/[pcre]/code/trunk/pcre_dfa_exec.c
ViewVC logotype

Diff of /code/trunk/pcre_dfa_exec.c

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 530 by ph10, Tue Jun 1 13:42:06 2010 UTC revision 1260 by ph10, Wed Feb 27 15:41:22 2013 UTC
# Line 7  and semantics are as close as possible t Line 7  and semantics are as close as possible t
7  below for why this module is different).  below for why this module is different).
8    
9                         Written by Philip Hazel                         Written by Philip Hazel
10             Copyright (c) 1997-2010 University of Cambridge             Copyright (c) 1997-2013 University of Cambridge
11    
12  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
13  Redistribution and use in source and binary forms, with or without  Redistribution and use in source and binary forms, with or without
# Line 38  POSSIBILITY OF SUCH DAMAGE. Line 38  POSSIBILITY OF SUCH DAMAGE.
38  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
39  */  */
40    
   
41  /* This module contains the external function pcre_dfa_exec(), which is an  /* This module contains the external function pcre_dfa_exec(), which is an
42  alternative matching function that uses a sort of DFA algorithm (not a true  alternative matching function that uses a sort of DFA algorithm (not a true
43  FSM). This is NOT Perl- compatible, but it has advantages in certain  FSM). This is NOT Perl-compatible, but it has advantages in certain
44  applications. */  applications. */
45    
46    
# Line 113  small value. Non-zero values in the tabl Line 112  small value. Non-zero values in the tabl
112  the character is to be found. ***NOTE*** If the start of this table is  the character is to be found. ***NOTE*** If the start of this table is
113  modified, the three tables that follow must also be modified. */  modified, the three tables that follow must also be modified. */
114    
115  static const uschar coptable[] = {  static const pcre_uint8 coptable[] = {
116    0,                             /* End                                    */    0,                             /* End                                    */
117    0, 0, 0, 0, 0,                 /* \A, \G, \K, \B, \b                     */    0, 0, 0, 0, 0,                 /* \A, \G, \K, \B, \b                     */
118    0, 0, 0, 0, 0, 0,              /* \D, \d, \S, \s, \W, \w                 */    0, 0, 0, 0, 0, 0,              /* \D, \d, \S, \s, \W, \w                 */
# Line 121  static const uschar coptable[] = { Line 120  static const uschar coptable[] = {
120    0, 0,                          /* \P, \p                                 */    0, 0,                          /* \P, \p                                 */
121    0, 0, 0, 0, 0,                 /* \R, \H, \h, \V, \v                     */    0, 0, 0, 0, 0,                 /* \R, \H, \h, \V, \v                     */
122    0,                             /* \X                                     */    0,                             /* \X                                     */
123    0, 0, 0, 0, 0,                 /* \Z, \z, Opt, ^, $                      */    0, 0, 0, 0, 0, 0,              /* \Z, \z, ^, ^M, $, $M                   */
124    1,                             /* Char                                   */    1,                             /* Char                                   */
125    1,                             /* Charnc                                 */    1,                             /* Chari                                  */
126    1,                             /* not                                    */    1,                             /* not                                    */
127      1,                             /* noti                                   */
128    /* Positive single-char repeats                                          */    /* Positive single-char repeats                                          */
129    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */
130    3, 3, 3,                       /* upto, minupto, exact                   */    1+IMM2_SIZE, 1+IMM2_SIZE,      /* upto, minupto                          */
131    1, 1, 1, 3,                    /* *+, ++, ?+, upto+                      */    1+IMM2_SIZE,                   /* exact                                  */
132      1, 1, 1, 1+IMM2_SIZE,          /* *+, ++, ?+, upto+                      */
133      1, 1, 1, 1, 1, 1,              /* *I, *?I, +I, +?I, ?I, ??I              */
134      1+IMM2_SIZE, 1+IMM2_SIZE,      /* upto I, minupto I                      */
135      1+IMM2_SIZE,                   /* exact I                                */
136      1, 1, 1, 1+IMM2_SIZE,          /* *+I, ++I, ?+I, upto+I                  */
137    /* Negative single-char repeats - only for chars < 256                   */    /* Negative single-char repeats - only for chars < 256                   */
138    1, 1, 1, 1, 1, 1,              /* NOT *, *?, +, +?, ?, ??                */    1, 1, 1, 1, 1, 1,              /* NOT *, *?, +, +?, ?, ??                */
139    3, 3, 3,                       /* NOT upto, minupto, exact               */    1+IMM2_SIZE, 1+IMM2_SIZE,      /* NOT upto, minupto                      */
