/[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 891 by ph10, Wed Jan 18 16:26:18 2012 UTC revision 1260 by ph10, Wed Feb 27 15:41:22 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) 1997-2012 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 93  because the offset vector is always a mu Line 106  because the offset vector is always a mu
106  static const char rep_min[] = { 0, 0, 1, 1, 0, 0 };  static const char rep_min[] = { 0, 0, 1, 1, 0, 0 };
107  static const char rep_max[] = { 0, 0, 0, 0, 1, 1 };  static const char rep_max[] = { 0, 0, 0, 0, 1, 1 };
108    
   
   
109  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
110  /*************************************************  /*************************************************
111  *        Debugging function to print chars       *  *        Debugging function to print chars       *
# Line 115  Returns:     nothing Line 126  Returns:     nothing
126  static void  static void
127  pchars(const pcre_uchar *p, int length, BOOL is_subject, match_data *md)  pchars(const pcre_uchar *p, int length, BOOL is_subject, match_data *md)
128  {  {
129  unsigned int c;  pcre_uint32 c;
130    BOOL utf = md->utf;
131  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;
132  while (length-- > 0)  while (length-- > 0)
133    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);
134  }  }
135  #endif  #endif
136    
# Line 140  Arguments: Line 152  Arguments:
152    md          points to match data block    md          points to match data block
153    caseless    TRUE if caseless    caseless    TRUE if caseless
154    
155  Returns:      < 0 if not matched, otherwise the number of subject bytes matched  Returns:      >= 0 the number of subject bytes matched
156                  -1 no match
157                  -2 partial match; always given if at end subject
158  */  */
159    
160  static int  static int
# Line 149  match_ref(int offset, register PCRE_PUCH Line 163  match_ref(int offset, register PCRE_PUCH
163  {  {
164  PCRE_PUCHAR eptr_start = eptr;  PCRE_PUCHAR eptr_start = eptr;
165  register PCRE_PUCHAR p = md->start_subject + md->offset_vector[offset];  register PCRE_PUCHAR p = md->start_subject + md->offset_vector[offset];
166    #ifdef SUPPORT_UTF
167    BOOL utf = md->utf;
168    #endif
169    
170  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
171  if (eptr >= md->end_subject)  if (eptr >= md->end_subject)
# Line 163  pchars(p, length, FALSE, md); Line 180  pchars(p, length, FALSE, md);
180  printf("\n");  printf("\n");
181  #endif  #endif
182    
183  /* Always fail if reference not set (and not JavaScript compatible). */  /* Always fail if reference not set (and not JavaScript compatible - in that
184    case the length is passed as zero). */
185    
186  if (length < 0) return -1;  if (length < 0) return -1;
187    
# Line 175  if (caseless) Line 193  if (caseless)
193    {    {
194  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
195  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
196    if (md->utf)    if (utf)
197      {      {
198      /* 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
199      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
200      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.
201      (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
202      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
203      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
204      reference, not along the subject (earlier code did this wrong). */      length along the reference, not along the subject (earlier code did this
205        wrong). */
206    
207      PCRE_PUCHAR endptr = p + length;      PCRE_PUCHAR endptr = p + length;
208      while (p < endptr)      while (p < endptr)
209        {        {
210        int c, d;        pcre_uint32 c, d;
211        if (eptr >= md->end_subject) return -1;        const ucd_record *ur;
212          if (eptr >= md->end_subject) return -2;   /* Partial match */
213        GETCHARINC(c, eptr);        GETCHARINC(c, eptr);
214        GETCHARINC(d, p);        GETCHARINC(d, p);
215        if (c != d && c != UCD_OTHERCASE(d)) return -1;        ur = GET_UCD(d);
216          if (c != d && c != d + ur->other_case)
217            {
218            const pcre_uint32 *pp = PRIV(ucd_caseless_sets) + ur->caseset;
219            for (;;)
220              {
221              if (c < *pp) return -1;
222              if (c == *pp++) break;
223              }
224            }
225        }        }
226      }      }
227    else    else
# Line 202  if (caseless) Line 231  if (caseless)
231    /* 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
232    is no UCP support. */    is no UCP support. */
233      {      {
     if (eptr + length > md->end_subject) return -1;  
234      while (length-- > 0)      while (length-- > 0)
235        {        {
236        if (TABLE_GET(*p, md->lcc, *p) != TABLE_GET(*eptr, md->lcc, *eptr)) return -1;        pcre_uint32 cc, cp;
237          if (eptr >= md->end_subject) return -2;   /* Partial match */
238          cc = RAWUCHARTEST(eptr);
239          cp = RAWUCHARTEST(p);
240          if (TABLE_GET(cp, md->lcc, cp) != TABLE_GET(cc, md->lcc, cc)) return -1;
241        p++;        p++;
242        eptr++;        eptr++;
243        }        }
# Line 217  are in UTF-8 mode. */ Line 249  are in UTF-8 mode. */
249    
250  else  else
251    {    {
252    if (eptr + length > md->end_subject) return -1;    while (length-- > 0)
253    while (length-- > 0) if (*p++ != *eptr++) return -1;      {
254        if (eptr >= md->end_subject) return -2;   /* Partial match */
255        if (RAWUCHARINCTEST(p) != RAWUCHARINCTEST(eptr)) return -1;
256        }
257    }    }
258    
259  return (int)(eptr - eptr_start);  return (int)(eptr - eptr_start);
# Line 273  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM Line 308  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM
308         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
309         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
310         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,
311         RM61,  RM62, RM63, RM64, RM65, RM66 };         RM61,  RM62, RM63, RM64, RM65, RM66, RM67, RM68 };
312    
313  /* 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
314  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 291  actually used in this definition. */ Line 326  actually used in this definition. */
326    }    }
327  #define RRETURN(ra) \  #define RRETURN(ra) \
328    { \    { \
329    printf("match() returned %d from line %d ", ra, __LINE__); \    printf("match() returned %d from line %d\n", ra, __LINE__); \
330    return ra; \    return ra; \
331    }    }
332  #else  #else
# Line 311  argument of match(), which never changes Line 346  argument of match(), which never changes
346    
347  #define RMATCH(ra,rb,rc,rd,re,rw)\  #define RMATCH(ra,rb,rc,rd,re,rw)\
348    {\    {\
349    heapframe *newframe = (heapframe *)(PUBL(stack_malloc))(sizeof(heapframe));\    heapframe *newframe = frame->Xnextframe;\
350    if (newframe == NULL) RRETURN(PCRE_ERROR_NOMEMORY);\    if (newframe == NULL)\
351    frame->Xwhere = rw; \      {\
352        newframe = (heapframe *)(PUBL(stack_malloc))(sizeof(heapframe));\
353        if (newframe == NULL) RRETURN(PCRE_ERROR_NOMEMORY);\
354        newframe->Xnextframe = NULL;\
355        frame->Xnextframe = newframe;\
356        }\
357      frame->Xwhere = rw;\
358    newframe->Xeptr = ra;\    newframe->Xeptr = ra;\
359    newframe->Xecode = rb;\    newframe->Xecode = rb;\
360    newframe->Xmstart = mstart;\    newframe->Xmstart = mstart;\
# Line 332  argument of match(), which never changes Line 373  argument of match(), which never changes
373    {\    {\
374    heapframe *oldframe = frame;\    heapframe *oldframe = frame;\
375    frame = oldframe->Xprevframe;\    frame = oldframe->Xprevframe;\
   (PUBL(stack_free))(oldframe);\  
376    if (frame != NULL)\    if (frame != NULL)\
377      {\      {\
378      rrc = ra;\      rrc = ra;\
# Line 346  argument of match(), which never changes Line 386  argument of match(), which never changes
386    
387  typedef struct heapframe {  typedef struct heapframe {
388    struct heapframe *Xprevframe;    struct heapframe *Xprevframe;
389      struct heapframe *Xnextframe;
390    
391    /* Function arguments that may change */    /* Function arguments that may change */
392    
# Line 376  typedef struct heapframe { Line 417  typedef struct heapframe {
417    
418  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
419    int Xprop_type;    int Xprop_type;
420    int Xprop_value;    unsigned int Xprop_value;
421    int Xprop_fail_result;    int Xprop_fail_result;
422    int Xoclength;    int Xoclength;
423    pcre_uchar Xocchars[6];    pcre_uchar Xocchars[6];
# Line 389  typedef struct heapframe { Line 430  typedef struct heapframe {
430    int Xlength;    int Xlength;
431    int Xmax;    int Xmax;
432    int Xmin;    int Xmin;
433    int Xnumber;    unsigned int Xnumber;
434    int Xoffset;    int Xoffset;
435    int Xop;    unsigned int Xop;
436    int Xsave_capture_last;    pcre_int32 Xsave_capture_last;
437    int Xsave_offset1, Xsave_offset2, Xsave_offset3;    int Xsave_offset1, Xsave_offset2, Xsave_offset3;
438    int Xstacksave[REC_STACK_SAVE_MAX];    int Xstacksave[REC_STACK_SAVE_MAX];
439    
# Line 477  so they can be ordinary variables in all Line 518  so they can be ordinary variables in all
518    
519  register int  rrc;         /* Returns from recursive calls */  register int  rrc;         /* Returns from recursive calls */
520  register int  i;           /* Used for loops not involving calls to RMATCH() */  register int  i;           /* Used for loops not involving calls to RMATCH() */
521  register unsigned int c;   /* Character values not kept over RMATCH() calls */  register pcre_uint32 c;    /* Character values not kept over RMATCH() calls */
522  register BOOL utf;         /* Local copy of UTF flag for speed */  register BOOL utf;         /* Local copy of UTF flag for speed */
523    
524  BOOL minimize, possessive; /* Quantifier options */  BOOL minimize, possessive; /* Quantifier options */
# Line 485  BOOL caseless; Line 526  BOOL caseless;
526  int condcode;  int condcode;
527    
528  /* 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
529  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
530  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
531  heap whenever RMATCH() does a "recursion". See the macro definitions above. */  whenever RMATCH() does a "recursion". See the macro definitions above. Putting
532    the top-level on the stack rather than malloc-ing them all gives a performance
533    boost in many cases where there is not much "recursion". */
534    
535  #ifdef NO_RECURSE  #ifdef NO_RECURSE
536  heapframe *frame = (heapframe *)(PUBL(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 */  
537    
538  /* Copy in the original argument variables */  /* Copy in the original argument variables */
539    
# Line 594  BOOL prev_is_word; Line 635  BOOL prev_is_word;
635    
636  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
637  int prop_type;  int prop_type;
638  int prop_value;  unsigned int prop_value;
639  int prop_fail_result;  int prop_fail_result;
640  int oclength;  int oclength;
641  pcre_uchar occhars[6];  pcre_uchar occhars[6];
# Line 605  int ctype; Line 646  int ctype;
646  int length;  int length;
647  int max;  int max;
648  int min;  int min;
649  int number;  unsigned int number;
650  int offset;  int offset;
651  int op;  unsigned int op;
652  int save_capture_last;  pcre_int32 save_capture_last;
653  int save_offset1, save_offset2, save_offset3;  int save_offset1, save_offset2, save_offset3;
654  int stacksave[REC_STACK_SAVE_MAX];  int stacksave[REC_STACK_SAVE_MAX];
655    
656  eptrblock newptrb;  eptrblock newptrb;
657    
658    /* There is a special fudge for calling match() in a way that causes it to
659    measure the size of its basic stack frame when the stack is being used for
660    recursion. The second argument (ecode) being NULL triggers this behaviour. It
661    cannot normally ever be NULL. The return is the negated value of the frame
662    size. */
663    
664    if (ecode == NULL)
665      {
666      if (rdepth == 0)
667        return match((PCRE_PUCHAR)&rdepth, NULL, NULL, 0, NULL, NULL, 1);
668      else
669        {
670        int len = (char *)&rdepth - (char *)eptr;
671        return (len > 0)? -len : len;
672        }
673      }
674  #endif     /* NO_RECURSE */  #endif     /* NO_RECURSE */
675    
676  /* 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 709  for (;;) Line 767  for (;;)
767      unaltered. */      unaltered. */
768    
769      else if (rrc == MATCH_SKIP_ARG &&      else if (rrc == MATCH_SKIP_ARG &&
770          STRCMP_UC_UC(ecode + 2, md->start_match_ptr) == 0)          STRCMP_UC_UC_TEST(ecode + 2, md->start_match_ptr) == 0)
771        {        {
772        md->start_match_ptr = eptr;        md->start_match_ptr = eptr;
773        RRETURN(MATCH_SKIP);        RRETURN(MATCH_SKIP);
# Line 819  for (;;) Line 877  for (;;)
877      case OP_ONCE_NC:      case OP_ONCE_NC:
878      prev = ecode;      prev = ecode;
879      saved_eptr = eptr;      saved_eptr = eptr;
880      save_mark = md->mark;      save_mark = md->mark;
881      do      do
882        {        {
