;          break;
2613    
2614          case PT_WORD:          case PT_WORD:
2615          if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
2616               _pcre_ucp_gentype[prop->chartype] == ucp_N ||               PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
2617               c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))               c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))
2618            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2619            break;
2620    
2621            case PT_CLIST:
2622            cp = PRIV(ucd_caseless_sets) + ecode[2];
2623            for (;;)
2624              {
2625              if (c < *cp)
2626                { if (op == OP_PROP) { RRETURN(MATCH_NOMATCH); } else break; }
2627              if (c == *cp++)
2628                { if (op == OP_PROP) break; else { RRETURN(MATCH_NOMATCH); } }
2629              }
2630            break;
2631    
2632            case PT_UCNC:
2633            if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
2634                 c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
2635                 c >= 0xe000) == (op == OP_NOTPROP))
2636              RRETURN(MATCH_NOMATCH);
2637          break;          break;
2638    
2639          /* This should never occur */          /* This should never occur */
# Line 2538  for (;;) Line 2653  for (;;)
2653      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
2654        {        {
2655        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2656        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2657        }        }
2658      GETCHARINCTEST(c, eptr);      else
     if (UCD_CATEGORY(c) == ucp_M) MRRETURN(MATCH_NOMATCH);  
     while (eptr < md->end_subject)  
2659        {        {
2660        int len = 1;        int lgb, rgb;
2661        if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }        GETCHARINCTEST(c, eptr);
2662        if (UCD_CATEGORY(c) != ucp_M) break;        lgb = UCD_GRAPHBREAK(c);
2663        eptr += len;        while (eptr < md->end_subject)
2664            {
2665            int len = 1;
2666            if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
2667            rgb = UCD_GRAPHBREAK(c);
2668            if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
2669            lgb = rgb;
2670            eptr += len;
2671            }
2672        }        }
2673        CHECK_PARTIAL();
2674      ecode++;      ecode++;
2675      break;      break;
2676  #endif  #endif  /* SUPPORT_UCP */
2677    
2678    
2679      /* Match a back reference, possibly repeatedly. Look past the end of the      /* Match a back reference, possibly repeatedly. Look past the end of the
# Line 2566  for (;;) Line 2688  for (;;)
2688      case OP_REFI:      case OP_REFI:
2689      caseless = op == OP_REFI;      caseless = op == OP_REFI;
2690      offset = GET2(ecode, 1) << 1;               /* Doubled ref number */      offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
2691      ecode += 3;      ecode += 1 + IMM2_SIZE;
2692    
2693      /* If the reference is unset, there are two possibilities:      /* If the reference is unset, there are two possibilities:
2694    
# Line 2606  for (;;) Line 2728  for (;;)
2728        case OP_CRMINRANGE:        case OP_CRMINRANGE:
2729        minimize = (*ecode == OP_CRMINRANGE);        minimize = (*ecode == OP_CRMINRANGE);
2730        min = GET2(ecode, 1);        min = GET2(ecode, 1);
2731        max = GET2(ecode, 3);        max = GET2(ecode, 1 + IMM2_SIZE);
2732        if (max == 0) max = INT_MAX;        if (max == 0) max = INT_MAX;
2733        ecode += 5;        ecode += 1 + 2 * IMM2_SIZE;
2734        break;        break;
2735    
2736        default:               /* No repeat follows */        default:               /* No repeat follows */
2737        if ((length = match_ref(offset, eptr, length, md, caseless)) < 0)        if ((length = match_ref(offset, eptr, length, md, caseless)) < 0)
2738          {          {
2739            if (length == -2) eptr = md->end_subject;   /* Partial match */
2740          CHECK_PARTIAL();          CHECK_PARTIAL();
2741          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2742          }          }
2743        eptr += length;        eptr += length;
2744        continue;              /* With the main loop */        continue;              /* With the main loop */
2745        }        }
2746    
2747      /* Handle repeated back references. If the length of the reference is      /* Handle repeated back references. If the length of the reference is
2748      zero, just continue with the main loop. */      zero, just continue with the main loop. If the length is negative, it
2749        means the reference is unset in non-Java-compatible mode. If the minimum is
2750        zero, we can continue at the same level without recursion. For any other
2751        minimum, carrying on will result in NOMATCH. */
2752    
2753      if (length == 0) continue;      if (length == 0) continue;
2754        if (length < 0 && min == 0) continue;
2755    
2756      /* First, ensure the minimum number of matches are present. We get back      /* First, ensure the minimum number of matches are present. We get back
2757      the length of the reference string explicitly rather than passing the      the length of the reference string explicitly rather than passing the
# Line 2635  for (;;) Line 2762  for (;;)
2762        int slength;        int slength;
2763        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2764          {          {
2765            if (slength == -2) eptr = md->end_subject;   /* Partial match */
2766          CHECK_PARTIAL();          CHECK_PARTIAL();
2767          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2768          }          }
2769        eptr += slength;        eptr += slength;
2770        }        }
# Line 2655  for (;;) Line 2783  for (;;)
2783          int slength;          int slength;
2784          RMATCH(eptr, ecode, offset_top, md, eptrb, RM14);          RMATCH(eptr, ecode, offset_top, md, eptrb, RM14);
2785          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2786          if (fi >= max) MRRETURN(MATCH_NOMATCH);          if (fi >= max) RRETURN(MATCH_NOMATCH);
2787          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2788            {            {
2789              if (slength == -2) eptr = md->end_subject;   /* Partial match */
2790            CHECK_PARTIAL();            CHECK_PARTIAL();
2791            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2792            }            }
2793          eptr += slength;          eptr += slength;
2794          }          }
# Line 2676  for (;;) Line 2805  for (;;)
2805          int slength;          int slength;
2806          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)          if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
2807            {            {
2808            CHECK_PARTIAL();            /* Can't use CHECK_PARTIAL because we don't want to update eptr in
2809              the soft partial matching case. */
2810    
2811              if (slength == -2 && md->partial != 0 &&
2812                  md->end_subject > md->start_used_ptr)
2813                {
2814                md->hitend = TRUE;
2815                if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
2816                }
2817            break;            break;
2818            }            }
2819          eptr += slength;          eptr += slength;
2820          }          }
2821    
2822        while (eptr >= pp)        while (eptr >= pp)
2823          {          {
2824          RMATCH(eptr, ecode, offset_top, md, eptrb, RM15);          RMATCH(eptr, ecode, offset_top, md, eptrb, RM15);
2825          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2826          eptr -= length;          eptr -= length;
2827          }          }
2828        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2829        }        }
2830      /* Control never gets here */      /* Control never gets here */
2831    
# Line 2705  for (;;) Line 2843  for (;;)
2843      case OP_NCLASS:      case OP_NCLASS:
2844      case OP_CLASS:      case OP_CLASS:
2845        {        {
2846          /* The data variable is saved across frames, so the byte map needs to
2847          be stored there. */
2848    #define BYTE_MAP ((pcre_uint8 *)data)
2849        data = ecode + 1;                /* Save for matching */        data = ecode + 1;                /* Save for matching */
2850        ecode += 33;                     /* Advance past the item */        ecode += 1 + (32 / sizeof(pcre_uchar)); /* Advance past the item */
2851    
2852        switch (*ecode)        switch (*ecode)
2853          {          {
# Line 2727  for (;;) Line 2868  for (;;)
2868          case OP_CRMINRANGE:          case OP_CRMINRANGE:
2869          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
2870          min = GET2(ecode, 1);          min = GET2(ecode, 1);
2871          max = GET2(ecode, 3);          max = GET2(ecode, 1 + IMM2_SIZE);
2872          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
2873          ecode += 5;          ecode += 1 + 2 * IMM2_SIZE;
2874          break;          break;
2875    
2876          default:               /* No repeat follows */          default:               /* No repeat follows */
# Line 2739  for (;;) Line 2880  for (;;)
2880    
2881        /* First, ensure the minimum number of matches are present. */        /* First, ensure the minimum number of matches are present. */
2882    
2883  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2884        /* UTF-8 mode */        if (utf)
       if (utf8)  
2885          {          {
2886          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2887            {            {
2888            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
2889              {              {
2890              SCHECK_PARTIAL();              SCHECK_PARTIAL();
2891              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2892              }              }
2893            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
2894            if (c > 255)            if (c > 255)
2895              {              {
2896              if (op == OP_CLASS) MRRETURN(MATCH_NOMATCH);              if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2897              }              }
2898            else            else
2899              {              if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
             if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  