140    1, 1, 1, 3,                    /* NOT *+, ++, ?+, updo+                  */    1+IMM2_SIZE,                   /* NOT exact                              */
141      1, 1, 1, 1+IMM2_SIZE,          /* NOT *+, ++, ?+, upto+                  */
142      1, 1, 1, 1, 1, 1,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */
143      1+IMM2_SIZE, 1+IMM2_SIZE,      /* NOT upto I, minupto I                  */
144      1+IMM2_SIZE,                   /* NOT exact I                            */
145      1, 1, 1, 1+IMM2_SIZE,          /* NOT *+I, ++I, ?+I, upto+I              */
146    /* Positive type repeats                                                 */    /* Positive type repeats                                                 */
147    1, 1, 1, 1, 1, 1,              /* Type *, *?, +, +?, ?, ??               */    1, 1, 1, 1, 1, 1,              /* Type *, *?, +, +?, ?, ??               */
148    3, 3, 3,                       /* Type upto, minupto, exact              */    1+IMM2_SIZE, 1+IMM2_SIZE,      /* Type upto, minupto                     */
149    1, 1, 1, 3,                    /* Type *+, ++, ?+, upto+                 */    1+IMM2_SIZE,                   /* Type exact                             */
150      1, 1, 1, 1+IMM2_SIZE,          /* Type *+, ++, ?+, upto+                 */
151    /* Character class & ref repeats                                         */    /* Character class & ref repeats                                         */
152    0, 0, 0, 0, 0, 0,              /* *, *?, +, +?, ?, ??                    */    0, 0, 0, 0, 0, 0,              /* *, *?, +, +?, ?, ??                    */
153    0, 0,                          /* CRRANGE, CRMINRANGE                    */    0, 0,                          /* CRRANGE, CRMINRANGE                    */
# Line 144  static const uschar coptable[] = { Line 155  static const uschar coptable[] = {
155    0,                             /* NCLASS                                 */    0,                             /* NCLASS                                 */
156    0,                             /* XCLASS - variable length               */    0,                             /* XCLASS - variable length               */
157    0,                             /* REF                                    */    0,                             /* REF                                    */
158      0,                             /* REFI                                   */
159    0,                             /* RECURSE                                */    0,                             /* RECURSE                                */
160    0,                             /* CALLOUT                                */    0,                             /* CALLOUT                                */
161    0,                             /* Alt                                    */    0,                             /* Alt                                    */
162    0,                             /* Ket                                    */    0,                             /* Ket                                    */
163    0,                             /* KetRmax                                */    0,                             /* KetRmax                                */
164    0,                             /* KetRmin                                */    0,                             /* KetRmin                                */
165      0,                             /* KetRpos                                */
166      0,                             /* Reverse                                */
167    0,                             /* Assert                                 */    0,                             /* Assert                                 */
168    0,                             /* Assert not                             */    0,                             /* Assert not                             */
169    0,                             /* Assert behind                          */    0,                             /* Assert behind                          */
170    0,                             /* Assert behind not                      */    0,                             /* Assert behind not                      */
171    0,                             /* Reverse                                */    0, 0,                          /* ONCE, ONCE_NC                          */
172    0, 0, 0, 0,                    /* ONCE, BRA, CBRA, COND                  */    0, 0, 0, 0, 0,                 /* BRA, BRAPOS, CBRA, CBRAPOS, COND       */
173    0, 0, 0,                       /* SBRA, SCBRA, SCOND                     */    0, 0, 0, 0, 0,                 /* SBRA, SBRAPOS, SCBRA, SCBRAPOS, SCOND  */
174    0, 0,                          /* CREF, NCREF                            */    0, 0,                          /* CREF, NCREF                            */
175    0, 0,                          /* RREF, NRREF                            */    0, 0,                          /* RREF, NRREF                            */
176    0,                             /* DEF                                    */    0,                             /* DEF                                    */
177    0, 0,                          /* BRAZERO, BRAMINZERO                    */    0, 0, 0,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */
178    0, 0, 0,                       /* MARK, PRUNE, PRUNE_ARG,                */    0, 0, 0,                       /* MARK, PRUNE, PRUNE_ARG                 */
179    0, 0, 0, 0,                    /* SKIP, SKIP_ARG, THEN, THEN_ARG,        */    0, 0, 0, 0,                    /* SKIP, SKIP_ARG, THEN, THEN_ARG         */
180    0, 0, 0, 0, 0                  /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO  */    0, 0, 0, 0,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */
181      0, 0                           /* CLOSE, SKIPZERO  */
182  };  };
183    
184  /* This table identifies those opcodes that inspect a character. It is used to  /* This table identifies those opcodes that inspect a character. It is used to
# Line 171  remember the fact that a character could Line 186  remember the fact that a character could
186  the subject is reached. ***NOTE*** If the start of this table is modified, the  the subject is reached. ***NOTE*** If the start of this table is modified, the
187  two tables that follow must also be modified. */  two tables that follow must also be modified. */
188    
189  static const uschar poptable[] = {  static const pcre_uint8 poptable[] = {
190    0,                             /* End                                    */    0,                             /* End                                    */
191    0, 0, 0, 1, 1,                 /* \A, \G, \K, \B, \b                     */    0, 0, 0, 1, 1,                 /* \A, \G, \K, \B, \b                     */
192    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */
# Line 179  static const uschar poptable[] = { Line 194  static const uschar poptable[] = {
194    1, 1,                          /* \P, \p                                 */    1, 1,                          /* \P, \p                                 */
195    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */
196    1,                             /* \X                                     */    1,                             /* \X                                     */
197    0, 0, 0, 0, 0,                 /* \Z, \z, Opt, ^, $                      */    0, 0, 0, 0, 0, 0,              /* \Z, \z, ^, ^M, $, $M                   */
198    1,                             /* Char                                   */    1,                             /* Char                                   */
199    1,                             /* Charnc                                 */    1,                             /* Chari                                  */
200    1,                             /* not                                    */    1,                             /* not                                    */
201      1,                             /* noti                                   */
202    /* Positive single-char repeats                                          */    /* Positive single-char repeats                                          */
203    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */
204    1, 1, 1,                       /* upto, minupto, exact                   */    1, 1, 1,                       /* upto, minupto, exact                   */
205    1, 1, 1, 1,                    /* *+, ++, ?+, upto+                      */    1, 1, 1, 1,                    /* *+, ++, ?+, upto+                      */
206      1, 1, 1, 1, 1, 1,              /* *I, *?I, +I, +?I, ?I, ??I              */
207      1, 1, 1,                       /* upto I, minupto I, exact I             */
208      1, 1, 1, 1,                    /* *+I, ++I, ?+I, upto+I                  */
209    /* Negative single-char repeats - only for chars < 256                   */    /* Negative single-char repeats - only for chars < 256                   */
210    1, 1, 1, 1, 1, 1,              /* NOT *, *?, +, +?, ?, ??                */    1, 1, 1, 1, 1, 1,              /* NOT *, *?, +, +?, ?, ??                */
211    1, 1, 1,                       /* NOT upto, minupto, exact               */    1, 1, 1,                       /* NOT upto, minupto, exact               */
212    1, 1, 1, 1,                    /* NOT *+, ++, ?+, upto+                  */    1, 1, 1, 1,                    /* NOT *+, ++, ?+, upto+                  */
213      1, 1, 1, 1, 1, 1,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */
214      1, 1, 1,                       /* NOT upto I, minupto I, exact I         */
215      1, 1, 1, 1,                    /* NOT *+I, ++I, ?+I, upto+I              */
216    /* Positive type repeats                                                 */    /* Positive type repeats                                                 */
217    1, 1, 1, 1, 1, 1,              /* Type *, *?, +, +?, ?, ??               */    1, 1, 1, 1, 1, 1,              /* Type *, *?, +, +?, ?, ??               */
218    1, 1, 1,                       /* Type upto, minupto, exact              */    1, 1, 1,                       /* Type upto, minupto, exact              */
# Line 202  static const uschar poptable[] = { Line 224  static const uschar poptable[] = {
224    1,                             /* NCLASS                                 */    1,                             /* NCLASS                                 */
225    1,                             /* XCLASS - variable length               */    1,                             /* XCLASS - variable length               */