883        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM64);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM64);
# Line 838  for (;;) Line 896  for (;;)
896    
897        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
898        ecode += GET(ecode,1);        ecode += GET(ecode,1);
899        md->mark = save_mark;        md->mark = save_mark;
900        }        }
901      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
902    
# Line 878  for (;;) Line 936  for (;;)
936        }        }
937      else  /* OP_KETRMAX */      else  /* OP_KETRMAX */
938        {        {
       md->match_function_type = MATCH_CBEGROUP;  
939        RMATCH(eptr, prev, offset_top, md, eptrb, RM66);        RMATCH(eptr, prev, offset_top, md, eptrb, RM66);
940        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
941        ecode += 1 + LINK_SIZE;        ecode += 1 + LINK_SIZE;
# Line 918  for (;;) Line 975  for (;;)
975        save_offset2 = md->offset_vector[offset+1];        save_offset2 = md->offset_vector[offset+1];
976        save_offset3 = md->offset_vector[md->offset_end - number];        save_offset3 = md->offset_vector[md->offset_end - number];
977        save_capture_last = md->capture_last;        save_capture_last = md->capture_last;
978        save_mark = md->mark;        save_mark = md->mark;
979    
980        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));        DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));
981        md->offset_vector[md->offset_end - number] =        md->offset_vector[md->offset_end - number] =
# Line 1007  for (;;) Line 1064  for (;;)
1064    
1065      for (;;)      for (;;)
1066        {        {
1067        if (op >= OP_SBRA || op == OP_ONCE) md->match_function_type = MATCH_CBEGROUP;        if (op >= OP_SBRA || op == OP_ONCE)
1068            md->match_function_type = MATCH_CBEGROUP;
1069    
1070        /* 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
1071        the pattern, and this is the final alternative, optimize as described        the pattern, and this is the final alternative, optimize as described
# Line 1022  for (;;) Line 1080  for (;;)
1080        /* In all other cases, we have to make another call to match(). */        /* In all other cases, we have to make another call to match(). */
1081    
1082        save_mark = md->mark;        save_mark = md->mark;
1083          save_capture_last = md->capture_last;
1084        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb,        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb,
1085          RM2);          RM2);
1086    
1087        /* See comment in the code for capturing groups above about handling        /* See comment in the code for capturing groups above about handling
1088        THEN. */        THEN. */
1089    
# Line 1051  for (;;) Line 1110  for (;;)
1110          RRETURN(rrc);          RRETURN(rrc);
1111          }          }
1112        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1113        md->mark = save_mark;        md->mark = save_mark;
1114        if (*ecode != OP_ALT) break;        if (*ecode != OP_ALT) break;
1115          md->capture_last = save_capture_last;
1116        }        }
1117    
1118      RRETURN(MATCH_NOMATCH);      RRETURN(MATCH_NOMATCH);
# Line 1174  for (;;) Line 1234  for (;;)
1234      POSSESSIVE_NON_CAPTURE:      POSSESSIVE_NON_CAPTURE:
1235      matched_once = FALSE;      matched_once = FALSE;
1236      code_offset = (int)(ecode - md->start_code);      code_offset = (int)(ecode - md->start_code);
1237        save_capture_last = md->capture_last;
1238    
1239      for (;;)      for (;;)
1240        {        {
# Line 1203  for (;;) Line 1264  for (;;)
1264        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
1265        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
1266        if (*ecode != OP_ALT) break;        if (*ecode != OP_ALT) break;
1267          md->capture_last = save_capture_last;
1268        }        }
1269    
1270      if (matched_once || allow_zero)      if (matched_once || allow_zero)
# Line 1234  for (;;) Line 1296  for (;;)
1296          cb.version          = 2;   /* Version 1 of the callout block */          cb.version          = 2;   /* Version 1 of the callout block */
1297          cb.callout_number   = ecode[LINK_SIZE+2];          cb.callout_number   = ecode[LINK_SIZE+2];
1298          cb.offset_vector    = md->offset_vector;          cb.offset_vector    = md->offset_vector;
1299  #ifdef COMPILE_PCRE8  #if defined COMPILE_PCRE8
1300          cb.subject          = (PCRE_SPTR)md->start_subject;          cb.subject          = (PCRE_SPTR)md->start_subject;
1301  #else  #elif defined COMPILE_PCRE16
1302          cb.subject          = (PCRE_SPTR16)md->start_subject;          cb.subject          = (PCRE_SPTR16)md->start_subject;
1303    #elif defined COMPILE_PCRE32
1304            cb.subject          = (PCRE_SPTR32)md->start_subject;
1305  #endif  #endif
1306          cb.subject_length   = (int)(md->end_subject - md->start_subject);          cb.subject_length   = (int)(md->end_subject - md->start_subject);
1307          cb.start_match      = (int)(mstart - md->start_subject);          cb.start_match      = (int)(mstart - md->start_subject);
# Line 1245  for (;;) Line 1309  for (;;)
1309          cb.pattern_position = GET(ecode, LINK_SIZE + 3);          cb.pattern_position = GET(ecode, LINK_SIZE + 3);
1310          cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);          cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);
1311          cb.capture_top      = offset_top/2;          cb.capture_top      = offset_top/2;
1312          cb.capture_last     = md->capture_last;          cb.capture_last     = md->capture_last & CAPLMASK;
1313            /* Internal change requires this for API compatibility. */
1314            if (cb.capture_last == 0) cb.capture_last = -1;
1315          cb.callout_data     = md->callout_data;          cb.callout_data     = md->callout_data;
1316          cb.mark             = md->nomatch_mark;          cb.mark             = md->nomatch_mark;
1317          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
# Line 1267  for (;;) Line 1333  for (;;)
1333          }          }
1334        else        else
1335          {          {
1336          int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/          unsigned int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/
1337          condition = (recno == RREF_ANY || recno == md->recursive->group_num);          condition = (recno == RREF_ANY || recno == md->recursive->group_num);
1338    
1339          /* 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
# Line 1339  for (;;) Line 1405  for (;;)
1405    
1406        if (!condition && condcode == OP_NCREF)        if (!condition && condcode == OP_NCREF)
1407          {          {
1408          int refno = offset >> 1;          unsigned int refno = offset >> 1;
1409          pcre_uchar *slotA = md->name_table;          pcre_uchar *slotA = md->name_table;
1410    
1411          for (i = 0; i < md->name_count; i++)          for (i = 0; i < md->name_count; i++)
# Line 1467  for (;;) Line 1533  for (;;)
1533      to close any currently open capturing brackets. */      to close any currently open capturing brackets. */
1534    
1535      case OP_CLOSE:      case OP_CLOSE:
1536      number = GET2(ecode, 1);      number = GET2(ecode, 1);   /* Must be less than 65536 */
1537      offset = number << 1;      offset = number << 1;
1538    
1539  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
# Line 1475  for (;;) Line 1541  for (;;)
1541        printf("\n");        printf("\n");
1542  #endif  #endif
1543    
1544      md->capture_last = number;      md->capture_last = (md->capture_last & OVFLMASK) | number;
1545      if (offset >= md->offset_max) md->offset_overflow = TRUE; else      if (offset >= md->offset_max) md->capture_last |= OVFLBIT; else
1546        {        {
1547        md->offset_vector[offset] =        md->offset_vector[offset] =
1548          md->offset_vector[md->offset_end - number];          md->offset_vector[md->offset_end - number];
# Line 1530  for (;;) Line 1596  for (;;)
1596    
1597      case OP_ASSERT:      case OP_ASSERT:
1598      case OP_ASSERTBACK:      case OP_ASSERTBACK:
1599      save_mark = md->mark;      save_mark = md->mark;
1600      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1601        {        {
1602        condassert = TRUE;        condassert = TRUE;
# Line 1546  for (;;) Line 1612  for (;;)
1612          mstart = md->start_match_ptr;   /* In case \K reset it */          mstart = md->start_match_ptr;   /* In case \K reset it */
1613          break;          break;
1614          }          }
1615          md->mark = save_mark;
1616    
1617        /* PCRE does not allow THEN to escape beyond an assertion; it is treated        /* A COMMIT failure must fail the entire assertion, without trying any
1618        as NOMATCH. */        subsequent branches. */
1619    
1620          if (rrc == MATCH_COMMIT) RRETURN(MATCH_NOMATCH);
1621    
1622          /* PCRE does not allow THEN to escape beyond an assertion; it
1623          is treated as NOMATCH. */
1624    
1625        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1626        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
       md->mark = save_mark;  
1627        }        }
1628      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
1629    
# Line 1576  for (;;) Line 1647  for (;;)
1647    
1648      case OP_ASSERT_NOT:      case OP_ASSERT_NOT:
1649      case OP_ASSERTBACK_NOT:      case OP_ASSERTBACK_NOT:
1650      save_mark = md->mark;      save_mark = md->mark;
1651      if (md->match_function_type == MATCH_CONDASSERT)      if (md->match_function_type == MATCH_CONDASSERT)
1652        {        {
1653        condassert = TRUE;        condassert = TRUE;
# Line 1587  for (;;) Line 1658  for (;;)
1658      do      do
1659        {        {
1660        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);
1661        md->mark = save_mark;        md->mark = save_mark;
1662        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) RRETURN(MATCH_NOMATCH);        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) RRETURN(MATCH_NOMATCH);
1663        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)        if (rrc == MATCH_SKIP || rrc == MATCH_PRUNE || rrc == MATCH_COMMIT)
1664          {          {
# Line 1652  for (;;) Line 1723  for (;;)
1723        cb.version          = 2;   /* Version 1 of the callout block */        cb.version          = 2;   /* Version 1 of the callout block */
1724        cb.callout_number   = ecode[1];        cb.callout_number   = ecode[1];
1725        cb.offset_vector    = md->offset_vector;        cb.offset_vector    = md->offset_vector;
1726  #ifdef COMPILE_PCRE8  #if defined COMPILE_PCRE8
1727        cb.subject          = (PCRE_SPTR)md->start_subject;        cb.subject          = (PCRE_SPTR)md->start_subject;
1728  #else  #elif defined COMPILE_PCRE16
1729        cb.subject          = (PCRE_SPTR16)md->start_subject;        cb.subject          = (PCRE_SPTR16)md->start_subject;
1730    #elif defined COMPILE_PCRE32
1731          cb.subject          = (PCRE_SPTR32)md->start_subject;
1732  #endif  #endif
1733        cb.subject_length   = (int)(md->end_subject - md->start_subject);        cb.subject_length   = (int)(md->end_subject - md->start_subject);
1734        cb.start_match      = (int)(mstart - md->start_subject);        cb.start_match      = (int)(mstart - md->start_subject);
# Line 1663  for (;;) Line 1736  for (;;)
1736        cb.pattern_position = GET(ecode, 2);        cb.pattern_position = GET(ecode, 2);
1737        cb.next_item_length = GET(ecode, 2 + LINK_SIZE);        cb.next_item_length = GET(ecode, 2 + LINK_SIZE);
1738        cb.capture_top      = offset_top/2;        cb.capture_top      = offset_top/2;
1739        cb.capture_last     = md->capture_last;        cb.capture_last     = md->capture_last & CAPLMASK;
1740          /* Internal change requires this for API compatibility. */
1741          if (cb.capture_last == 0) cb.capture_last = -1;
1742        cb.callout_data     = md->callout_data;        cb.callout_data     = md->callout_data;
1743        cb.mark             = md->nomatch_mark;        cb.mark             = md->nomatch_mark;
1744        if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);        if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
# Line 1692  for (;;) Line 1767  for (;;)
1767      case OP_RECURSE:      case OP_RECURSE:
1768        {        {
1769        recursion_info *ri;        recursion_info *ri;
1770        int recno;        unsigned int recno;
1771    
1772        callpat = md->start_code + GET(ecode, 1);        callpat = md->start_code + GET(ecode, 1);
1773        recno = (callpat == md->start_code)? 0 :        recno = (callpat == md->start_code)? 0 :
# Line 1709  for (;;) Line 1784  for (;;)
1784        /* Add to "recursing stack" */        /* Add to "recursing stack" */
1785    
1786        new_recursive.group_num = recno;        new_recursive.group_num = recno;
1787          new_recursive.saved_capture_last = md->capture_last;
1788        new_recursive.subject_position = eptr;        new_recursive.subject_position = eptr;
1789        new_recursive.prevrec = md->recursive;        new_recursive.prevrec = md->recursive;
1790        md->recursive = &new_recursive;        md->recursive = &new_recursive;