             }  
2900            }            }
2901          }          }
2902        else        else
2903  #endif  #endif
2904        /* Not UTF-8 mode */        /* Not UTF mode */
2905          {          {
2906          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
2907            {            {
2908            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
2909              {              {
2910              SCHECK_PARTIAL();              SCHECK_PARTIAL();
2911              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
2912              }              }
2913            c = *eptr++;            c = *eptr++;
2914            if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  #ifndef COMPILE_PCRE8
2915              if (c > 255)
2916                {
2917                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2918                }
2919              else
2920    #endif
2921                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2922            }            }
2923          }          }
2924    
# Line 2787  for (;;) Line 2932  for (;;)
2932    
2933        if (minimize)        if (minimize)
2934          {          {
2935  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2936          /* UTF-8 mode */          if (utf)
         if (utf8)  
2937            {            {
2938            for (fi = min;; fi++)            for (fi = min;; fi++)
2939              {              {
2940              RMATCH(eptr, ecode, offset_top, md, eptrb, RM16);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM16);
2941              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2942              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2943              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
2944                {                {
2945                SCHECK_PARTIAL();                SCHECK_PARTIAL();
2946                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2947                }                }
2948              GETCHARINC(c, eptr);              GETCHARINC(c, eptr);
2949              if (c > 255)              if (c > 255)
2950                {                {
2951                if (op == OP_CLASS) MRRETURN(MATCH_NOMATCH);                if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2952                }                }
2953              else              else
2954                {                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
               if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  
               }  
2955              }              }
2956            }            }
2957          else          else
2958  #endif  #endif
2959          /* Not UTF-8 mode */          /* Not UTF mode */
2960            {            {
2961            for (fi = min;; fi++)            for (fi = min;; fi++)
2962              {              {
2963              RMATCH(eptr, ecode, offset_top, md, eptrb, RM17);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM17);
2964              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2965              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
2966              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
2967                {                {
2968                SCHECK_PARTIAL();                SCHECK_PARTIAL();
2969                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
2970                }                }
2971              c = *eptr++;              c = *eptr++;
2972              if ((data[c/8] & (1 << (c&7))) == 0) MRRETURN(MATCH_NOMATCH);  #ifndef COMPILE_PCRE8
2973                if (c > 255)
2974                  {
2975                  if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
2976                  }
2977                else
2978    #endif
2979                  if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
2980              }              }
2981            }            }
2982          /* Control never gets here */          /* Control never gets here */
# Line 2839  for (;;) Line 2988  for (;;)
2988          {          {
2989          pp = eptr;          pp = eptr;
2990    
2991  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2992          /* UTF-8 mode */          if (utf)
         if (utf8)  
2993            {            {
2994            for (i = min; i < max; i++)            for (i = min; i < max; i++)
2995              {              {
# Line 2857  for (;;) Line 3005  for (;;)
3005                if (op == OP_CLASS) break;                if (op == OP_CLASS) break;
3006                }                }
3007              else              else
3008                {                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
               if ((data[c/8] & (1 << (c&7))) == 0) break;  
               }  
3009              eptr += len;              eptr += len;
3010              }              }
3011            for (;;)            for (;;)
# Line 2872  for (;;) Line 3018  for (;;)
3018            }            }
3019          else          else
3020  #endif  #endif
3021            /* Not UTF-8 mode */            /* Not UTF mode */
3022            {            {
3023            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3024              {              {
# Line 2882  for (;;) Line 3028  for (;;)
3028                break;                break;
3029                }                }
3030              c = *eptr;              c = *eptr;
3031              if ((data[c/8] & (1 << (c&7))) == 0) break;  #ifndef COMPILE_PCRE8
3032                if (c > 255)
3033                  {
3034                  if (op == OP_CLASS) break;
3035                  }
3036                else
3037    #endif
3038                  if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
3039              eptr++;              eptr++;
3040              }              }
3041            while (eptr >= pp)            while (eptr >= pp)
# Line 2893  for (;;) Line 3046  for (;;)
3046              }              }
3047            }            }
3048    
3049          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3050          }          }
3051    #undef BYTE_MAP
3052        }        }
3053      /* Control never gets here */      /* Control never gets here */
3054    
# Line 2903  for (;;) Line 3057  for (;;)
3057      when UTF-8 mode mode is supported. Nevertheless, we may not be in UTF-8      when UTF-8 mode mode is supported. Nevertheless, we may not be in UTF-8
3058      mode, because Unicode properties are supported in non-UTF-8 mode. */      mode, because Unicode properties are supported in non-UTF-8 mode. */
3059    
3060  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
3061      case OP_XCLASS:      case OP_XCLASS:
3062        {        {
3063        data = ecode + 1 + LINK_SIZE;                /* Save for matching */        data = ecode + 1 + LINK_SIZE;                /* Save for matching */
# Line 2928  for (;;) Line 3082  for (;;)
3082          case OP_CRMINRANGE:          case OP_CRMINRANGE:
3083          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
3084          min = GET2(ecode, 1);          min = GET2(ecode, 1);
3085          max = GET2(ecode, 3);          max = GET2(ecode, 1 + IMM2_SIZE);
3086          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
3087          ecode += 5;          ecode += 1 + 2 * IMM2_SIZE;
3088          break;          break;
3089    
3090          default:               /* No repeat follows */          default:               /* No repeat follows */
# Line 2945  for (;;) Line 3099  for (;;)
3099          if (eptr >= md->end_subject)          if (eptr >= md->end_subject)
3100            {            {
3101            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3102            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3103            }            }
3104          GETCHARINCTEST(c, eptr);          GETCHARINCTEST(c, eptr);
3105          if (!_pcre_xclass(c, data)) MRRETURN(MATCH_NOMATCH);          if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH);
3106          }          }
3107    
3108        /* If max == min we can continue with the main loop without the        /* If max == min we can continue with the main loop without the
# Line 2965  for (;;) Line 3119  for (;;)
3119            {            {
3120            RMATCH(eptr, ecode, offset_top, md, eptrb, RM20);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM20);
3121            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3122            if (fi >= max) MRRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3123            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3124              {              {
3125              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3126              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3127              }              }
3128            GETCHARINCTEST(c, eptr);            GETCHARINCTEST(c, eptr);
3129            if (!_pcre_xclass(c, data)) MRRETURN(MATCH_NOMATCH);            if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH);
3130            }            }
3131          /* Control never gets here */          /* Control never gets here */
3132          }          }
# Line 2990  for (;;) Line 3144  for (;;)
3144              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3145              break;              break;
3146              }              }
3147    #ifdef SUPPORT_UTF
3148            GETCHARLENTEST(c, eptr, len);            GETCHARLENTEST(c, eptr, len);
3149            if (!_pcre_xclass(c, data)) break;  #else
3150              c = *eptr;
3151    #endif
3152              if (!PRIV(xclass)(c, data, utf)) break;
3153            eptr += len;            eptr += len;
3154            }            }
3155          for(;;)          for(;;)
# Line 2999  for (;;) Line 3157  for (;;)
3157            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);
3158            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3159            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
3160            if (utf8) BACKCHAR(eptr);  #ifdef SUPPORT_UTF
3161              if (utf) BACKCHAR(eptr);
3162    #endif
3163            }            }
3164          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3165          }          }
3166    
3167        /* Control never gets here */        /* Control never gets here */
# Line 3011  for (;;) Line 3171  for (;;)
3171      /* Match a single character, casefully */      /* Match a single character, casefully */
3172    
3173      case OP_CHAR:      case OP_CHAR:
3174  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3175      if (utf8)      if (utf)
3176        {        {
3177        length = 1;        length = 1;
3178        ecode++;        ecode++;
# Line 3020  for (;;) Line 3180  for (;;)
3180        if (length > md->end_subject - eptr)        if (length > md->end_subject - eptr)
3181          {          {
3182          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
3183          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3184          }          }
3185        while (length-- > 0) if (*ecode++ != *eptr++) MRRETURN(MATCH_NOMATCH);        while (length-- > 0) if (*ecode++ != RAWUCHARINC(eptr)) RRETURN(MATCH_NOMATCH);
3186        }        }
3187      else      else
3188  #endif  #endif
3189        /* Not UTF mode */
     /* Non-UTF-8 mode */  
3190        {        {
3191        if (md->end_subject - eptr < 1)        if (md->end_subject - eptr < 1)
3192          {          {
3193          SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */          SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
3194          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3195          }          }
3196        if (ecode[1] != *eptr++) MRRETURN(MATCH_NOMATCH);        if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);
3197        ecode += 2;        ecode += 2;
3198        }        }
3199      break;      break;
3200    
3201      /* Match a single character, caselessly */      /* Match a single character, caselessly. If we are at the end of the
3202        subject, give up immediately. */
3203    
3204      case OP_CHARI:      case OP_CHARI:
3205  #ifdef SUPPORT_UTF8      if (eptr >= md->end_subject)
3206      if (utf8)        {
3207          SCHECK_PARTIAL();
3208          RRETURN(MATCH_NOMATCH);
3209          }
3210    
3211    #ifdef SUPPORT_UTF
3212        if (utf)
3213        {        {
3214        length = 1;        length = 1;
3215        ecode++;        ecode++;
3216        GETCHARLEN(fc, ecode, length);        GETCHARLEN(fc, ecode, length);
3217    
       if (length > md->end_subject - eptr)  
         {  
         CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */  
         MRRETURN(MATCH_NOMATCH);  
         }  
   
3218        /* 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
3219        can use the fast lookup table. */        we know that its other case must also be one byte long, so we can use the
3220          fast lookup table. We know that there is at least one byte left in the
3221          subject. */
3222    
3223        if (fc < 128)        if (fc < 128)
3224          {          {
3225          if (md->lcc[*ecode++] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);          pcre_uint32 cc = RAWUCHAR(eptr);
3226            if (md->lcc[fc] != TABLE_GET(cc, md->lcc, cc)) RRETURN(MATCH_NOMATCH);
3227            ecode++;
3228            eptr++;
3229          }          }
3230    
3231        /* Otherwise we must pick up the subject character */        /* Otherwise we must pick up the subject character. Note that we cannot
3232          use the value of "length" to check for sufficient bytes left, because the
3233          other case of the character may have more or fewer bytes.  */
3234    
3235        else        else
3236          {          {
3237          unsigned int dc;          pcre_uint32 dc;
3238          GETCHARINC(dc, eptr);          GETCHARINC(dc, eptr);
3239          ecode += length;          ecode += length;
3240    
# Line 3079  for (;;) Line 3246  for (;;)
3246  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3247            if (dc != UCD_OTHERCASE(fc))            if (dc != UCD_OTHERCASE(fc))
3248  #endif  #endif
3249              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3250            }            }
3251          }          }
3252        }        }
3253      else      else
3254  #endif   /* SUPPORT_UTF8 */  #endif   /* SUPPORT_UTF */
3255    
3256      /* Non-UTF-8 mode */      /* Not UTF mode */
3257        {        {
3258        if (md->end_subject - eptr < 1)        if (TABLE_GET(ecode[1], md->lcc, ecode[1])
3259          {            != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH);
3260          SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */        eptr++;
         MRRETURN(MATCH_NOMATCH);  
         }  
       if (md->lcc[ecode[1]] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);  
3261        ecode += 2;        ecode += 2;
3262        }        }
3263      break;      break;
# Line 3103  for (;;) Line 3267  for (;;)
3267      case OP_EXACT:      case OP_EXACT:
3268      case OP_EXACTI:      case OP_EXACTI:
3269      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3270      ecode += 3;      ecode += 1 + IMM2_SIZE;
3271      goto REPEATCHAR;      goto REPEATCHAR;
3272    
3273      case OP_POSUPTO:      case OP_POSUPTO:
# Line 3118  for (;;) Line 3282  for (;;)
3282      min = 0;      min = 0;
3283      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3284      minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI;      minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI;
3285      ecode += 3;      ecode += 1 + IMM2_SIZE;
3286      goto REPEATCHAR;      goto REPEATCHAR;
3287    
3288      case OP_POSSTAR:      case OP_POSSTAR:
# Line 3166  for (;;) Line 3330  for (;;)
3330      /* Common code for all repeated single-character matches. */      /* Common code for all repeated single-character matches. */
3331    
3332      REPEATCHAR:      REPEATCHAR:
3333  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3334      if (utf8)      if (utf)
3335        {        {
3336        length = 1;        length = 1;
3337        charptr = ecode;        charptr = ecode;
# Line 3180  for (;;) Line 3344  for (;;)
3344        if (length > 1)        if (length > 1)
3345          {          {
3346  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3347          unsigned int othercase;          pcre_uint32 othercase;
3348          if (op >= OP_STARI &&     /* Caseless */          if (op >= OP_STARI &&     /* Caseless */
3349              (othercase = UCD_OTHERCASE(fc)) != fc)              (othercase = UCD_OTHERCASE(fc)) != fc)
3350            oclength = _pcre_ord2utf8(othercase, occhars);            oclength = PRIV(ord2utf)(othercase, occhars);
3351          else oclength = 0;          else oclength = 0;
3352  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3353    
3354          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3355            {            {
3356            if (eptr <= md->end_subject - length &&            if (eptr <= md->end_subject - length &&
3357              memcmp(eptr, charptr, length) == 0) eptr += length;              memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3358  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3359            else if (oclength > 0 &&            else if (oclength > 0 &&
3360                     eptr <= md->end_subject - oclength &&                     eptr <= md->end_subject - oclength &&
3361                     memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                     memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3362  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3363            else            else
3364              {              {
3365              CHECK_PARTIAL();              CHECK_PARTIAL();
3366              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3367              }              }
3368            }            }
3369    
# Line 3211  for (;;) Line 3375  for (;;)
3375              {              {
3376              RMATCH(eptr, ecode, offset_top, md, eptrb, RM22);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM22);
3377              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3378              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3379              if (eptr <= md->end_subject - length &&              if (eptr <= md->end_subject - length &&
3380                memcmp(eptr, charptr, length) == 0) eptr += length;                memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3381  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3382              else if (oclength > 0 &&              else if (oclength > 0 &&
3383                       eptr <= md->end_subject - oclength &&                       eptr <= md->end_subject - oclength &&
3384                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                       memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3385  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3386              else              else
3387                {                {
3388                CHECK_PARTIAL();                CHECK_PARTIAL();
3389                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3390                }                }
3391              }              }
3392            /* Control never gets here */            /* Control never gets here */
# Line 3234  for (;;) Line 3398  for (;;)
3398            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3399              {              {
3400              if (eptr <= md->end_subject - length &&              if (eptr <= md->end_subject - length &&
3401                  memcmp(eptr, charptr, length) == 0) eptr += length;                  memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
3402  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3403              else if (oclength > 0 &&              else if (oclength > 0 &&
3404                       eptr <= md->end_subject - oclength &&                       eptr <= md->end_subject - oclength &&
3405                       memcmp(eptr, occhars, oclength) == 0) eptr += oclength;                       memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
3406  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3407              else              else
3408                {                {
# Line 3253  for (;;) Line 3417  for (;;)
3417              {              {
3418              RMATCH(eptr, ecode, offset_top, md, eptrb, RM23);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM23);
3419              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3420              if (eptr == pp) { MRRETURN(MATCH_NOMATCH); }              if (eptr == pp) { RRETURN(MATCH_NOMATCH); }
3421  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3422              eptr--;              eptr--;
3423              BACKCHAR(eptr);              BACKCHAR(eptr);
# Line 3270  for (;;) Line 3434  for (;;)
3434        value of fc will always be < 128. */        value of fc will always be < 128. */