226    0,                             /* REF                                    */    0,                             /* REF                                    */
227      0,                             /* REFI                                   */
228    0,                             /* RECURSE                                */    0,                             /* RECURSE                                */
229    0,                             /* CALLOUT                                */    0,                             /* CALLOUT                                */
230    0,                             /* Alt                                    */    0,                             /* Alt                                    */
231    0,                             /* Ket                                    */    0,                             /* Ket                                    */
232    0,                             /* KetRmax                                */    0,                             /* KetRmax                                */
233    0,                             /* KetRmin                                */    0,                             /* KetRmin                                */
234      0,                             /* KetRpos                                */
235      0,                             /* Reverse                                */
236    0,                             /* Assert                                 */    0,                             /* Assert                                 */
237    0,                             /* Assert not                             */    0,                             /* Assert not                             */
238    0,                             /* Assert behind                          */    0,                             /* Assert behind                          */
239    0,                             /* Assert behind not                      */    0,                             /* Assert behind not                      */
240    0,                             /* Reverse                                */    0, 0,                          /* ONCE, ONCE_NC                          */
241    0, 0, 0, 0,                    /* ONCE, BRA, CBRA, COND                  */    0, 0, 0, 0, 0,                 /* BRA, BRAPOS, CBRA, CBRAPOS, COND       */
242    0, 0, 0,                       /* SBRA, SCBRA, SCOND                     */    0, 0, 0, 0, 0,                 /* SBRA, SBRAPOS, SCBRA, SCBRAPOS, SCOND  */
243    0, 0,                          /* CREF, NCREF                            */    0, 0,                          /* CREF, NCREF                            */
244    0, 0,                          /* RREF, NRREF                            */    0, 0,                          /* RREF, NRREF                            */
245    0,                             /* DEF                                    */    0,                             /* DEF                                    */
246    0, 0,                          /* BRAZERO, BRAMINZERO                    */    0, 0, 0,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */
247    0, 0, 0,                       /* MARK, PRUNE, PRUNE_ARG,                */    0, 0, 0,                       /* MARK, PRUNE, PRUNE_ARG                 */
248    0, 0, 0, 0,                    /* SKIP, SKIP_ARG, THEN, THEN_ARG,        */    0, 0, 0, 0,                    /* SKIP, SKIP_ARG, THEN, THEN_ARG         */
249    0, 0, 0, 0, 0                  /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO  */    0, 0, 0, 0,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */
250      0, 0                           /* CLOSE, SKIPZERO                        */
251  };  };
252    
253  /* These 2 tables allow for compact code for testing for \D, \d, \S, \s, \W,  /* These 2 tables allow for compact code for testing for \D, \d, \S, \s, \W,
254  and \w */  and \w */
255    
256  static const uschar toptable1[] = {  static const pcre_uint8 toptable1[] = {
257    0, 0, 0, 0, 0, 0,    0, 0, 0, 0, 0, 0,
258    ctype_digit, ctype_digit,    ctype_digit, ctype_digit,
259    ctype_space, ctype_space,    ctype_space, ctype_space,
# Line 235  static const uschar toptable1[] = { Line 261  static const uschar toptable1[] = {
261    0, 0                            /* OP_ANY, OP_ALLANY */    0, 0                            /* OP_ANY, OP_ALLANY */
262  };  };
263    
264  static const uschar toptable2[] = {  static const pcre_uint8 toptable2[] = {
265    0, 0, 0, 0, 0, 0,    0, 0, 0, 0, 0, 0,
266    ctype_digit, 0,    ctype_digit, 0,
267    ctype_space, 0,    ctype_space, 0,
# Line 252  these structures in, is a vector of ints Line 278  these structures in, is a vector of ints
278  typedef struct stateblock {  typedef struct stateblock {
279    int offset;                     /* Offset to opcode */    int offset;                     /* Offset to opcode */
280    int count;                      /* Count for repeats */    int count;                      /* Count for repeats */
   int ims;                        /* ims flag bits */  
281    int data;                       /* Some use extra data */    int data;                       /* Some use extra data */
282  } stateblock;  } stateblock;