# Line 1732  for (;;) Line 1808  for (;;)
1808              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
1809    
1810        /* OK, now we can do the recursion. After processing each alternative,        /* OK, now we can do the recursion. After processing each alternative,
1811        restore the offset data. If there were nested recursions, md->recursive        restore the offset data and the last captured value. If there were nested
1812        might be changed, so reset it before looping. */        recursions, md->recursive might be changed, so reset it before looping.
1813          */
1814    
1815        DPRINTF(("Recursing into group %d\n", new_recursive.group_num));        DPRINTF(("Recursing into group %d\n", new_recursive.group_num));
1816        cbegroup = (*callpat >= OP_SBRA);        cbegroup = (*callpat >= OP_SBRA);
# Line 1744  for (;;) Line 1821  for (;;)
1821            md, eptrb, RM6);            md, eptrb, RM6);
1822          memcpy(md->offset_vector, new_recursive.offset_save,          memcpy(md->offset_vector, new_recursive.offset_save,
1823              new_recursive.saved_max * sizeof(int));              new_recursive.saved_max * sizeof(int));
1824            md->capture_last = new_recursive.saved_capture_last;
1825          md->recursive = new_recursive.prevrec;          md->recursive = new_recursive.prevrec;
1826          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)          if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
1827            {            {
# Line 1760  for (;;) Line 1838  for (;;)
1838            goto RECURSION_MATCHED;        /* Exit loop; end processing */            goto RECURSION_MATCHED;        /* Exit loop; end processing */
1839            }            }
1840    
1841          /* PCRE does not allow THEN to escape beyond a recursion; it is treated          /* PCRE does not allow THEN or COMMIT to escape beyond a recursion; it
1842          as NOMATCH. */          is treated as NOMATCH. */
1843    
1844          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN &&
1845                     rrc != MATCH_COMMIT)
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)
# Line 1893  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 1974  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 1983  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 2040  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)) RRETURN(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) RRETURN(MATCH_NOMATCH);        if (md->noteol) RRETURN(MATCH_NOMATCH);
# Line 2072  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          {
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);        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 2200  for (;;) Line 2302  for (;;)
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)) RRETURN(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. */
# Line 2345  for (;;) Line 2460  for (;;)
2460        {        {
2461        default: RRETURN(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    #endif  /* Not EBCDIC */
2481        if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);        if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
2482        break;        break;
2483        }        }
# Line 2372  for (;;) Line 2493  for (;;)
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 */  
       RRETURN(MATCH_NOMATCH);  
2498        }        }
2499      ecode++;      ecode++;
2500      break;      break;
# Line 2406  for (;;) Line 2508  for (;;)
2508      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2509      switch(c)      switch(c)
2510        {        {
2511          HSPACE_CASES: break;  /* Byte and multibyte cases */
2512        default: RRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
       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 */  
       break;  
2513        }        }
2514      ecode++;      ecode++;
2515      break;      break;
# Line 2440  for (;;) Line 2523  for (;;)
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 */  
       RRETURN(MATCH_NOMATCH);  
2528        }        }
2529      ecode++;      ecode++;
2530      break;      break;
# Line 2462  for (;;) Line 2538  for (;;)
2538      GETCHARINCTEST(c, eptr);      GETCHARINCTEST(c, eptr);
2539      switch(c)      switch(c)
2540        {        {
2541          VSPACE_CASES: break;
2542        default: RRETURN(MATCH_NOMATCH);        default: RRETURN(MATCH_NOMATCH);
       case 0x0a:      /* LF */  
       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 2488  for (;;) Line 2557  for (;;)
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])
# Line 2548  for (;;) Line 2618  for (;;)
2618            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2619          break;          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;
2638    
2639          /* This should never occur */          /* This should never occur */
2640    
2641          default:          default:
# Line 2567  for (;;) Line 2655  for (;;)
2655        SCHECK_PARTIAL();        SCHECK_PARTIAL();
2656        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
2657        }        }
2658      GETCHARINCTEST(c, eptr);      else
     if (UCD_CATEGORY(c) == ucp_M) RRETURN(MATCH_NOMATCH);  
     while (eptr < md->end_subject)  
2659        {        {
2660        int len = 1;        int lgb, rgb;
2661        if (!utf) 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 2641  for (;;) Line 2736  for (;;)
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          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2742          }          }
# Line 2666  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          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2768          }          }
# Line 2689  for (;;) Line 2786  for (;;)
2786          if (fi >= max) RRETURN(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            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
2792            }            }
# Line 2707  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);
# Line 3075  for (;;) Line 3182  for (;;)
3182          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
3183          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3184          }          }
3185        while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);        while (length-- > 0) if (*ecode++ != RAWUCHARINC(eptr)) RRETURN(MATCH_NOMATCH);
3186        }        }
3187      else      else
3188  #endif  #endif
# Line 3115  for (;;) Line 3222  for (;;)
3222    
3223        if (fc < 128)        if (fc < 128)
3224          {          {
3225          if (md->lcc[fc]          pcre_uint32 cc = RAWUCHAR(eptr);
3226              != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH);          if (md->lcc[fc] != TABLE_GET(cc, md->lcc, cc)) RRETURN(MATCH_NOMATCH);
3227          ecode++;          ecode++;
3228          eptr++;          eptr++;
3229          }          }
# Line 3127  for (;;) Line 3234  for (;;)
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 3237  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 = PRIV(ord2utf)(othercase, occhars);            oclength = PRIV(ord2utf)(othercase, occhars);
# Line 3341  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        {        {
# Line 3364  for (;;) Line 3471  for (;;)
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            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3479            }            }
3480          if (fc != *eptr && foc != *eptr) RRETURN(MATCH_NOMATCH);          cc = RAWUCHARTEST(eptr);
3481            if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
3482          eptr++;          eptr++;
3483          }          }
3484        if (min == max) continue;        if (min == max) continue;
# Line 3377  for (;;) Line 3486  for (;;)
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) RRETURN(MATCH_NOMATCH);            if (fi >= max) RRETURN(MATCH_NOMATCH);
# Line 3385  for (;;) Line 3495  for (;;)
3495              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3496              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3497              }              }
3498            if (fc != *eptr && foc != *eptr) RRETURN(MATCH_NOMATCH);            cc = RAWUCHARTEST(eptr);
3499              if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
3500            eptr++;            eptr++;
3501            }            }
3502          /* Control never gets here */          /* Control never gets here */
# Line 3395  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 != *eptr && foc != *eptr) break;            cc = RAWUCHARTEST(eptr);
3516              if (fc != cc && foc != cc) break;
3517            eptr++;            eptr++;
3518            }            }
3519    
# Line 3428  for (;;) Line 3541  for (;;)
3541            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3542            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3543            }            }
3544          if (fc != *eptr++) RRETURN(MATCH_NOMATCH);          if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);
3545          }          }
3546    
3547        if (min == max) continue;        if (min == max) continue;
# Line 3445  for (;;) Line 3558  for (;;)
3558              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3559              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3560              }              }
3561            if (fc != *eptr++) RRETURN(MATCH_NOMATCH);            if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);
3562            }            }
3563          /* Control never gets here */          /* Control never gets here */
3564          }          }
# Line 3459  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 3485  for (;;) Line 3598  for (;;)
3598        SCHECK_PARTIAL();        SCHECK_PARTIAL();
3599        RRETURN(MATCH_NOMATCH);        RRETURN(MATCH_NOMATCH);
3600        }        }
     ecode++;  
     GETCHARINCTEST(c, eptr);  
     if (op == OP_NOTI)         /* The caseless case */  
       {  
       register int ch, och;  
       ch = *ecode++;  
 #ifdef COMPILE_PCRE8  
       /* ch must be < 128 if UTF is enabled. */  
       och = md->fcc[ch];  
 #else  
3601  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3602        if (utf)
3603          {
3604          register pcre_uint32 ch, och;
3605    
3606          ecode++;
3607          GETCHARINC(ch, ecode);
3608          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  #ifdef SUPPORT_UCP
3617        if (utf && ch > 127)          if (ch > 127)
3618          och = UCD_OTHERCASE(ch);            och = UCD_OTHERCASE(ch);
3619  #else  #else
3620        if (utf && ch > 127)          if (ch > 127)
3621          och = ch;            och = ch;
3622  #endif /* SUPPORT_UCP */  #endif /* SUPPORT_UCP */
3623        else          else
3624  #endif /* SUPPORT_UTF */            och = TABLE_GET(ch, md->fcc, ch);
3625          och = TABLE_GET(ch, md->fcc, ch);          if (ch == c || och == c) RRETURN(MATCH_NOMATCH);
3626  #endif /* COMPILE_PCRE8 */          }
       if (ch == c || och == c) RRETURN(MATCH_NOMATCH);  
3627        }        }
3628      else    /* Caseful */      else
3629    #endif
3630        {        {
3631        if (*ecode++ == c) RRETURN(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 3591  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 3602  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        {        {
 #ifdef COMPILE_PCRE8  
       /* fc must be < 128 if UTF is enabled. */  
       foc = md->fcc[fc];  
 #else  
3730  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3731  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3732        if (utf && fc > 127)        if (utf && fc > 127)
# Line 3621  for (;;) Line 3738  for (;;)
3738        else        else
3739  #endif /* SUPPORT_UTF */  #endif /* SUPPORT_UTF */
3740          foc = TABLE_GET(fc, md->fcc, fc);          foc = TABLE_GET(fc, md->fcc, fc);
 #endif /* COMPILE_PCRE8 */  
3741    
3742  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3743        if (utf)        if (utf)
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)
# Line 3635  for (;;) Line 3751  for (;;)
3751              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3752              }              }
3753            GETCHARINC(d, eptr);            GETCHARINC(d, eptr);
3754            if (fc == d || foc == d) RRETURN(MATCH_NOMATCH);            if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH);
3755            }            }
3756          }          }
3757        else        else
# Line 3661  for (;;) Line 3777  for (;;)
3777  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3778          if (utf)          if (utf)
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);
# Line 3673  for (;;) Line 3789  for (;;)
3789                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
3790                }                }
3791              GETCHARINC(d, eptr);              GETCHARINC(d, eptr);
3792              if (fc == d || foc == d) RRETURN(MATCH_NOMATCH);              if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH);
3793              }              }
3794            }            }
3795          else          else
# Line 3706  for (;;) Line 3822  for (;;)
3822  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3823          if (utf)          if (utf)