3435        }        }
3436      else      else
3437  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
3438          /* When not in UTF-8 mode, load a single-byte character. */
3439          fc = *ecode++;
3440    
3441      /* When not in UTF-8 mode, load a single-byte character. */      /* The value of fc at this point is always one character, though we may
3442        or may not be in UTF mode. The code is duplicated for the caseless and
     fc = *ecode++;  
   
     /* The value of fc at this point is always less than 256, though we may or  
     may not be in UTF-8 mode. The code is duplicated for the caseless and  
3443      caseful cases, for speed, since matching characters is likely to be quite      caseful cases, for speed, since matching characters is likely to be quite
3444      common. First, ensure the minimum number of matches are present. If min =      common. First, ensure the minimum number of matches are present. If min =
3445      max, continue at the same level without recursing. Otherwise, if      max, continue at the same level without recursing. Otherwise, if
# Line 3286  for (;;) Line 3448  for (;;)
3448      maximizing, find the maximum number of characters and work backwards. */      maximizing, find the maximum number of characters and work backwards. */
3449    
3450      DPRINTF(("matching %c{%d,%d} against subject %.*s\n", fc, min, max,      DPRINTF(("matching %c{%d,%d} against subject %.*s\n", fc, min, max,
3451        max, eptr));        max, (char *)eptr));
3452    
3453      if (op >= OP_STARI)  /* Caseless */      if (op >= OP_STARI)  /* Caseless */
3454        {        {
3455        fc = md->lcc[fc];  #ifdef COMPILE_PCRE8
3456          /* fc must be < 128 if UTF is enabled. */
3457          foc = md->fcc[fc];
3458    #else
3459    #ifdef SUPPORT_UTF
3460    #ifdef SUPPORT_UCP
3461          if (utf && fc > 127)
3462            foc = UCD_OTHERCASE(fc);
3463    #else
3464          if (utf && fc > 127)
3465            foc = fc;
3466    #endif /* SUPPORT_UCP */
3467          else
3468    #endif /* SUPPORT_UTF */
3469            foc = TABLE_GET(fc, md->fcc, fc);
3470    #endif /* COMPILE_PCRE8 */
3471    
3472        for (i = 1; i <= min; i++)        for (i = 1; i <= min; i++)
3473          {          {
3474            pcre_uint32 cc;                 /* Faster than pcre_uchar */
3475          if (eptr >= md->end_subject)          if (eptr >= md->end_subject)
3476            {            {
3477            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3478            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3479            }            }
3480          if (fc != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);          cc = RAWUCHARTEST(eptr);
3481            if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
3482            eptr++;
3483          }          }
3484        if (min == max) continue;        if (min == max) continue;
3485        if (minimize)        if (minimize)
3486          {          {
3487          for (fi = min;; fi++)          for (fi = min;; fi++)
3488            {            {
3489              pcre_uint32 cc;               /* Faster than pcre_uchar */
3490            RMATCH(eptr, ecode, offset_top, md, eptrb, RM24);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM24);
3491            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3492            if (fi >= max) MRRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3493            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3494              {              {
3495              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3496              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3497              }              }
3498            if (fc != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);            cc = RAWUCHARTEST(eptr);
3499              if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
3500              eptr++;
3501            }            }
3502          /* Control never gets here */          /* Control never gets here */
3503          }          }
# Line 3322  for (;;) Line 3506  for (;;)
3506          pp = eptr;          pp = eptr;
3507          for (i = min; i < max; i++)          for (i = min; i < max; i++)
3508            {            {
3509              pcre_uint32 cc;               /* Faster than pcre_uchar */
3510            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3511              {              {
3512              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3513              break;              break;
3514              }              }
3515            if (fc != md->lcc[*eptr]) break;            cc = RAWUCHARTEST(eptr);
3516              if (fc != cc && foc != cc) break;
3517            eptr++;            eptr++;
3518            }            }
3519    
# Line 3339  for (;;) Line 3525  for (;;)
3525            eptr--;            eptr--;
3526            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3527            }            }
3528          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3529          }          }
3530        /* Control never gets here */        /* Control never gets here */
3531        }        }
# Line 3353  for (;;) Line 3539  for (;;)
3539          if (eptr >= md->end_subject)          if (eptr >= md->end_subject)
3540            {            {
3541            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3542            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3543            }            }
3544          if (fc != *eptr++) MRRETURN(MATCH_NOMATCH);          if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);
3545          }          }
3546    
3547        if (min == max) continue;        if (min == max) continue;
# Line 3366  for (;;) Line 3552  for (;;)
3552            {            {
3553            RMATCH(eptr, ecode, offset_top, md, eptrb, RM26);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM26);
3554            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3555            if (fi >= max) MRRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
3556            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3557              {              {
3558              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3559              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3560              }              }
3561            if (fc != *eptr++) MRRETURN(MATCH_NOMATCH);            if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);
3562            }            }
3563          /* Control never gets here */          /* Control never gets here */
3564          }          }
# Line 3386  for (;;) Line 3572  for (;;)
3572              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3573              break;              break;
3574              }              }
3575            if (fc != *eptr) break;            if (fc != RAWUCHARTEST(eptr)) break;
3576            eptr++;            eptr++;
3577            }            }
3578          if (possessive) continue;          if (possessive) continue;
# Line 3397  for (;;) Line 3583  for (;;)
3583            eptr--;            eptr--;
3584            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3585            }            }
3586          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3587          }          }
3588        }        }
3589      /* Control never gets here */      /* Control never gets here */
# Line 3410  for (;;) Line 3596  for (;;)
3596      if (eptr >= md->end_subject)      if (eptr >= md->end_subject)
3597        {        {
3598        SCHECK_PARTIAL();        SCHECK_PARTIAL();
3599        MRRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
3600        }        }
3601      ecode++;  #ifdef SUPPORT_UTF
3602      GETCHARINCTEST(c, eptr);      if (utf)
     if (op == OP_NOTI)         /* The caseless case */  
3603        {        {
3604  #ifdef SUPPORT_UTF8        register pcre_uint32 ch, och;
3605        if (c < 256)  
3606  #endif        ecode++;
3607        c = md->lcc[c];        GETCHARINC(ch, ecode);
3608        if (md->lcc[*ecode++] == c) MRRETURN(MATCH_NOMATCH);        GETCHARINC(c, eptr);
3609    
3610          if (op == OP_NOT)
3611            {
3612            if (ch == c) RRETURN(MATCH_NOMATCH);
3613            }
3614          else
3615            {
3616    #ifdef SUPPORT_UCP
3617            if (ch > 127)
3618              och = UCD_OTHERCASE(ch);
3619    #else
3620            if (ch > 127)
3621              och = ch;
3622    #endif /* SUPPORT_UCP */
3623            else
3624              och = TABLE_GET(ch, md->fcc, ch);
3625            if (ch == c || och == c) RRETURN(MATCH_NOMATCH);
3626            }
3627        }        }
3628      else    /* Caseful */      else
3629    #endif
3630        {        {
3631        if (*ecode++ == c) MRRETURN(MATCH_NOMATCH);        register pcre_uint32 ch = ecode[1];
3632          c = *eptr++;
3633          if (ch == c || (op == OP_NOTI && TABLE_GET(ch, md->fcc, ch) == c))
3634            RRETURN(MATCH_NOMATCH);
3635          ecode += 2;
3636        }        }
3637      break;      break;
3638    
# Line 3438  for (;;) Line 3646  for (;;)
3646      case OP_NOTEXACT:      case OP_NOTEXACT:
3647      case OP_NOTEXACTI:      case OP_NOTEXACTI:
3648      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
3649      ecode += 3;      ecode += 1 + IMM2_SIZE;
3650      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3651    
3652      case OP_NOTUPTO:      case OP_NOTUPTO:
# Line 3448  for (;;) Line 3656  for (;;)
3656      min = 0;      min = 0;
3657      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3658      minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI;      minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI;
3659      ecode += 3;      ecode += 1 + IMM2_SIZE;
3660      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3661    
3662      case OP_NOTPOSSTAR:      case OP_NOTPOSSTAR:
# Line 3480  for (;;) Line 3688  for (;;)
3688      possessive = TRUE;      possessive = TRUE;
3689      min = 0;      min = 0;
3690      max = GET2(ecode, 1);      max = GET2(ecode, 1);
3691      ecode += 3;      ecode += 1 + IMM2_SIZE;
3692      goto REPEATNOTCHAR;      goto REPEATNOTCHAR;
3693    
3694      case OP_NOTSTAR:      case OP_NOTSTAR:
# Line 3504  for (;;) Line 3712  for (;;)
3712      /* Common code for all repeated single-byte matches. */      /* Common code for all repeated single-byte matches. */
3713    
3714      REPEATNOTCHAR:      REPEATNOTCHAR:
3715      fc = *ecode++;      GETCHARINCTEST(fc, ecode);
3716    
3717      /* The code is duplicated for the caseless and caseful cases, for speed,      /* The code is duplicated for the caseless and caseful cases, for speed,
3718      since matching characters is likely to be quite common. First, ensure the      since matching characters is likely to be quite common. First, ensure the
# Line 3515  for (;;) Line 3723  for (;;)
3723      characters and work backwards. */      characters and work backwards. */
3724    
3725      DPRINTF(("negative matching %c{%d,%d} against subject %.*s\n", fc, min, max,      DPRINTF(("negative matching %c{%d,%d} against subject %.*s\n", fc, min, max,
3726        max, eptr));        max, (char *)eptr));
3727    
3728      if (op >= OP_NOTSTARI)     /* Caseless */      if (op >= OP_NOTSTARI)     /* Caseless */
3729        {        {
3730        fc = md->lcc[fc];  #ifdef SUPPORT_UTF
3731    #ifdef SUPPORT_UCP
3732          if (utf && fc > 127)
3733            foc = UCD_OTHERCASE(fc);
3734    #else
3735          if (utf && fc > 127)
3736            foc = fc;
3737    #endif /* SUPPORT_UCP */
3738          else
3739    #endif /* SUPPORT_UTF */
3740            foc = TABLE_GET(fc, md->fcc, fc);
3741    
3742  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3743        /* UTF-8 mode */        if (utf)
       if (utf8)  
3744          {          {
3745          register unsigned int d;          register pcre_uint32 d;
3746          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3747            {            {
3748            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3749              {              {
3750              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3751              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3752              }              }
3753            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3754            if (d < 256) d = md->lcc[d];            if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH);
           if (fc == d) MRRETURN(MATCH_NOMATCH);  
3755            }            }
3756          }          }
3757        else        else
3758  #endif  #endif
3759          /* Not UTF mode */
       /* Not UTF-8 mode */  
3760          {          {
3761          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3762            {            {
3763            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3764              {              {
3765              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3766              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3767              }              }
3768            if (fc == md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);            if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
3769              eptr++;
3770            }            }
3771          }          }
3772    
# Line 3558  for (;;) Line 3774  for (;;)
3774    
3775        if (minimize)        if (minimize)
3776          {          {
3777  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3778          /* UTF-8 mode */          if (utf)
         if (utf8)  
3779            {            {
3780            register unsigned int d;            register pcre_uint32 d;
3781            for (fi = min;; fi++)            for (fi = min;; fi++)
3782              {              {
3783              RMATCH(eptr, ecode, offset_top, md, eptrb, RM28);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM28);
3784              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3785              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3786              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3787                {                {
3788                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3789                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3790                }                }
3791              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3792              if (d < 256) d = md->lcc[d];              if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH);
             if (fc == d) MRRETURN(MATCH_NOMATCH);  
3793              }              }
3794            }            }
3795          else          else
3796  #endif  #endif
3797          /* Not UTF-8 mode */          /* Not UTF mode */
3798            {            {
3799            for (fi = min;; fi++)            for (fi = min;; fi++)
3800              {              {
3801              RMATCH(eptr, ecode, offset_top, md, eptrb, RM29);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM29);
3802              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3803              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3804              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3805                {                {
3806                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3807                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3808                }                }
3809              if (fc == md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);              if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
3810                eptr++;
3811              }              }
3812            }            }
3813          /* Control never gets here */          /* Control never gets here */
# Line 3604  for (;;) Line 3819  for (;;)
3819          {          {
3820          pp = eptr;          pp = eptr;
3821    
3822  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3823          /* UTF-8 mode */          if (utf)
         if (utf8)  
3824            {            {
3825            register unsigned int d;            register pcre_uint32 d;
3826            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3827              {              {
3828              int len = 1;              int len = 1;
# Line 3618  for (;;) Line 3832  for (;;)
3832                break;                break;
3833                }                }
3834              GETCHARLEN(d, eptr, len);              GETCHARLEN(d, eptr, len);
3835              if (d < 256) d = md->lcc[d];              if (fc == d || (unsigned int)foc == d) break;
             if (fc == d) break;  
3836              eptr += len;              eptr += len;
3837              }              }
3838          if (possessive) continue;            if (possessive) continue;
3839          for(;;)            for(;;)
3840              {              {
3841              RMATCH(eptr, ecode, offset_top, md, eptrb, RM30);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM30);
3842              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
# Line 3633  for (;;) Line 3846  for (;;)
3846            }            }
3847          else          else
3848  #endif  #endif
3849          /* Not UTF-8 mode */          /* Not UTF mode */
3850            {            {
3851            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3852              {              {
# Line 3642  for (;;) Line 3855  for (;;)
3855                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3856                break;                break;
3857                }                }
3858              if (fc == md->lcc[*eptr]) break;              if (fc == *eptr || foc == *eptr) break;
3859              eptr++;              eptr++;
3860              }              }
3861            if (possessive) continue;            if (possessive) continue;
# Line 3654  for (;;) Line 3867  for (;;)
3867              }              }
3868            }            }
3869    
3870          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3871          }          }
3872        /* Control never gets here */        /* Control never gets here */
3873        }        }
# Line 3663  for (;;) Line 3876  for (;;)
3876    
3877      else      else
3878        {        {
3879  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3880        /* UTF-8 mode */        if (utf)
       if (utf8)  
3881          {          {
3882          register unsigned int d;          register pcre_uint32 d;
3883          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3884            {            {
3885            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3886              {              {
3887              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3888              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3889              }              }
3890            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3891            if (fc == d) MRRETURN(MATCH_NOMATCH);            if (fc == d) RRETURN(MATCH_NOMATCH);
3892            }            }
3893          }          }
3894        else        else
3895  #endif  #endif
3896        /* Not UTF-8 mode */        /* Not UTF mode */
3897          {          {
3898          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3899            {            {
3900            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
3901              {              {
3902              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3903              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3904              }              }
3905            if (fc == *eptr++) MRRETURN(MATCH_NOMATCH);            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
3906            }            }
3907          }          }
3908    
# Line 3698  for (;;) Line 3910  for (;;)
3910    
3911        if (minimize)        if (minimize)
3912          {          {
3913  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3914          /* UTF-8 mode */          if (utf)
         if (utf8)  
3915            {            {
3916            register unsigned int d;            register pcre_uint32 d;
3917            for (fi = min;; fi++)            for (fi = min;; fi++)
3918              {              {
3919              RMATCH(eptr, ecode, offset_top, md, eptrb, RM32);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM32);
3920              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3921              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3922              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3923                {                {
3924                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3925                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3926                }                }
3927              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3928              if (fc == d) MRRETURN(MATCH_NOMATCH);              if (fc == d) RRETURN(MATCH_NOMATCH);