283    
284  #define INTS_PER_STATEBLOCK  (sizeof(stateblock)/sizeof(int))  #define INTS_PER_STATEBLOCK  (int)(sizeof(stateblock)/sizeof(int))
285    
286    
287  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
# Line 275  Returns:       nothing Line 300  Returns:       nothing
300  */  */
301    
302  static void  static void
303  pchars(unsigned char *p, int length, FILE *f)  pchars(const pcre_uchar *p, int length, FILE *f)
304  {  {
305  int c;  pcre_uint32 c;
306  while (length-- > 0)  while (length-- > 0)
307    {    {
308    if (isprint(c = *(p++)))    if (isprint(c = *(p++)))
309      fprintf(f, "%c", c);      fprintf(f, "%c", c);
310    else    else
311      fprintf(f, "\\x%02x", c);      fprintf(f, "\\x{%02x}", c);
312    }    }
313  }  }
314  #endif  #endif
# Line 308  Arguments: Line 333  Arguments:
333    offsetcount       size of same    offsetcount       size of same
334    workspace         vector of workspace    workspace         vector of workspace
335    wscount           size of same    wscount           size of same
   ims               the current ims flags  
336    rlevel            function call recursion level    rlevel            function call recursion level
   recursing         regex recursive call level  
337    
338  Returns:            > 0 => number of match offset pairs placed in offsets  Returns:            > 0 => number of match offset pairs placed in offsets
339                      = 0 => offsets overflowed; longest matches are present                      = 0 => offsets overflowed; longest matches are present
# Line 325  for the current character, one for the f Line 348  for the current character, one for the f
348      { \      { \
349      next_active_state->offset = (x); \      next_active_state->offset = (x); \
350      next_active_state->count  = (y); \      next_active_state->count  = (y); \
     next_active_state->ims    = ims; \  
351      next_active_state++; \      next_active_state++; \
352      DPRINTF(("%.*sADD_ACTIVE(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \      DPRINTF(("%.*sADD_ACTIVE(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \
353      } \      } \
# Line 336  for the current character, one for the f Line 358  for the current character, one for the f
358      { \      { \
359      next_active_state->offset = (x); \      next_active_state->offset = (x); \
360      next_active_state->count  = (y); \      next_active_state->count  = (y); \
     next_active_state->ims    = ims; \  
361      next_active_state->data   = (z); \      next_active_state->data   = (z); \
362      next_active_state++; \      next_active_state++; \
363      DPRINTF(("%.*sADD_ACTIVE_DATA(%d,%d,%d)\n", rlevel*2-2, SP, (x), (y), (z))); \      DPRINTF(("%.*sADD_ACTIVE_DATA(%d,%d,%d)\n", rlevel*2-2, SP, (x), (y), (z))); \
# Line 348  for the current character, one for the f Line 369  for the current character, one for the f
369      { \      { \
370      next_new_state->offset = (x); \      next_new_state->offset = (x); \
371      next_new_state->count  = (y); \      next_new_state->count  = (y); \
     next_new_state->ims    = ims; \  
372      next_new_state++; \      next_new_state++; \
373      DPRINTF(("%.*sADD_NEW(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \      DPRINTF(("%.*sADD_NEW(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \
374      } \      } \
# Line 359  for the current character, one for the f Line 379  for the current character, one for the f
379      { \      { \
380      next_new_state->offset = (x); \      next_new_state->offset = (x); \
381      next_new_state->count  = (y); \      next_new_state->count  = (y); \
     next_new_state->ims    = ims; \  
382      next_new_state->data   = (z); \      next_new_state->data   = (z); \
383      next_new_state++; \      next_new_state++; \
384      DPRINTF(("%.*sADD_NEW_DATA(%d,%d,%d)\n", rlevel*2-2, SP, (x), (y), (z))); \      DPRINTF(("%.*sADD_NEW_DATA(%d,%d,%d) line %d\n", rlevel*2-2, SP, \
385          (x), (y), (z), __LINE__)); \
386      } \      } \
387    else return PCRE_ERROR_DFA_WSSIZE    else return PCRE_ERROR_DFA_WSSIZE
388    
# Line 371  for the current character, one for the f Line 391  for the current character, one for the f
391  static int  static int
392  internal_dfa_exec(  internal_dfa_exec(
393    dfa_match_data *md,    dfa_match_data *md,
394    const uschar *this_start_code,    const pcre_uchar *this_start_code,
395    const uschar *current_subject,    const pcre_uchar *current_subject,
396    int start_offset,    int start_offset,
397    int *offsets,    int *offsets,
398    int offsetcount,    int offsetcount,
399    int *workspace,    int *workspace,
400    int wscount,    int wscount,
401    int ims,    int  rlevel)
   int  rlevel,  
   int  recursing)  
402  {  {
403  stateblock *active_states, *new_states, *temp_states;  stateblock *active_states, *new_states, *temp_states;
404  stateblock *next_active_state, *next_new_state;  stateblock *next_active_state, *next_new_state;
405    
406  const uschar *ctypes, *lcc, *fcc;  const pcre_uint8 *ctypes, *lcc, *fcc;
407  const uschar *ptr;  const pcre_uchar *ptr;
408  const uschar *end_code, *first_op;  const pcre_uchar *end_code, *first_op;
409    
410    dfa_recursion_info new_recursive;
411    