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 3716  for (;;) Line 3832  for (;;)
3832                break;                break;
3833                }                }
3834              GETCHARLEN(d, eptr, len);              GETCHARLEN(d, eptr, len);
3835              if (fc == d || foc == d) break;              if (fc == d || (unsigned int)foc == d) break;
3836              eptr += len;              eptr += len;
3837              }              }
3838            if (possessive) continue;            if (possessive) continue;
# Line 3763  for (;;) Line 3879  for (;;)
3879  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3880        if (utf)        if (utf)
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)
# Line 3797  for (;;) Line 3913  for (;;)
3913  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3914          if (utf)          if (utf)
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);
# Line 3841  for (;;) Line 3957  for (;;)
3957  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3958          if (utf)          if (utf)
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 4117  for (;;) Line 4233  for (;;)
4233              }              }
4234            break;            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;
4272    
4273            /* This should not occur */            /* This should not occur */
4274    
4275            default:            default:
# Line 4136  for (;;) Line 4289  for (;;)
4289              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4290              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4291              }              }
4292            GETCHARINCTEST(c, eptr);            else
           if (UCD_CATEGORY(c) == ucp_M) RRETURN(MATCH_NOMATCH);  
           while (eptr < md->end_subject)  
4293              {              {
4294              int len = 1;              int lgb, rgb;
4295              if (!utf) 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 4165  for (;;) Line 4325  for (;;)
4325              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4326              }              }
4327            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);            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                }
4337            eptr++;            eptr++;
4338            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
4339            }            }
# Line 4201  for (;;) Line 4370  for (;;)
4370              {              {
4371              default: RRETURN(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    #endif  /* Not EBCDIC */
4387              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
4388              break;              break;
4389              }              }
# Line 4230  for (;;) Line 4401  for (;;)
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 */  
             RRETURN(MATCH_NOMATCH);  
4406              }              }
4407            }            }
4408          break;          break;
# Line 4266  for (;;) Line 4418  for (;;)
4418            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4419            switch(c)            switch(c)
4420              {              {
4421                HSPACE_CASES: break;  /* Byte and multibyte cases */
4422              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
             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 */  
             break;  
4423              }              }
4424            }            }
4425          break;          break;
# Line 4302  for (;;) Line 4435  for (;;)
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 */  
             RRETURN(MATCH_NOMATCH);  
4440              }              }
4441            }            }
4442          break;          break;
# Line 4326  for (;;) Line 4452  for (;;)
4452            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
4453            switch(c)            switch(c)
4454              {              {
4455                VSPACE_CASES: break;
4456              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
             case 0x0a:      /* LF */  
             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 4356  for (;;) Line 4475  for (;;)
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              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4483              }              }
4484            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_digit) == 0)            cc = RAWUCHAR(eptr);
4485              if (cc >= 128 || (md->ctypes[cc] & ctype_digit) == 0)
4486              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4487            eptr++;            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 */
# Line 4371  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              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4500              }              }
4501            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)            cc = RAWUCHAR(eptr);
4502              if (cc < 128 && (md->ctypes[cc] & ctype_space) != 0)
4503              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4504            eptr++;            eptr++;
4505            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
# Line 4386  for (;;) Line 4509  for (;;)
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              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4517              }              }
4518            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_space) == 0)            cc = RAWUCHAR(eptr);
4519              if (cc >= 128 || (md->ctypes[cc] & ctype_space) == 0)
4520              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4521            eptr++;            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 */
# Line 4401  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              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4534              }              }
4535            if (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0)            cc = RAWUCHAR(eptr);
4536              if (cc < 128 && (md->ctypes[cc] & ctype_word) != 0)
4537              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4538            eptr++;            eptr++;
4539            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);            ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
# Line 4416  for (;;) Line 4543  for (;;)
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              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4551              }              }
4552            if (*eptr >= 128 || (md->ctypes[*eptr] & ctype_word) == 0)            cc = RAWUCHAR(eptr);
4553              if (cc >= 128 || (md->ctypes[cc] & ctype_word) == 0)
4554              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4555            eptr++;            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 */
# Line 4449  for (;;) Line 4578  for (;;)
4578              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4579              }              }
4580            if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);            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                }
4590            eptr++;            eptr++;
4591            }            }
4592          break;          break;
# Line 4483  for (;;) Line 4621  for (;;)
4621              {              {
4622              default: RRETURN(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  #ifdef COMPILE_PCRE16  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
4635              case 0x2028:              case 0x2028:
4636              case 0x2029:              case 0x2029:
4637  #endif  #endif
# Line 4514  for (;;) Line 4652  for (;;)
4652            switch(*eptr++)            switch(*eptr++)
4653              {              {
4654              default: break;              default: break;
4655              case 0x09:      /* HT */              HSPACE_BYTE_CASES:
4656              case 0x20:      /* SPACE */  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
4657              case 0xa0:      /* NBSP */              HSPACE_MULTIBYTE_CASES:
 #ifdef COMPILE_PCRE16  
             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 */  
4658  #endif  #endif
4659              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4660              }              }
# Line 4551  for (;;) Line 4672  for (;;)
4672            switch(*eptr++)            switch(*eptr++)
4673              {              {
4674              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4675              case 0x09:      /* HT */              HSPACE_BYTE_CASES:
4676              case 0x20:      /* SPACE */  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
4677              case 0xa0:      /* NBSP */              HSPACE_MULTIBYTE_CASES:
 #ifdef COMPILE_PCRE16  
             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 */  
4678  #endif  #endif
4679              break;              break;
4680              }              }
# Line 4587  for (;;) Line 4691  for (;;)
4691              }              }
4692            switch(*eptr++)            switch(*eptr++)
4693              {              {
4694              default: break;              VSPACE_BYTE_CASES:
4695              case 0x0a:      /* LF */  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
4696              case 0x0b:      /* VT */              VSPACE_MULTIBYTE_CASES:
             case 0x0c:      /* FF */  
             case 0x0d:      /* CR */  
             case 0x85:      /* NEL */  
 #ifdef COMPILE_PCRE16  
             case 0x2028:    /* LINE SEPARATOR */  
             case 0x2029:    /* PARAGRAPH SEPARATOR */  
4697  #endif  #endif
4698              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4699                default: break;
4700              }              }
4701            }            }
4702          break;          break;
# Line 4613  for (;;) Line 4712  for (;;)
4712            switch(*eptr++)            switch(*eptr++)
4713              {              {
4714              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4715              case 0x0a:      /* LF */              VSPACE_BYTE_CASES:
4716              case 0x0b:      /* VT */  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
4717              case 0x0c:      /* FF */              VSPACE_MULTIBYTE_CASES:
             case 0x0d:      /* CR */  
             case 0x85:      /* NEL */  
 #ifdef COMPILE_PCRE16  
             case 0x2028:    /* LINE SEPARATOR */  
             case 0x2029:    /* PARAGRAPH SEPARATOR */  
4718  #endif  #endif
4719              break;              break;
4720              }              }
# Line 4898  for (;;) Line 4992  for (;;)
4992              }              }
4993            /* Control never gets here */            /* Control never gets here */
4994    
4995            /* This should never occur */            case PT_CLIST:
4996              for (fi = min;; fi++)
4997                {
4998                const pcre_uint32 *cp;
4999                RMATCH(eptr, ecode, offset_top, md, eptrb, RM67);
5000                if (rrc != MATCH_NOMATCH) RRETURN(rrc);
5001                if (fi >= max) RRETURN(MATCH_NOMATCH);
5002                if (eptr >= md->end_subject)
5003                  {
5004                  SCHECK_PARTIAL();
5005                  RRETURN(MATCH_NOMATCH);
5006                  }
5007                GETCHARINCTEST(c, eptr);
5008                cp = PRIV(ucd_caseless_sets) + prop_value;
5009                for (;;)
5010                  {
5011                  if (c < *cp)
5012                    { if (prop_fail_result) break; else { RRETURN(MATCH_NOMATCH); } }
5013                  if (c == *cp++)
5014                    { if (prop_fail_result) { RRETURN(MATCH_NOMATCH); } else break; }
5015                  }
5016                }
5017              /* Control never gets here */
5018    
5019              case PT_UCNC:
5020              for (fi = min;; fi++)
5021                {
5022                RMATCH(eptr, ecode, offset_top, md, eptrb, RM68);
5023                if (rrc != MATCH_NOMATCH) RRETURN(rrc);
5024                if (fi >= max) RRETURN(MATCH_NOMATCH);
5025                if (eptr >= md->end_subject)
5026                  {
5027                  SCHECK_PARTIAL();
5028                  RRETURN(MATCH_NOMATCH);
5029                  }
5030                GETCHARINCTEST(c, eptr);
5031                if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
5032                     c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
5033                     c >= 0xe000) == prop_fail_result)
5034                  RRETURN(MATCH_NOMATCH);
5035                }
5036              /* Control never gets here */
5037    
5038              /* This should never occur */
5039            default:            default:
5040            RRETURN(PCRE_ERROR_INTERNAL);            RRETURN(PCRE_ERROR_INTERNAL);
5041            }            }
# Line 4920  for (;;) Line 5056  for (;;)
5056              SCHECK_PARTIAL();              SCHECK_PARTIAL();
5057              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
5058              }              }
5059            GETCHARINCTEST(c, eptr);            else
           if (UCD_CATEGORY(c) == ucp_M) RRETURN(MATCH_NOMATCH);  
           while (eptr < md->end_subject)  