3929              }              }
3930            }            }
3931          else          else
3932  #endif  #endif
3933          /* Not UTF-8 mode */          /* Not UTF mode */
3934            {            {
3935            for (fi = min;; fi++)            for (fi = min;; fi++)
3936              {              {
3937              RMATCH(eptr, ecode, offset_top, md, eptrb, RM33);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM33);
3938              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3939              if (fi >= max) MRRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
3940              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
3941                {                {
3942                SCHECK_PARTIAL();                SCHECK_PARTIAL();
3943                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3944                }                }
3945              if (fc == *eptr++) MRRETURN(MATCH_NOMATCH);              if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
3946              }              }
3947            }            }
3948          /* Control never gets here */          /* Control never gets here */
# Line 3743  for (;;) Line 3954  for (;;)
3954          {          {
3955          pp = eptr;          pp = eptr;
3956    
3957  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3958          /* UTF-8 mode */          if (utf)
         if (utf8)  
3959            {            {
3960            register unsigned int d;            register pcre_uint32 d;
3961            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3962              {              {
3963              int len = 1;              int len = 1;
# Line 3771  for (;;) Line 3981  for (;;)
3981            }            }
3982          else          else
3983  #endif  #endif
3984          /* Not UTF-8 mode */          /* Not UTF mode */
3985            {            {
3986            for (i = min; i < max; i++)            for (i = min; i < max; i++)
3987              {              {
# Line 3792  for (;;) Line 4002  for (;;)
4002              }              }
4003            }            }
4004    
4005          MRRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
4006          }          }
4007        }        }
4008      /* Control never gets here */      /* Control never gets here */
# Line 3804  for (;;) Line 4014  for (;;)
4014      case OP_TYPEEXACT:      case OP_TYPEEXACT:
4015      min = max = GET2(ecode, 1);      min = max = GET2(ecode, 1);
4016      minimize = TRUE;      minimize = TRUE;
4017      ecode += 3;      ecode += 1 + IMM2_SIZE;
4018      goto REPEATTYPE;      goto REPEATTYPE;
4019    
4020      case OP_TYPEUPTO:      case OP_TYPEUPTO:
# Line 3812  for (;;) Line 4022  for (;;)
4022      min = 0;      min = 0;
4023      max = GET2(ecode, 1);      max = GET2(ecode, 1);
4024      minimize = *ecode == OP_TYPEMINUPTO;      minimize = *ecode == OP_TYPEMINUPTO;
4025      ecode += 3;      ecode += 1 + IMM2_SIZE;
4026      goto REPEATTYPE;      goto REPEATTYPE;
4027    
4028      case OP_TYPEPOSSTAR:      case OP_TYPEPOSSTAR:
# Line 3840  for (;;) Line 4050  for (;;)
4050      possessive = TRUE;      possessive = TRUE;
4051      min = 0;      min = 0;
4052      max = GET2(ecode, 1);      max = GET2(ecode, 1);
4053      ecode += 3;      ecode += 1 + IMM2_SIZE;
4054      goto REPEATTYPE;      goto REPEATTYPE;
4055    
4056      case OP_TYPESTAR:      case OP_TYPESTAR:
# Line 3886  for (;;) Line 4096  for (;;)
4096          switch(prop_type)          switch(prop_type)
4097            {            {
4098            case PT_ANY:            case PT_ANY:
4099            if (prop_fail_result) MRRETURN(MATCH_NOMATCH);            if (prop_fail_result) RRETURN(MATCH_NOMATCH);
4100            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
4101              {              {
4102              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4103                {                {
4104                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4105                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4106                }                }
4107              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4108              }              }
# Line 3905  for (;;) Line 4115  for (;;)
4115              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4116                {                {
4117                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4118                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4119                }                }
4120              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4121              chartype = UCD_CHARTYPE(c);              chartype = UCD_CHARTYPE(c);
4122              if ((chartype == ucp_Lu ||              if ((chartype == ucp_Lu ||
4123                   chartype == ucp_Ll ||                   chartype == ucp_Ll ||
4124                   chartype == ucp_Lt) == prop_fail_result)                   chartype == ucp_Lt) == prop_fail_result)
4125                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4126              }              }
4127            break;            break;
4128    
# Line 3922  for (;;) Line 4132  for (;;)
4132              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4133                {                {
4134                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4135                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4136                }                }
4137              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4138              if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)              if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)
4139                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4140              }              }
4141            break;            break;
4142    
# Line 3936  for (;;) Line 4146  for (;;)
4146              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4147                {                {
4148                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4149                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4150                }                }
4151              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4152              if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)              if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)
4153                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4154              }              }
4155            break;            break;
4156    
# Line 3950  for (;;) Line 4160  for (;;)
4160              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4161                {                {
4162                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4163                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4164                }                }
4165              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4166              if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result)              if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result)
4167                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4168              }              }
4169            break;            break;
4170    
# Line 3965  for (;;) Line 4175  for (;;)
4175              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4176                {                {
4177                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4178                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4179                }                }
4180              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4181              category = UCD_CATEGORY(c);              category = UCD_CATEGORY(c);
4182              if ((category == ucp_L || category == ucp_N) == prop_fail_result)              if ((category == ucp_L || category == ucp_N) == prop_fail_result)
4183                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4184              }              }
4185            break;            break;
4186    
# Line 3980  for (;;) Line 4190  for (;;)
4190              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4191                {                {
4192                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4193                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4194                }                }
4195              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4196              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
4197                   c == CHAR_FF || c == CHAR_CR)                   c == CHAR_FF || c == CHAR_CR)
4198                     == prop_fail_result)                     == prop_fail_result)
4199                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4200              }              }
4201            break;            break;
4202    
# Line 3996  for (;;) Line 4206  for (;;)
4206              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4207                {                {
4208                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4209                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4210                }                }
4211              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4212              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
4213                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)
4214                     == prop_fail_result)                     == prop_fail_result)
4215                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4216              }              }
4217            break;            break;
4218    
# Line 4013  for (;;) Line 4223  for (;;)
4223              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
4224                {                {
4225                SCHECK_PARTIAL();                SCHECK_PARTIAL();
4226                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4227                }                }
4228              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4229              category = UCD_CATEGORY(c);              category = UCD_CATEGORY(c);
4230              if ((category == ucp_L || category == ucp_N || c == CHAR_UNDERSCORE)              if ((category == ucp_L || category == ucp_N || c == CHAR_UNDERSCORE)
4231                     == prop_fail_result)                     == prop_fail_result)
4232                MRRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4233                }