412  int active_count, new_count, match_count;  int active_count, new_count, match_count;
413    
414  /* Some fields in the md block are frequently referenced, so we load them into  /* Some fields in the md block are frequently referenced, so we load them into
415  independent variables in the hope that this will perform better. */  independent variables in the hope that this will perform better. */
416    
417  const uschar *start_subject = md->start_subject;  const pcre_uchar *start_subject = md->start_subject;
418  const uschar *end_subject = md->end_subject;  const pcre_uchar *end_subject = md->end_subject;
419  const uschar *start_code = md->start_code;  const pcre_uchar *start_code = md->start_code;
420    
421  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
422  BOOL utf8 = (md->poptions & PCRE_UTF8) != 0;  BOOL utf = (md->poptions & PCRE_UTF8) != 0;
423  #else  #else
424  BOOL utf8 = FALSE;  BOOL utf = FALSE;
425  #endif  #endif
426    
427    BOOL reset_could_continue = FALSE;
428    
429  rlevel++;  rlevel++;
430  offsetcount &= (-2);  offsetcount &= (-2);
431    
# Line 412  wscount = (wscount - (wscount % (INTS_PE Line 434  wscount = (wscount - (wscount % (INTS_PE
434            (2 * INTS_PER_STATEBLOCK);            (2 * INTS_PER_STATEBLOCK);
435    
436  DPRINTF(("\n%.*s---------------------\n"  DPRINTF(("\n%.*s---------------------\n"
437    "%.*sCall to internal_dfa_exec f=%d r=%d\n",    "%.*sCall to internal_dfa_exec f=%d\n",
438    rlevel*2-2, SP, rlevel*2-2, SP, rlevel, recursing));    rlevel*2-2, SP, rlevel*2-2, SP, rlevel));
439    
440  ctypes = md->tables + ctypes_offset;  ctypes = md->tables + ctypes_offset;
441  lcc = md->tables + lcc_offset;  lcc = md->tables + lcc_offset;
# Line 426  next_new_state = new_states = active_sta Line 448  next_new_state = new_states = active_sta
448  new_count = 0;  new_count = 0;
449    
450  first_op = this_start_code + 1 + LINK_SIZE +  first_op = this_start_code + 1 + LINK_SIZE +
451    ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA)? 2:0);    ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA ||
452        *this_start_code == OP_CBRAPOS || *this_start_code == OP_SCBRAPOS)
453        ? IMM2_SIZE:0);
454    
455  /* The first thing in any (sub) pattern is a bracket of some sort. Push all  /* The first thing in any (sub) pattern is a bracket of some sort. Push all
456  the alternative states onto the list, and find out where the end is. This  the alternative states onto the list, and find out where the end is. This
# Line 454  if (*first_op == OP_REVERSE) Line 478  if (*first_op == OP_REVERSE)
478    /* If we can't go back the amount required for the longest lookbehind    /* If we can't go back the amount required for the longest lookbehind
479    pattern, go back as far as we can; some alternatives may still be viable. */    pattern, go back as far as we can; some alternatives may still be viable. */
480    
481  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
482    /* In character mode we have to step back character by character */    /* In character mode we have to step back character by character */
483    
484    if (utf8)    if (utf)
485      {      {
486      for (gone_back = 0; gone_back < max_back; gone_back++)      for (gone_back = 0; gone_back < max_back; gone_back++)
487        {        {
488        if (current_subject <= start_subject) break;        if (current_subject <= start_subject) break;
489        current_subject--;        current_subject--;
490        while (current_subject > start_subject &&        ACROSSCHAR(current_subject > start_subject, *current_subject, current_subject--);
              (*current_subject & 0xc0) == 0x80)  
         current_subject--;  
491        }        }
492      }      }
493    else    else
# Line 525  else Line 547  else
547    else    else
548      {      {
549      int length = 1 + LINK_SIZE +      int length = 1 + LINK_SIZE +
550        ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA)? 2:0);        ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA ||
551            *this_start_code == OP_CBRAPOS || *this_start_code == OP_SCBRAPOS)
552            ? IMM2_SIZE:0);
553      do      do
554        {        {
555        ADD_NEW((int)(end_code - start_code + length), 0);        ADD_NEW((int)(end_code - start_code + length), 0);
# Line 538  else Line 562  else
562    
563  workspace[0] = 0;    /* Bit indicating which vector is current */  workspace[0] = 0;    /* Bit indicating which vector is current */
564    
565  DPRINTF(("%.*sEnd state = %d\n", rlevel*2-2, SP, end_code - start_code));  DPRINTF(("%.*sEnd state = %d\n", rlevel*2-2, SP, (int)(end_code - start_code)));
566    
567  /* Loop for scanning the subject */  /* Loop for scanning the subject */
568    
# Line 547  for (;;) Line 571  for (;;)
571    {    {
572    int i, j;    int i, j;
573    int clen, dlen;    int clen, dlen;
574    unsigned int c, d;    pcre_uint32 c, d;
575    int forced_fail = 0;    int forced_fail = 0;
576    BOOL could_continue = FALSE;    BOOL partial_newline = FALSE;
577      BOOL could_continue = reset_could_continue;
578      reset_could_continue = FALSE;
579    
580    /* Make the new state list into the active state list and empty the    /* Make the new state list into the active state list and empty the