5060              {              {
5061              int len = 1;              int lgb, rgb;
5062              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }              GETCHARINCTEST(c, eptr);
5063              if (UCD_CATEGORY(c) != ucp_M) break;              lgb = UCD_GRAPHBREAK(c);
5064              eptr += len;              while (eptr < md->end_subject)
5065                  {
5066                  int len = 1;
5067                  if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
5068                  rgb = UCD_GRAPHBREAK(c);
5069                  if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
5070                  lgb = rgb;
5071                  eptr += len;
5072                  }
5073              }              }
5074              CHECK_PARTIAL();
5075            }            }
5076          }          }
5077        else        else
# Line 4952  for (;;) Line 5095  for (;;)
5095            GETCHARINC(c, eptr);            GETCHARINC(c, eptr);
5096            switch(ctype)            switch(ctype)
5097              {              {
5098              case OP_ANY:        /* This is the non-NL case */              case OP_ANY:               /* This is the non-NL case */
5099                if (md->partial != 0 &&    /* Take care with CRLF partial */
5100                    eptr >= md->end_subject &&
5101                    NLBLOCK->nltype == NLTYPE_FIXED &&
5102                    NLBLOCK->nllen == 2 &&
5103                    c == NLBLOCK->nl[0])
5104                  {
5105                  md->hitend = TRUE;
5106                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5107                  }
5108                break;
5109    
5110              case OP_ALLANY:              case OP_ALLANY:
5111              case OP_ANYBYTE:              case OP_ANYBYTE:
5112              break;              break;
# Line 4961  for (;;) Line 5115  for (;;)
5115              switch(c)              switch(c)
5116                {                {
5117                default: RRETURN(MATCH_NOMATCH);                default: RRETURN(MATCH_NOMATCH);
5118                case 0x000d:                case CHAR_CR:
5119                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && RAWUCHAR(eptr) == CHAR_LF) eptr++;
5120                break;                break;
5121                case 0x000a:  
5122                  case CHAR_LF:
5123                break;                break;
5124    
5125                case 0x000b:                case CHAR_VT:
5126                case 0x000c:                case CHAR_FF:
5127                case 0x0085:                case CHAR_NEL:
5128    #ifndef EBCDIC
5129                case 0x2028:                case 0x2028:
5130                case 0x2029:                case 0x2029:
5131    #endif  /* Not EBCDIC */
5132                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
5133                break;                break;
5134                }                }
# Line 4980  for (;;) Line 5137  for (;;)
5137              case OP_NOT_HSPACE:              case OP_NOT_HSPACE:
5138              switch(c)              switch(c)
5139                {                {
5140                  HSPACE_CASES: RRETURN(MATCH_NOMATCH);
5141                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 */  
               RRETURN(MATCH_NOMATCH);  
5142                }                }
5143              break;              break;
5144    
5145              case OP_HSPACE:              case OP_HSPACE:
5146              switch(c)              switch(c)
5147                {                {
5148                  HSPACE_CASES: break;
5149                default: RRETURN(MATCH_NOMATCH);                default: RRETURN(MATCH_NOMATCH);
               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 */  
               break;  
5150                }                }
5151              break;              break;
5152    
5153              case OP_NOT_VSPACE:              case OP_NOT_VSPACE:
5154              switch(c)              switch(c)
5155                {                {
5156                  VSPACE_CASES: RRETURN(MATCH_NOMATCH);
5157                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 */  
               RRETURN(MATCH_NOMATCH);  
5158                }                }
5159              break;              break;
5160    
5161              case OP_VSPACE:              case OP_VSPACE:
5162              switch(c)              switch(c)
5163                {                {
5164                  VSPACE_CASES: break;
5165                default: RRETURN(MATCH_NOMATCH);                default: RRETURN(MATCH_NOMATCH);
               case 0x0a:      /* LF */  
               case 0x0b:      /* VT */  
               case 0x0c:      /* FF */  
               case 0x0d:      /* CR */  
               case 0x85:      /* NEL */  
               case 0x2028:    /* LINE SEPARATOR */  
               case 0x2029:    /* PARAGRAPH SEPARATOR */  
               break;  
5166                }                }
5167              break;              break;
5168    
# Line 5115  for (;;) Line 5220  for (;;)
5220            c = *eptr++;            c = *eptr++;
5221            switch(ctype)            switch(ctype)
5222              {              {
5223              case OP_ANY:     /* This is the non-NL case */              case OP_ANY:               /* This is the non-NL case */
5224                if (md->partial != 0 &&    /* Take care with CRLF partial */
5225                    eptr >= md->end_subject &&
5226                    NLBLOCK->nltype == NLTYPE_FIXED &&
5227                    NLBLOCK->nllen == 2 &&
5228                    c == NLBLOCK->nl[0])
5229                  {
5230                  md->hitend = TRUE;
5231                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5232                  }
5233                break;
5234    
5235              case OP_ALLANY:              case OP_ALLANY:
5236              case OP_ANYBYTE:              case OP_ANYBYTE:
5237              break;              break;
# Line 5124  for (;;) Line 5240  for (;;)
5240              switch(c)              switch(c)
5241                {                {
5242                default: RRETURN(MATCH_NOMATCH);                default: RRETURN(MATCH_NOMATCH);
5243                case 0x000d:                case CHAR_CR:
5244                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == CHAR_LF) eptr++;
5245                break;                break;
5246    
5247                case 0x000a:                case CHAR_LF:
5248                break;                break;
5249    
5250                case 0x000b:                case CHAR_VT:
5251                case 0x000c:                case CHAR_FF:
5252                case 0x0085:                case CHAR_NEL:
5253  #ifdef COMPILE_PCRE16  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
5254                case 0x2028:                case 0x2028:
5255                case 0x2029:                case 0x2029:
5256  #endif  #endif
# Line 5147  for (;;) Line 5263  for (;;)
5263              switch(c)              switch(c)
5264                {                {
5265                default: break;                default: break;
5266                case 0x09:      /* HT */                HSPACE_BYTE_CASES:
5267                case 0x20:      /* SPACE */  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
5268                case 0xa0:      /* NBSP */                HSPACE_MULTIBYTE_CASES:
 #ifdef COMPILE_PCRE16  
               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 */  
5269  #endif  #endif
5270                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
5271                }                }
# Line 5176  for (;;) Line 5275  for (;;)
5275              switch(c)              switch(c)
5276                {                {
5277                default: RRETURN(MATCH_NOMATCH);                default: RRETURN(MATCH_NOMATCH);
5278                case 0x09:      /* HT */                HSPACE_BYTE_CASES:
5279                case 0x20:      /* SPACE */  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
5280                case 0xa0:      /* NBSP */                HSPACE_MULTIBYTE_CASES:
 #ifdef COMPILE_PCRE16  
               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 */  
5281  #endif  #endif
5282                break;                break;
5283                }                }
# Line 5205  for (;;) Line 5287  for (;;)
5287              switch(c)              switch(c)
5288                {                {
5289                default: break;                default: break;
5290                case 0x0a:      /* LF */                VSPACE_BYTE_CASES:
5291                case 0x0b:      /* VT */  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
5292                case 0x0c:      /* FF */                VSPACE_MULTIBYTE_CASES:
               case 0x0d:      /* CR */  
               case 0x85:      /* NEL */  
 #ifdef COMPILE_PCRE16  
               case 0x2028:    /* LINE SEPARATOR */  
               case 0x2029:    /* PARAGRAPH SEPARATOR */  
5293  #endif  #endif
5294                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
5295                }                }
# Line 5222  for (;;) Line 5299  for (;;)
5299              switch(c)              switch(c)
5300                {                {
5301                default: RRETURN(MATCH_NOMATCH);                default: RRETURN(MATCH_NOMATCH);
5302                case 0x0a:      /* LF */                VSPACE_BYTE_CASES:
5303                case 0x0b:      /* VT */  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
5304                case 0x0c:      /* FF */                VSPACE_MULTIBYTE_CASES:
               case 0x0d:      /* CR */  
               case 0x85:      /* NEL */  
 #ifdef COMPILE_PCRE16  
               case 0x2028:    /* LINE SEPARATOR */  
               case 0x2029:    /* PARAGRAPH SEPARATOR */  
5305  #endif  #endif
5306                break;                break;
5307                }                }
# Line 5433  for (;;) Line 5505  for (;;)
5505              }              }
5506            break;            break;
5507    
5508              case PT_CLIST:
5509              for (i = min; i < max; i++)
5510                {
5511                const pcre_uint32 *cp;
5512                int len = 1;
5513                if (eptr >= md->end_subject)
5514                  {
5515                  SCHECK_PARTIAL();
5516                  break;
5517                  }
5518                GETCHARLENTEST(c, eptr, len);
5519                cp = PRIV(ucd_caseless_sets) + prop_value;
5520                for (;;)
5521                  {
5522                  if (c < *cp)
5523                    { if (prop_fail_result) break; else goto GOT_MAX; }
5524                  if (c == *cp++)
5525                    { if (prop_fail_result) goto GOT_MAX; else break; }
5526                  }
5527                eptr += len;
5528                }
5529              GOT_MAX:
5530              break;
5531    
5532              case PT_UCNC:
5533              for (i = min; i < max; i++)
5534                {
5535                int len = 1;
5536                if (eptr >= md->end_subject)
5537                  {
5538                  SCHECK_PARTIAL();
5539                  break;
5540                  }
5541                GETCHARLENTEST(c, eptr, len);
5542                if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
5543                     c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
5544                     c >= 0xe000) == prop_fail_result)
5545                  break;
5546                eptr += len;
5547                }
5548              break;
5549    
5550            default:            default:
5551            RRETURN(PCRE_ERROR_INTERNAL);            RRETURN(PCRE_ERROR_INTERNAL);
5552            }            }
# Line 5456  for (;;) Line 5570  for (;;)
5570          {          {
5571          for (i = min; i < max; i++)          for (i = min; i < max; i++)
5572            {            {
           int len = 1;  
5573            if (eptr >= md->end_subject)            if (eptr >= md->end_subject)
5574              {              {
5575              SCHECK_PARTIAL();              SCHECK_PARTIAL();
5576              break;              break;
5577              }              }
5578            if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }            else
           if (UCD_CATEGORY(c) == ucp_M) break;  
           eptr += len;  
           while (eptr < md->end_subject)  
5579              {              {
5580              len = 1;              int lgb, rgb;
5581              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }              GETCHARINCTEST(c, eptr);
5582              if (UCD_CATEGORY(c) != ucp_M) break;              lgb = UCD_GRAPHBREAK(c);
5583              eptr += len;              while (eptr < md->end_subject)
5584                  {
5585                  int len = 1;
5586                  if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
5587                  rgb = UCD_GRAPHBREAK(c);
5588                  if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
5589                  lgb = rgb;
5590                  eptr += len;
5591                  }
5592              }              }
5593              CHECK_PARTIAL();
5594            }            }
5595    