4234              break;
4235    
4236              case PT_CLIST:
4237              for (i = 1; i <= min; i++)
4238                {
4239                const pcre_uint32 *cp;
4240                if (eptr >= md->end_subject)
4241                  {
4242                  SCHECK_PARTIAL();
4243                  RRETURN(MATCH_NOMATCH);
4244                  }
4245                GETCHARINCTEST(c, eptr);
4246                cp = PRIV(ucd_caseless_sets) + prop_value;
4247                for (;;)
4248                  {
4249                  if (c < *cp)
4250                    { if (prop_fail_result) break; else { RRETURN(MATCH_NOMATCH); } }
4251                  if (c == *cp++)
4252                    { if (prop_fail_result) { RRETURN(MATCH_NOMATCH); } else break; }
4253                  }
4254                }
4255              break;
4256    
4257              case PT_UCNC:
4258              for (i = 1; i <= min; i++)
4259                {
4260                if (eptr >= md->end_subject)
4261                  {
4262                  SCHECK_PARTIAL();
4263                  RRETURN(MATCH_NOMATCH);
4264                  }
4265                GETCHARINCTEST(c, eptr);
4266                if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
4267                     c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
4268                     c >= 0xe000) == prop_fail_result)
4269                  RRETURN(MATCH_NOMATCH);
4270              }              }
4271            break;            break;
4272    
# Line 4040  for (;;) Line 4287  for (;;)
4287            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4288              {              {
4289              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4290              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4291              }              }
4292            GETCHARINCTEST(c, eptr);            else
           if (UCD_CATEGORY(c) == ucp_M) MRRETURN(MATCH_NOMATCH);  
           while (eptr < md->end_subject)  
4293              {              {
4294              int len = 1;              int lgb, rgb;
4295              if (!utf8) c = *eptr; else { GETCHARLEN(c, eptr, len); }              GETCHARINCTEST(c, eptr);
4296              if (UCD_CATEGORY(c) != ucp_M) break;              lgb = UCD_GRAPHBREAK(c);
4297              eptr += len;             while (eptr < md->end_subject)
4298                  {
4299                  int len = 1;
4300                  if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
4301                  rgb = UCD_GRAPHBREAK(c);
4302                  if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
4303                  lgb = rgb;
4304                  eptr += len;
4305                  }
4306              }              }
4307              CHECK_PARTIAL();
4308            }            }
4309          }          }
4310    
# Line 4059  for (;;) Line 4313  for (;;)
4313    
4314  /* Handle all other cases when the coding is UTF-8 */  /* Handle all other cases when the coding is UTF-8 */
4315    
4316  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
4317        if (utf8) switch(ctype)        if (utf) switch(ctype)
4318          {          {
4319          case OP_ANY:          case OP_ANY:
4320          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
# Line 4068  for (;;) Line 4322  for (;;)
4322            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4323              {              {
4324              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4325              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4326                }
4327              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
4328              if (md->partial != 0 &&
4329                  eptr + 1 >= md->end_subject &&
4330                  NLBLOCK->nltype == NLTYPE_FIXED &&
4331                  NLBLOCK->nllen == 2 &&
4332                  RAWUCHAR(eptr) == NLBLOCK->nl[0])
4333                {
4334                md->hitend = TRUE;
4335                if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
4336              }              }
           if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);  
4337            eptr++;            eptr++;
4338            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4339            }            }
4340          break;          break;
4341    
# Line 4082  for (;;) Line 4345  for (;;)
4345            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4346              {              {
4347              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4348              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4349              }              }
4350            eptr++;            eptr++;
4351            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4352            }            }
4353          break;          break;
4354    
4355          case OP_ANYBYTE:          case OP_ANYBYTE:
4356          if (eptr > md->end_subject - min) MRRETURN(MATCH_NOMATCH);          if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
4357          eptr += min;          eptr += min;
4358          break;          break;
4359    
# Line 4100  for (;;) Line 4363  for (;;)
4363            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4364              {              {
4365              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4366              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4367              }              }
4368            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4369            switch(c)            switch(c)
4370              {              {
4371              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4372    
4373              case 0x000d:              case CHAR_CR:
4374              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && RAWUCHAR(eptr) == CHAR_LF) eptr++;
4375              break;              break;
4376    
4377              case 0x000a:              case CHAR_LF:
4378              break;              break;
4379    
4380              case 0x000b:              case CHAR_VT:
4381              case 0x000c:              case CHAR_FF:
4382              case 0x0085:              case CHAR_NEL:
4383    #ifndef EBCDIC
4384              case 0x2028:              case 0x2028:
4385              case 0x2029:              case 0x2029:
4386              if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);  #endif  /* Not EBCDIC */
4387                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4388              break;              break;
4389              }              }
4390            }            }
# Line 4131  for (;;) Line 4396  for (;;)
4396            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4397              {              {
4398              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4399              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4400              }              }
4401            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4402            switch(c)            switch(c)
4403              {              {
4404                HSPACE_CASES: RRETURN(MATCH_NOMATCH);  /* Byte and multibyte cases */
4405              default: break;              default: break;
             case 0x09:      /* HT */  
             case 0x20:      /* SPACE */  
             case 0xa0:      /* NBSP */  
             case 0x1680:    /* OGHAM SPACE MARK */  
             case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */  
             case 0x2000:    /* EN QUAD */  
             case 0x2001:    /* EM QUAD */  
             case 0x2002:    /* EN SPACE */  
             case 0x2003:    /* EM SPACE */  
             case 0x2004:    /* THREE-PER-EM SPACE */  
             case 0x2005:    /* FOUR-PER-EM SPACE */  
             case 0x2006:    /* SIX-PER-EM SPACE */  
             case 0x2007:    /* FIGURE SPACE */  
             case 0x2008:    /* PUNCTUATION SPACE */  
             case 0x2009:    /* THIN SPACE */  
             case 0x200A:    /* HAIR SPACE */  
             case 0x202f:    /* NARROW NO-BREAK SPACE */  
             case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */  
             case 0x3000:    /* IDEOGRAPHIC SPACE */  
             MRRETURN(MATCH_NOMATCH);  
4406              }              }
4407            }            }
4408          break;          break;
# Line 4167  for (;;) Line 4413  for (;;)
4413            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4414              {              {
4415              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4416              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4417              }              }
4418            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4419            switch(c)            switch(c)
4420              {              {
4421              default: MRRETURN(MATCH_NOMATCH);              HSPACE_CASES: break;  /* Byte and multibyte cases */
4422              case 0x09:      /* HT */              default: RRETURN(MATCH_NOMATCH);
             case 0x20:      /* SPACE */  
             case 0xa0:      /* NBSP */  
             case 0x1680:    /* OGHAM SPACE MARK */  
             case 0x180e:    /* MONGOLIAN VOWEL SEPARATOR */  
             case 0x2000:    /* EN QUAD */  
             case 0x2001:    /* EM QUAD */  
             case 0x2002:    /* EN SPACE */  
             case 0x2003:    /* EM SPACE */  
             case 0x2004:    /* THREE-PER-EM SPACE */  
             case 0x2005:    /* FOUR-PER-EM SPACE */  
             case 0x2006:    /* SIX-PER-EM SPACE */  
             case 0x2007:    /* FIGURE SPACE */  
             case 0x2008:    /* PUNCTUATION SPACE */  
             case 0x2009:    /* THIN SPACE */  
             case 0x200A:    /* HAIR SPACE */  
             case 0x202f:    /* NARROW NO-BREAK SPACE */  
             case 0x205f:    /* MEDIUM MATHEMATICAL SPACE */  
             case 0x3000:    /* IDEOGRAPHIC SPACE */  
             break;  
4423              }              }
4424            }            }
4425          break;          break;
# Line 4203  for (;;) Line 4430  for (;;)
4430            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4431              {              {
4432              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4433              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4434              }              }
4435            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4436            switch(c)            switch(c)
4437              {              {