581    new state list. */    new state list. */
# Line 565  for (;;) Line 591  for (;;)
591    
592  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
593    printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP);    printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP);
594    pchars((uschar *)ptr, strlen((char *)ptr), stdout);    pchars(ptr, STRLEN_UC(ptr), stdout);
595    printf("\"\n");    printf("\"\n");
596    
597    printf("%.*sActive states: ", rlevel*2-2, SP);    printf("%.*sActive states: ", rlevel*2-2, SP);
# Line 585  for (;;) Line 611  for (;;)
611    
612    if (ptr < end_subject)    if (ptr < end_subject)
613      {      {
614      clen = 1;        /* Number of bytes in the character */      clen = 1;        /* Number of data items in the character */
615  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
616      if (utf8) { GETCHARLEN(c, ptr, clen); } else      GETCHARLENTEST(c, ptr, clen);
617  #endif  /* SUPPORT_UTF8 */  #else
618      c = *ptr;      c = *ptr;
619    #endif  /* SUPPORT_UTF */
620      }      }
621    else    else
622      {      {
# Line 605  for (;;) Line 632  for (;;)
632    for (i = 0; i < active_count; i++)    for (i = 0; i < active_count; i++)
633      {      {
634      stateblock *current_state = active_states + i;      stateblock *current_state = active_states + i;
635      const uschar *code;      BOOL caseless = FALSE;
636        const pcre_uchar *code;
637      int state_offset = current_state->offset;      int state_offset = current_state->offset;
638      int count, codevalue, rrc;      int codevalue, rrc;
639        unsigned int count;
640    
641  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
642      printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset);      printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset);
# Line 616  for (;;) Line 645  for (;;)
645          else printf("0x%02x\n", c);          else printf("0x%02x\n", c);
646  #endif  #endif
647    
     /* This variable is referred to implicity in the ADD_xxx macros. */  
   
     ims = current_state->ims;  
   
648      /* A negative offset is a special case meaning "hold off going to this      /* A negative offset is a special case meaning "hold off going to this
649      (negated) state until the number of characters in the data field have      (negated) state until the number of characters in the data field have
650      been skipped". */      been skipped". If the could_continue flag was passed over from a previous
651        state, arrange for it to passed on. */
652    
653      if (state_offset < 0)      if (state_offset < 0)
654        {        {
# Line 631  for (;;) Line 657  for (;;)
657          DPRINTF(("%.*sSkipping this character\n", rlevel*2-2, SP));          DPRINTF(("%.*sSkipping this character\n", rlevel*2-2, SP));
658          ADD_NEW_DATA(state_offset, current_state->count,          ADD_NEW_DATA(state_offset, current_state->count,
659            current_state->data - 1);            current_state->data - 1);
660            if (could_continue) reset_could_continue = TRUE;
661          continue;          continue;
662          }          }
663        else        else
# Line 670  for (;;) Line 697  for (;;)
697      permitted.      permitted.
698    
699      We also use this mechanism for opcodes such as OP_TYPEPLUS that take an      We also use this mechanism for opcodes such as OP_TYPEPLUS that take an
700      argument that is not a data character - but is always one byte long. We      argument that is not a data character - but is always one byte long because
701      have to take special action to deal with  \P, \p, \H, \h, \V, \v and \X in      the values are small. We have to take special action to deal with  \P, \p,
702      this case. To keep the other cases fast, convert these ones to new opcodes.      \H, \h, \V, \v and \X in this case. To keep the other cases fast, convert
703      */      these ones to new opcodes. */
704    
705      if (coptable[codevalue] > 0)      if (coptable[codevalue] > 0)
706        {        {
707        dlen = 1;        dlen = 1;
708  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
709        if (utf8) { GETCHARLEN(d, (code + coptable[codevalue]), dlen); } else        if (utf) { GETCHARLEN(d, (code + coptable[codevalue]), dlen); } else
710  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
711        d = code[coptable[codevalue]];        d = code[coptable[codevalue]];
712        if (codevalue >= OP_TYPESTAR)        if (codevalue >= OP_TYPESTAR)
713          {          {
# Line 725  for (;;) Line 752  for (;;)
752    
753  /* ========================================================================== */  /* ========================================================================== */
754        /* Reached a closing bracket. If not at the end of the pattern, carry        /* Reached a closing bracket. If not at the end of the pattern, carry
755        on with the next opcode. Otherwise, unless we have an empty string and        on with the next opcode. For repeating opcodes, also add the repeat
756          state. Note that KETRPOS will always be encountered at the end of the
757          subpattern, because the possessive subpattern repeats are always handled
758          using recursive calls. Thus, it never adds any new states.