5596          /* eptr is now past the end of the maximum run */          /* eptr is now past the end of the maximum run */
# Line 5515  for (;;) Line 5634  for (;;)
5634                  break;                  break;
5635                  }                  }
5636                if (IS_NEWLINE(eptr)) break;                if (IS_NEWLINE(eptr)) break;
5637                  if (md->partial != 0 &&    /* Take care with CRLF partial */
5638                      eptr + 1 >= md->end_subject &&
5639                      NLBLOCK->nltype == NLTYPE_FIXED &&
5640                      NLBLOCK->nllen == 2 &&
5641                      RAWUCHAR(eptr) == NLBLOCK->nl[0])
5642                    {
5643                    md->hitend = TRUE;
5644                    if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5645                    }
5646                eptr++;                eptr++;
5647                ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);                ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
5648                }                }
# Line 5532  for (;;) Line 5660  for (;;)
5660                  break;                  break;
5661                  }                  }
5662                if (IS_NEWLINE(eptr)) break;                if (IS_NEWLINE(eptr)) break;
5663                  if (md->partial != 0 &&    /* Take care with CRLF partial */
5664                      eptr + 1 >= md->end_subject &&
5665                      NLBLOCK->nltype == NLTYPE_FIXED &&
5666                      NLBLOCK->nllen == 2 &&
5667                      RAWUCHAR(eptr) == NLBLOCK->nl[0])
5668                    {
5669                    md->hitend = TRUE;
5670                    if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5671                    }
5672                eptr++;                eptr++;
5673                ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);                ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
5674                }                }
# Line 5581  for (;;) Line 5718  for (;;)
5718                break;                break;
5719                }                }
5720              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
5721              if (c == 0x000d)              if (c == CHAR_CR)
5722                {                {
5723                if (++eptr >= md->end_subject) break;                if (++eptr >= md->end_subject) break;
5724                if (*eptr == 0x000a) eptr++;                if (RAWUCHAR(eptr) == CHAR_LF) eptr++;
5725                }                }
5726              else              else
5727                {                {
5728                if (c != 0x000a &&                if (c != CHAR_LF &&
5729                    (md->bsr_anycrlf ||                    (md->bsr_anycrlf ||
5730                     (c != 0x000b && c != 0x000c &&                     (c != CHAR_VT && c != CHAR_FF && c != CHAR_NEL
5731                      c != 0x0085 && c != 0x2028 && c != 0x2029)))  #ifndef EBCDIC
5732                        && c != 0x2028 && c != 0x2029
5733    #endif  /* Not EBCDIC */
5734                        )))
5735                  break;                  break;
5736                eptr += len;                eptr += len;
5737                }                }
# Line 5612  for (;;) Line 5752  for (;;)
5752              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
5753              switch(c)              switch(c)
5754                {                {
5755                  HSPACE_CASES: gotspace = TRUE; break;
5756                default: gotspace = FALSE; break;                default: gotspace = FALSE; 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 */  
               gotspace = TRUE;  
               break;  
5757                }                }
5758              if (gotspace == (ctype == OP_NOT_HSPACE)) break;              if (gotspace == (ctype == OP_NOT_HSPACE)) break;
5759              eptr += len;              eptr += len;
# Line 5654  for (;;) Line 5774  for (;;)
5774              GETCHARLEN(c, eptr, len);              GETCHARLEN(c, eptr, len);
5775              switch(c)              switch(c)
5776                {                {
5777                  VSPACE_CASES: gotspace = TRUE; break;
5778                default: gotspace = FALSE; break;                default: gotspace = FALSE; break;
               case 0x0a:      /* LF */  
               case 0x0b:      /* VT */  
               case 0x0c:      /* FF */  
               case 0x0d:      /* CR */  
               case 0x85:      /* NEL */  
               case 0x2028:    /* LINE SEPARATOR */  
               case 0x2029:    /* PARAGRAPH SEPARATOR */  
               gotspace = TRUE;  
               break;  
5779                }                }
5780              if (gotspace == (ctype == OP_NOT_VSPACE)) break;              if (gotspace == (ctype == OP_NOT_VSPACE)) break;
5781              eptr += len;              eptr += len;
# Line 5777  for (;;) Line 5889  for (;;)
5889            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
5890            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
5891            BACKCHAR(eptr);            BACKCHAR(eptr);
5892            if (ctype == OP_ANYNL && eptr > pp  && *eptr == '\n' &&            if (ctype == OP_ANYNL && eptr > pp  && RAWUCHAR(eptr) == CHAR_NL &&
5893                eptr[-1] == '\r') eptr--;                RAWUCHAR(eptr - 1) == CHAR_CR) eptr--;
5894            }            }
5895          }          }
5896        else        else
# Line 5796  for (;;) Line 5908  for (;;)
5908                break;                break;
5909                }                }
5910              if (IS_NEWLINE(eptr)) break;              if (IS_NEWLINE(eptr)) break;
5911                if (md->partial != 0 &&    /* Take care with CRLF partial */
5912                    eptr + 1 >= md->end_subject &&
5913                    NLBLOCK->nltype == NLTYPE_FIXED &&
5914                    NLBLOCK->nllen == 2 &&
5915                    *eptr == NLBLOCK->nl[0])
5916                  {
5917                  md->hitend = TRUE;
5918                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
5919                  }
5920              eptr++;              eptr++;
5921              }              }
5922            break;            break;
# Line 5820  for (;;) Line 5941  for (;;)
5941                break;                break;
5942                }                }
5943              c = *eptr;              c = *eptr;
5944              if (c == 0x000d)              if (c == CHAR_CR)
5945                {                {
5946                if (++eptr >= md->end_subject) break;                if (++eptr >= md->end_subject) break;
5947                if (*eptr == 0x000a) eptr++;                if (*eptr == CHAR_LF) eptr++;
5948                }                }
5949              else              else
5950                {                {
5951                if (c != 0x000a && (md->bsr_anycrlf ||                if (c != CHAR_LF && (md->bsr_anycrlf ||
5952                  (c != 0x000b && c != 0x000c && c != 0x0085                   (c != CHAR_VT && c != CHAR_FF && c != CHAR_NEL
5953  #ifdef COMPILE_PCRE16  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
5954                  && c != 0x2028 && c != 0x2029                   && c != 0x2028 && c != 0x2029
5955  #endif  #endif
5956                  ))) break;                   ))) break;
5957                eptr++;                eptr++;
5958                }                }
5959              }              }
# Line 5846  for (;;) Line 5967  for (;;)
5967                SCHECK_PARTIAL();                SCHECK_PARTIAL();
5968                break;                break;
5969                }                }
5970              c = *eptr;              switch(*eptr)
5971              if (c == 0x09 || c == 0x20 || c == 0xa0                {
5972  #ifdef COMPILE_PCRE16                default: eptr++; break;
5973                || c == 0x1680 || c == 0x180e || (c >= 0x2000 && c <= 0x200A)                HSPACE_BYTE_CASES:
5974                || c == 0x202f || c == 0x205f || c == 0x3000  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
5975                  HSPACE_MULTIBYTE_CASES:
5976  #endif  #endif
5977                ) break;                goto ENDLOOP00;
5978              eptr++;                }
5979              }              }
5980              ENDLOOP00:
5981            break;            break;
5982    
5983            case OP_HSPACE:            case OP_HSPACE:
# Line 5865  for (;;) Line 5988  for (;;)
5988                SCHECK_PARTIAL();                SCHECK_PARTIAL();
5989                break;                break;
5990                }                }
5991              c = *eptr;              switch(*eptr)
5992              if (c != 0x09 && c != 0x20 && c != 0xa0                {
5993  #ifdef COMPILE_PCRE16                default: goto ENDLOOP01;
5994                && c != 0x1680 && c != 0x180e && (c < 0x2000 || c > 0x200A)                HSPACE_BYTE_CASES:
5995                && c != 0x202f && c != 0x205f && c != 0x3000  #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
5996                  HSPACE_MULTIBYTE_CASES:
5997  #endif  #endif
5998                ) break;                eptr++; break;
5999              eptr++;                }
6000              }              }
6001              ENDLOOP01:
6002            break;            break;
6003    
6004            case OP_NOT_VSPACE:            case OP_NOT_VSPACE:
# Line 5884  for (;;) Line 6009  for (;;)
6009                SCHECK_PARTIAL();                SCHECK_PARTIAL();
6010                break;                break;
6011                }                }
6012              c = *eptr;              switch(*eptr)
6013              if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85                {
6014  #ifdef COMPILE_PCRE16                default: eptr++; break;
6015                || c == 0x2028 || c == 0x2029                VSPACE_BYTE_CASES:
6016    #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
6017                  VSPACE_MULTIBYTE_CASES:
6018  #endif  #endif
6019                ) break;                goto ENDLOOP02;
6020              eptr++;                }
6021              }              }
6022              ENDLOOP02:
6023            break;            break;
6024    
6025            case OP_VSPACE:            case OP_VSPACE:
# Line 5902  for (;;) Line 6030  for (;;)
6030                SCHECK_PARTIAL();                SCHECK_PARTIAL();
6031                break;                break;
6032                }                }
6033              c = *eptr;              switch(*eptr)
6034              if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85                {
6035  #ifdef COMPILE_PCRE16                default: goto ENDLOOP03;
6036                && c != 0x2028 && c != 0x2029                VSPACE_BYTE_CASES:
6037    #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
6038                  VSPACE_MULTIBYTE_CASES:
6039  #endif  #endif
6040                ) break;                eptr++; break;
6041              eptr++;                }
6042              }              }
6043              ENDLOOP03:
6044            break;            break;
6045    
6046            case OP_NOT_DIGIT:            case OP_NOT_DIGIT:
# Line 6006  for (;;) Line 6137  for (;;)
6137            RMATCH(eptr, ecode, offset_top, md, eptrb, RM47);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM47);
6138            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
6139            eptr--;            eptr--;
6140            if (ctype == OP_ANYNL && eptr > pp  && *eptr == '\n' &&            if (ctype == OP_ANYNL && eptr > pp  && *eptr == CHAR_LF &&
6141                eptr[-1] == '\r') eptr--;                eptr[-1] == CHAR_CR) eptr--;
6142            }            }
6143          }          }
6144    
# Line 6057  switch (frame->Xwhere) Line 6188  switch (frame->Xwhere)
6188    LBL(32) LBL(34) LBL(42) LBL(46)    LBL(32) LBL(34) LBL(42) LBL(46)
6189  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
6190    LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)    LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
6191    LBL(59) LBL(60) LBL(61) LBL(62)    LBL(59) LBL(60) LBL(61) LBL(62) LBL(67) LBL(68)
6192  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
6193  #endif  /* SUPPORT_UTF */  #endif  /* SUPPORT_UTF */
6194    default:    default:
6195    DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));    DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));
   
 printf("+++jump error in pcre match: label %d non-existent\n", frame->Xwhere);  
   
6196    return PCRE_ERROR_INTERNAL;    return PCRE_ERROR_INTERNAL;
6197    }    }
6198  #undef LBL  #undef LBL
# Line 6126  Undefine all the macros that were define Line 6254  Undefine all the macros that were define
6254  ***************************************************************************/  ***************************************************************************/
6255    
6256    
6257    #ifdef NO_RECURSE
6258    /*************************************************
6259    *          Release allocated heap frames         *
6260    *************************************************/
6261    
6262    /* This function releases all the allocated frames. The base frame is on the
6263    machine stack, and so must not be freed.