4438                VSPACE_CASES: RRETURN(MATCH_NOMATCH);
4439              default: break;              default: break;
             case 0x0a:      /* LF */  
             case 0x0b:      /* VT */  
             case 0x0c:      /* FF */  
             case 0x0d:      /* CR */  
             case 0x85:      /* NEL */  
             case 0x2028:    /* LINE SEPARATOR */  
             case 0x2029:    /* PARAGRAPH SEPARATOR */  
             MRRETURN(MATCH_NOMATCH);  
4440              }              }
4441            }            }
4442          break;          break;
# Line 4227  for (;;) Line 4447  for (;;)
4447            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4448              {              {
4449              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4450              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4451              }              }
4452            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4453            switch(c)            switch(c)
4454              {              {
4455              default: MRRETURN(MATCH_NOMATCH);              VSPACE_CASES: break;
4456              case 0x0a:      /* LF */              default: RRETURN(MATCH_NOMATCH);
             case 0x0b:      /* VT */  
             case 0x0c:      /* FF */  
             case 0x0d:      /* CR */  
             case 0x85:      /* NEL */  
             case 0x2028:    /* LINE SEPARATOR */  
             case 0x2029:    /* PARAGRAPH SEPARATOR */  
             break;  
4457              }              }
4458            }            }
4459          break;          break;
# Line 4251  for (;;) Line 4464  for (;;)
4464            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4465              {              {
4466              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4467              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4468              }              }
4469            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4470            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)            if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)
4471              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4472            }            }
4473          break;          break;
4474    
4475          case OP_DIGIT:          case OP_DIGIT:
4476          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4477            {            {
4478              pcre_uint32 cc;
4479            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4480              {              {
4481              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4482              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4483              }              }
4484            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0)            cc = RAWUCHAR(eptr);
4485              MRRETURN(MATCH_NOMATCH);            if (cc >= 128 || (md->ctypes[cc] & ctype_digit) == 0)
4486                RRETURN(MATCH_NOMATCH);
4487              eptr++;
4488            /* 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 */
4489            }            }
4490          break;          break;
# Line 4276  for (;;) Line 4492  for (;;)
4492          case OP_NOT_WHITESPACE:          case OP_NOT_WHITESPACE:
4493          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4494            {            {
4495              pcre_uint32 cc;
4496            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4497              {              {
4498              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4499              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4500              }              }
4501            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)            cc = RAWUCHAR(eptr);
4502              MRRETURN(MATCH_NOMATCH);            if (cc < 128 && (md->ctypes[cc] & ctype_space) != 0)
4503            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);              RRETURN(MATCH_NOMATCH);
4504              eptr++;
4505              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4506            }            }
4507          break;          break;
4508    
4509          case OP_WHITESPACE:          case OP_WHITESPACE:
4510          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4511            {            {
4512              pcre_uint32 cc;
4513            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4514              {              {
4515              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4516              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4517              }              }
4518            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0)            cc = RAWUCHAR(eptr);
4519              MRRETURN(MATCH_NOMATCH);            if (cc >= 128 || (md->ctypes[cc] & ctype_space) == 0)
4520                RRETURN(MATCH_NOMATCH);
4521              eptr++;
4522            /* 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 */
4523            }            }
4524          break;          break;
# Line 4304  for (;;) Line 4526  for (;;)
4526          case OP_NOT_WORDCHAR:          case OP_NOT_WORDCHAR:
4527          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4528            {            {
4529              pcre_uint32 cc;
4530            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4531              {              {
4532              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4533              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4534              }              }
4535            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0)            cc = RAWUCHAR(eptr);
4536              MRRETURN(MATCH_NOMATCH);            if (cc < 128 && (md->ctypes[cc] & ctype_word) != 0)
4537            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);              RRETURN(MATCH_NOMATCH);
4538              eptr++;
4539              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4540            }            }
4541          break;          break;
4542    
4543          case OP_WORDCHAR:          case OP_WORDCHAR:
4544          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
4545            {            {
4546              pcre_uint32 cc;
4547            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4548              {              {
4549              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4550              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4551              }              }
4552            if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0)            cc = RAWUCHAR(eptr);
4553              MRRETURN(MATCH_NOMATCH);            if (cc >= 128 || (md->ctypes[cc] & ctype_word) == 0)
4554                RRETURN(MATCH_NOMATCH);
4555              eptr++;
4556            /* 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 */
4557            }            }
4558          break;          break;
# Line 4334  for (;;) Line 4562  for (;;)
4562          }  /* End switch(ctype) */          }  /* End switch(ctype) */
4563    
4564        else        else
4565  #endif     /* SUPPORT_UTF8 */  #endif     /* SUPPORT_UTF */
4566    
4567        /* 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
4568        than OP_PROP and OP_NOTPROP. */        than OP_PROP and OP_NOTPROP. */
# Line 4347  for (;;) Line 4575  for (;;)
4575            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4576              {              {
4577              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4578              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4579                }
4580              if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
4581              if (md->partial != 0 &&
4582                  eptr + 1 >= md->end_subject &&
4583                  NLBLOCK->nltype == NLTYPE_FIXED &&
4584                  NLBLOCK->nllen == 2 &&
4585                  *eptr == NLBLOCK->nl[0])
4586                {
4587                md->hitend = TRUE;
4588                if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
4589              }              }
           if (IS_NEWLINE(eptr)) MRRETURN(MATCH_NOMATCH);  
4590            eptr++;            eptr++;
4591            }            }
4592          break;          break;
# Line 4358  for (;;) Line 4595  for (;;)
4595          if (eptr > md->end_subject - min)          if (eptr > md->end_subject - min)
4596            {            {
4597            SCHECK_PARTIAL();            SCHECK_PARTIAL();
4598            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
4599            }            }
4600          eptr += min;          eptr += min;
4601          break;          break;
# Line 4367  for (;;) Line 4604  for (;;)
4604          if (eptr > md->end_subject - min)          if (eptr > md->end_subject - min)
4605            {            {
4606            SCHECK_PARTIAL();            SCHECK_PARTIAL();
4607            MRRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
4608            }            }
4609          eptr += min;          eptr += min;
4610          break;          break;
# Line 4378  for (;;) Line 4615  for (;;)
4615            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
4616              {              {
4617              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4618              MRRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4619              }              }
4620            switch(*eptr++)            switch(*eptr++)
4621              {              {
4622              default: MRRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4623    
4624              case 0x000d:              case CHAR_CR:
4625              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == CHAR_LF) eptr++;
4626              break;              break;
4627    
4628              case 0x000a:              case CHAR_LF:
4629              break;              break;
4630    
4631              case 0x000b:              case CHAR_VT:
4632              case 0x000c:              case CHAR_FF:
4633              case 0x0085:              case CHAR_NEL:
4634              if (md->bsr_anycrlf) MRRETURN(MATCH_NOMATCH);  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
4635                case 0x2028:
4636                case 0x2029:
4637    #endif
4638                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4639              break;              break;
4640              }              }
4641            }            }
# Line 4406  for (;;) Line 4647  for (;;)
46