6264    
6265    Argument: the address of the base frame
6266    Returns:  nothing
6267    */
6268    
6269    static void
6270    release_match_heapframes (heapframe *frame_base)
6271    {
6272    heapframe *nextframe = frame_base->Xnextframe;
6273    while (nextframe != NULL)
6274      {
6275      heapframe *oldframe = nextframe;
6276      nextframe = nextframe->Xnextframe;
6277      (PUBL(stack_free))(oldframe);
6278      }
6279    }
6280    #endif
6281    
6282    
6283  /*************************************************  /*************************************************
6284  *         Execute a Regular Expression           *  *         Execute a Regular Expression           *
# Line 6151  Returns:          > 0 => success; value Line 6304  Returns:          > 0 => success; value
6304                   < -1 => some kind of unexpected problem                   < -1 => some kind of unexpected problem
6305  */  */
6306    
6307  #ifdef COMPILE_PCRE8  #if defined COMPILE_PCRE8
6308  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
6309  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,  pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,
6310    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,    PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,
6311    int offsetcount)    int offsetcount)
6312  #else  #elif defined COMPILE_PCRE16
6313  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
6314  pcre16_exec(const pcre16 *argument_re, const pcre16_extra *extra_data,  pcre16_exec(const pcre16 *argument_re, const pcre16_extra *extra_data,
6315    PCRE_SPTR16 subject, int length, int start_offset, int options, int *offsets,    PCRE_SPTR16 subject, int length, int start_offset, int options, int *offsets,
6316    int offsetcount)    int offsetcount)
6317    #elif defined COMPILE_PCRE32
6318    PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
6319    pcre32_exec(const pcre32 *argument_re, const pcre32_extra *extra_data,
6320      PCRE_SPTR32 subject, int length, int start_offset, int options, int *offsets,
6321      int offsetcount)
6322  #endif  #endif
6323  {  {
6324  int rc, ocount, arg_offset_max;  int rc, ocount, arg_offset_max;
# Line 6183  const pcre_uint8 *start_bits = NULL; Line 6341  const pcre_uint8 *start_bits = NULL;
6341  PCRE_PUCHAR start_match = (PCRE_PUCHAR)subject + start_offset;  PCRE_PUCHAR start_match = (PCRE_PUCHAR)subject + start_offset;
6342  PCRE_PUCHAR end_subject;  PCRE_PUCHAR end_subject;
6343  PCRE_PUCHAR start_partial = NULL;  PCRE_PUCHAR start_partial = NULL;
6344    PCRE_PUCHAR match_partial;
6345  PCRE_PUCHAR req_char_ptr = start_match - 1;  PCRE_PUCHAR req_char_ptr = start_match - 1;
6346    
6347  const pcre_study_data *study;  const pcre_study_data *study;
6348  const REAL_PCRE *re = (const REAL_PCRE *)argument_re;  const REAL_PCRE *re = (const REAL_PCRE *)argument_re;
6349    
6350    #ifdef NO_RECURSE
6351    heapframe frame_zero;
6352    frame_zero.Xprevframe = NULL;            /* Marks the top level */
6353    frame_zero.Xnextframe = NULL;            /* None are allocated yet */
6354    md->match_frames_base = &frame_zero;
6355    #endif
6356    
6357    /* Check for the special magic call that measures the size of the stack used
6358    per recursive call of match(). Without the funny casting for sizeof, a Windows
6359    compiler gave this error: "unary minus operator applied to unsigned type,
6360    result still unsigned". Hopefully the cast fixes that. */
6361    
6362    if (re == NULL && extra_data == NULL && subject == NULL && length == -999 &&
6363        start_offset == -999)
6364    #ifdef NO_RECURSE
6365      return -((int)sizeof(heapframe));
6366    #else
6367      return match(NULL, NULL, NULL, 0, NULL, NULL, 0);
6368    #endif
6369    
6370  /* Plausibility checks */  /* Plausibility checks */
6371    
6372  if ((options & ~PUBLIC_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION;  if ((options & ~PUBLIC_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION;
6373  if (re == NULL || subject == NULL ||  if (re == NULL || subject == NULL || (offsets == NULL && offsetcount > 0))
6374     (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL;    return PCRE_ERROR_NULL;
6375  if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;  if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;
6376    if (length < 0) return PCRE_ERROR_BADLENGTH;
6377  if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET;  if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET;
6378    
6379    /* Check that the first field in the block is the magic number. If it is not,
6380    return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to
6381    REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which
6382    means that the pattern is likely compiled with different endianness. */
6383    
6384    if (re->magic_number != MAGIC_NUMBER)
6385      return re->magic_number == REVERSED_MAGIC_NUMBER?
6386        PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC;
6387    if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
6388    
6389  /* These two settings are used in the code for checking a UTF-8 string that  /* These two settings are used in the code for checking a UTF-8 string that
6390  follows immediately afterwards. Other values in the md block are used only  follows immediately afterwards. Other values in the md block are used only
6391  during "normal" pcre_exec() processing, not when the JIT support is in use,  during "normal" pcre_exec() processing, not when the JIT support is in use,
# Line 6221  if (utf && (options & PCRE_NO_UTF8_CHECK Line 6411  if (utf && (options & PCRE_NO_UTF8_CHECK
6411        offsets[0] = erroroffset;        offsets[0] = erroroffset;
6412        offsets[1] = errorcode;        offsets[1] = errorcode;
6413        }        }
6414  #ifdef COMPILE_PCRE16  #if defined COMPILE_PCRE8
     return (errorcode <= PCRE_UTF16_ERR1 && md->partial > 1)?  
       PCRE_ERROR_SHORTUTF16 : PCRE_ERROR_BADUTF16;  
 #else  
6415      return (errorcode <= PCRE_UTF8_ERR5 && md->partial > 1)?      return (errorcode <= PCRE_UTF8_ERR5 && md->partial > 1)?
6416        PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8;        PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8;
6417    #elif defined COMPILE_PCRE16
6418        return (errorcode <= PCRE_UTF16_ERR1 && md->partial > 1)?
6419          PCRE_ERROR_SHORTUTF16 : PCRE_ERROR_BADUTF16;
6420    #elif defined COMPILE_PCRE32
6421        return PCRE_ERROR_BADUTF32;
6422  #endif  #endif
6423      }      }
6424    #if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
6425    /* Check that a start_offset points to the start of a UTF character. */    /* Check that a start_offset points to the start of a UTF character. */
6426    if (start_offset > 0 && start_offset < length &&    if (start_offset > 0 && start_offset < length &&
6427        NOT_FIRSTCHAR(((PCRE_PUCHAR)subject)[start_offset]))        NOT_FIRSTCHAR(((PCRE_PUCHAR)subject)[start_offset]))
6428      return PCRE_ERROR_BADUTF8_OFFSET;      return PCRE_ERROR_BADUTF8_OFFSET;
6429    #endif
6430    }    }
6431  #endif  #endif
6432    
6433  /* If the pattern was successfully studied with JIT support, run the JIT  /* If the pattern was successfully studied with JIT support, run the JIT
6434  executable instead of the rest of this function. Most options must be set at  executable instead of the rest of this function. Most options must be set at
6435  compile time for the JIT code to be usable. Fallback to the normal code path if  compile time for the JIT code to be usable. Fallback to the normal code path if
6436  an unsupported flag is set. In particular, JIT does not support partial  an unsupported flag is set. */
 matching. */  
6437    
6438  #ifdef SUPPORT_JIT  #ifdef SUPPORT_JIT
6439  if (extra_data != NULL  if (extra_data != NULL
6440      && (extra_data->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0      && (extra_data->flags & (PCRE_EXTRA_EXECUTABLE_JIT |
6441                                 PCRE_EXTRA_TABLES)) == PCRE_EXTRA_EXECUTABLE_JIT
6442      && extra_data->executable_jit != NULL      && extra_data->executable_jit != NULL
6443      && (extra_data->flags & PCRE_EXTRA_TABLES) == 0      && (options & ~PUBLIC_JIT_EXEC_OPTIONS) == 0)
6444      && (options & ~(PCRE_NO_UTF8_CHECK | PCRE_NOTBOL | PCRE_NOTEOL |    {
6445                      PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART)) == 0)    rc = PRIV(jit_exec)(extra_data, (const pcre_uchar *)subject, length,
6446    return PRIV(jit_exec)(re, extra_data->executable_jit,         start_offset, options, offsets, offsetcount);
6447      (const pcre_uchar *)subject, length, start_offset, options,  
6448      ((extra_data->flags & PCRE_EXTRA_MATCH_LIMIT) == 0)    /* PCRE_ERROR_NULL means that the selected normal or partial matching
6449      ? MATCH_LIMIT : extra_data->match_limit, offsets, offsetcount);    mode is not compiled. In this case we simply fallback to interpreter. */
6450    
6451      if (rc != PCRE_ERROR_JIT_BADOPTION) return rc;
6452      }
6453  #endif  #endif
6454    
6455  /* Carry on with non-JIT matching. This information is for finding all the  /* Carry on with non-JIT matching. This information is for finding all the
# Line 6295  in other programs later. */ Line 6491  in other programs later. */
6491    
6492  if (tables == NULL) tables = PRIV(default_tables);  if (tables == NULL) tables = PRIV(default_tables);
6493    
 /* Check that the first field in the block is the magic number. If it is not,  
 return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to  
 REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which  
 means that the pattern is likely compiled with different endianness. */  
   
 if (re->magic_number != MAGIC_NUMBER)  
   return re->magic_number == REVERSED_MAGIC_NUMBER?  
     PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC;  
 if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;  
   
6494  /* Set up other data */  /* Set up other data */
6495    
6496  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
# Line 6434  if (re->top_backref > 0 && re->top_backr Line 6620  if (re->top_backref > 0 && re->top_backr
6620    DPRINTF(("Got memory to hold back references\n"));    DPRINTF(("Got memory to hold back references\n"));
6621    }    }
6622  else md->offset_vector = offsets;  else md->offset_vector = offsets;
   
6623  md->offset_end = ocount;  md->offset_end = ocount;
6624  md->offset_max = (2*ocount)/3;  md->offset_max = (2*ocount)/3;
6625  md->offset_overflow = FALSE;  md->capture_last = 0;
 md->capture_last = -1;  
6626    
6627  /* Reset the working variable associated with each extraction. These should  /* Reset the working variable associated with each extraction. These should
6628  never be used unless previously set, but they get saved and restored, and so we  never be used unless previously set, but they get saved and restored, and so we
# Line 6466  if (!anchored) Line 6650  if (!anchored)
6650    if ((re->flags & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
6651      {      {
6652      has_first_char = TRUE;      has_first_char = TRUE;
6653      first_char = first_char2 = re->first_char;      first_char = first_char2 = (pcre_uchar)(re->first_char);
6654      if ((re->flags & PCRE_FCH_CASELESS) != 0)      if ((re->flags & PCRE_FCH_CASELESS) != 0)
6655        {        {
6656        first_char2 = TABLE_GET(first_char, md->fcc, first_char);        first_char2 = TABLE_GET(first_char, md->fcc, first_char);
# Line 6488  character" set. */ Line 6672  character" set. */
6672  if ((re->flags & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
6673    {    {
6674    has_req_char = TRUE;    has_req_char = TRUE;
6675    req_char = req_char2 = re->req_char;    req_char = req_char2 = (pcre_uchar)(re->req_char);
6676    if ((re->flags & PCRE_RCH_CASELESS) != 0)    if ((re->flags & PCRE_RCH_CASELESS) != 0)
6677      {      {
6678      req_char2 = TABLE_GET(req_char, md->fcc, req_char);      req_char2 = TABLE_GET(req_char, md->fcc, req_char);
# Line 6546  for(;;) Line 6730  for(;;)
6730    
6731      if (has_first_char)      if (has_first_char)
6732        {        {
6733          pcre_uchar smc;
6734    
6735        if (first_char != first_char2)        if (first_char != first_char2)
6736          while (start_match < end_subject &&          while (start_match < end_subject &&
6737              *start_match != first_char && *start_match != first_char2)            (smc = RAWUCHARTEST(start_match)) != first_char && smc != first_char2)
6738            start_match++;            start_match++;
6739        else        else
6740          while (start_match < end_subject && *start_match != first_char)          while (start_match < end_subject && RAWUCHARTEST(start_match) != first_char)
6741            start_match++;            start_match++;
6742        }        }
6743    
# Line 6583  for(;;) Line 6769  for(;;)
6769          if (start_match[-1] == CHAR_CR &&          if (start_match[-1] == CHAR_CR &&
6770               (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&               (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
6771               start_match < end_subject &&               start_match < end_subject &&
6772               *start_match == CHAR_NL)               RAWUCHARTEST(start_match) == CHAR_NL)
6773            start_match++;            start_match++;
6774          }          }
6775        }        }
# Line 6594  for(;;) Line 6780  for(;;)
6780        {        {
6781        while (start_match < end_subject)        while (start_match < end_subject)
6782          {          {
6783          register unsigned int c = *start_match;          register pcre_uint32 c = RAWUCHARTEST(start_match);
6784  #ifndef COMPILE_PCRE8  #ifndef COMPILE_PCRE8
6785          if (c > 255) c = 255;          if (c > 255) c = 255;
6786  #endif  #endif
# Line 6662  for(;;) Line 6848  for(;;)
6848            {            {
6849            while (p < end_subject)            while (p < end_subject)
6850              {              {
6851              register int pp = *p++;              register pcre_uint32 pp = RAWUCHARINCTEST(p);
6852              if (pp == req_char || pp == req_char2) { p--; break; }              if (pp == req_char || pp == req_char2) { p--; break; }
6853              }              }
6854            }            }
# Line 6670  for(;;) Line 6856  for(;;)
6856            {            {
6857            while (p < end_subject)            while (p < end_subject)
6858              {              {
6859              if (*p++ == req_char) { p--; break; }              if (RAWUCHARINCTEST(p) == req_char) { p--; break; }
6860              }              }
6861            }            }
6862    
# Line 6707  for(;;) Line 6893  for(;;)
6893    md->match_function_type = 0;    md->match_function_type = 0;
6894    md->end_offset_top = 0;    md->end_offset_top = 0;
6895    rc = match(start_match, md->start_code, start_match, 2, md, NULL, 0);    rc = match(start_match, md->start_code, start_match, 2, md, NULL, 0);
6896    if (md->hitend && start_partial == NULL) start_partial = md->start_used_ptr;    if (md->hitend && start_partial == NULL)
6897        {
6898        start_partial = md->start_used_ptr;
6899        match_partial = start_match;
6900        }
6901    
6902    switch(rc)    switch(rc)
6903      {      {
# Line 6833  if (rc == MATCH_MATCH || rc == MATCH_ACC Line 7023  if (rc == MATCH_MATCH || rc == MATCH_ACC
7023          (arg_offset_max - 2) * sizeof(int));          (arg_offset_max - 2) * sizeof(int));
7024        DPRINTF(("Copied offsets from temporary memory\n"));        DPRINTF(("Copied offsets from temporary memory\n"));
7025        }        }
7026      if (md->end_offset_top > arg_offset_max) md->offset_overflow = TRUE;      if (md->end_offset_top > arg_offset_max) md->capture_last |= OVFLBIT;
7027      DPRINTF(("Freeing temporary memory\n"));      DPRINTF(("Freeing temporary memory\n"));
7028      (PUBL(free))(md->offset_vector);      (PUBL(free))(md->offset_vector);
7029      }      }
# Line 6841  if (rc == MATCH_MATCH || rc == MATCH_ACC Line 7031  if (rc == MATCH_MATCH || rc == MATCH_ACC
7031    /* Set the return code to the number of captured strings, or 0 if there were    /* Set the return code to the number of captured strings, or 0 if there were
7032    too many to fit into the vector. */    too many to fit into the vector. */
7033    
7034    rc = (md->offset_overflow && md->end_offset_top >= arg_offset_max)?    rc = ((md->capture_last & OVFLBIT) != 0 &&
7035             md->end_offset_top >= arg_offset_max)?
7036      0 : md->end_offset_top/2;      0 : md->end_offset_top/2;
7037    
7038    /* If there is space in the offset vector, set any unused pairs at the end of    /* If there is space in the offset vector, set any unused pairs at the end of
# Line 6857  if (rc == MATCH_MATCH || rc == MATCH_ACC Line 7048  if (rc == MATCH_MATCH || rc == MATCH_ACC
7048      {      {
7049      register int *iptr, *iend;      register int *iptr, *iend;
7050      int resetcount = 2 + re->top_bracket * 2;      int resetcount = 2 + re->top_bracket * 2;
7051      if (resetcount > offsetcount) resetcount = ocount;      if (resetcount > offsetcount) resetcount = offsetcount;
7052      iptr = offsets + md->end_offset_top;      iptr = offsets + md->end_offset_top;
7053      iend = offsets + resetcount;      iend = offsets + resetcount;
7054      while (iptr < iend) *iptr++ = -1;      while (iptr < iend) *iptr++ = -1;
# Line 6878  if (rc == MATCH_MATCH || rc == MATCH_ACC Line 7069  if (rc == MATCH_MATCH || rc == MATCH_ACC
7069    if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)    if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)
7070      *(extra_data->mark) = (pcre_uchar *)md->mark;      *(extra_data->mark) = (pcre_uchar *)md->mark;
7071    DPRINTF((">>>> returning %d\n", rc));    DPRINTF((">>>> returning %d\n", rc));
7072    #ifdef NO_RECURSE
7073      release_match_heapframes(&frame_zero);
7074    #endif
7075    return rc;    return rc;
7076    }    }
7077    
# Line 6895  if (using_temporary_offsets) Line 7089  if (using_temporary_offsets)
7089  if (rc != MATCH_NOMATCH && rc != PCRE_ERROR_PARTIAL)  if (rc != MATCH_NOMATCH && rc != PCRE_ERROR_PARTIAL)
7090    {    {
7091    DPRINTF((">>>> error: returning %d\n", rc));    DPRINTF((">>>> error: returning %d\n", rc));
7092    #ifdef NO_RECURSE
7093      release_match_heapframes(&frame_zero);
7094    #endif
7095    return rc;    return rc;
7096    }    }
7097    
# Line 6908  if (start_partial != NULL) Line 7105  if (start_partial != NULL)
7105      {      {
7106      offsets[0] = (int)(start_partial - (PCRE_PUCHAR)subject);      offsets[0] = (int)(start_partial - (PCRE_PUCHAR)subject);
7107      offsets[1] = (int)(end_subject - (PCRE_PUCHAR)subject);      offsets[1] = (int)(end_subject - (PCRE_PUCHAR)subject);
7108        if (offsetcount > 2)
7109          offsets[2] = (int)(match_partial - (PCRE_PUCHAR)subject);
7110      }      }
7111    rc = PCRE_ERROR_PARTIAL;    rc = PCRE_ERROR_PARTIAL;
7112    }    }
# Line 6924  else Line 7123  else
7123    
7124  if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)  if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)
7125    *(extra_data->mark) = (pcre_uchar *)md->nomatch_mark;    *(extra_data->mark) = (pcre_uchar *)md->nomatch_mark;
7126    #ifdef NO_RECURSE
7127      release_match_heapframes(&frame_zero);
7128    #endif
7129  return rc;  return rc;
7130  }  }
7131    

Legend:
Removed from v.891  
changed lines
  Added in v.1260

  ViewVC Help
Powered by ViewVC 1.1.5