759    
760          At the end of the (sub)pattern, unless we have an empty string and
761        PCRE_NOTEMPTY is set, or PCRE_NOTEMPTY_ATSTART is set and we are at the        PCRE_NOTEMPTY is set, or PCRE_NOTEMPTY_ATSTART is set and we are at the
762        start of the subject, save the match data, shifting up all previous        start of the subject, save the match data, shifting up all previous
763        matches so we always have the longest first. */        matches so we always have the longest first. */
# Line 733  for (;;) Line 765  for (;;)
765        case OP_KET:        case OP_KET:
766        case OP_KETRMIN:        case OP_KETRMIN:
767        case OP_KETRMAX:        case OP_KETRMAX:
768          case OP_KETRPOS:
769        if (code != end_code)        if (code != end_code)
770          {          {
771          ADD_ACTIVE(state_offset + 1 + LINK_SIZE, 0);          ADD_ACTIVE(state_offset + 1 + LINK_SIZE, 0);
# Line 749  for (;;) Line 782  for (;;)
782                  current_subject > start_subject + md->start_offset)))                  current_subject > start_subject + md->start_offset)))
783            {            {
784            if (match_count < 0) match_count = (offsetcount >= 2)? 1 : 0;            if (match_count < 0) match_count = (offsetcount >= 2)? 1 : 0;
785              else if (match_count > 0 && ++match_count * 2 >= offsetcount)              else if (match_count > 0 && ++match_count * 2 > offsetcount)
786                match_count = 0;                match_count = 0;
787            count = ((match_count == 0)? offsetcount : match_count * 2) - 2;            count = ((match_count == 0)? offsetcount : match_count * 2) - 2;
788            if (count > 0) memmove(offsets + 2, offsets, count * sizeof(int));            if (count > 0) memmove(offsets + 2, offsets, count * sizeof(int));
# Line 758  for (;;) Line 791  for (;;)
791              offsets[0] = (int)(current_subject - start_subject);              offsets[0] = (int)(current_subject - start_subject);
792              offsets[1] = (int)(ptr - start_subject);              offsets[1] = (int)(ptr - start_subject);
793              DPRINTF(("%.*sSet matched string = \"%.*s\"\n", rlevel*2-2, SP,              DPRINTF(("%.*sSet matched string = \"%.*s\"\n", rlevel*2-2, SP,
794                offsets[1] - offsets[0], current_subject));                offsets[1] - offsets[0], (char *)current_subject));
795              }              }
796            if ((md->moptions & PCRE_DFA_SHORTEST) != 0)            if ((md->moptions & PCRE_DFA_SHORTEST) != 0)
797              {              {
# Line 795  for (;;) Line 828  for (;;)
828        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
829        case OP_CBRA:        case OP_CBRA:
830        case OP_SCBRA:        case OP_SCBRA:
831        ADD_ACTIVE((int)(code - start_code + 3 + LINK_SIZE),  0);        ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE + IMM2_SIZE),  0);
832        code += GET(code, 1);        code += GET(code, 1);
833        while (*code == OP_ALT)        while (*code == OP_ALT)
834          {          {
# Line 822  for (;;) Line 855  for (;;)
855    
856        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
857        case OP_CIRC:        case OP_CIRC:
858        if ((ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) ||        if (ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0)
           ((ims & PCRE_MULTILINE) != 0 &&  
             ptr != end_subject &&  
             WAS_NEWLINE(ptr)))  
859          { ADD_ACTIVE(state_offset + 1, 0); }          { ADD_ACTIVE(state_offset + 1, 0); }
860        break;        break;
861    
862        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
863        case OP_EOD:        case OP_CIRCM:
864        if (ptr >= end_subject) { ADD_ACTIVE(state_offset + 1, 0); }        if ((ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) ||
865              (ptr != end_subject && WAS_NEWLINE(ptr)))
866            { ADD_ACTIVE(state_offset + 1, 0); }
867        break;        break;
868    
869        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
870        case OP_OPT:        case OP_EOD:
871        ims = code[1];        if (ptr >= end_subject)
872        ADD_ACTIVE(state_offset + 2, 0);          {
873            if ((md->moptions & PCRE_PARTIAL_HARD) != 0)
874              could_continue = TRUE;
875            else { ADD_ACTIVE(state_offset + 1, 0); }
876            }
877        break;        break;
878    
879        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
# Line 860  for (;;) Line 896  for (;;)
896        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
897        case OP_ANY:        case OP_ANY:
898        if (clen > 0 && !IS_NEWLINE(ptr))        if (clen > 0 && !IS_NEWLINE(ptr))
899          { ADD_NEW(state_offset + 1, 0); }          {
900            if (ptr + 1 >= md->end_subject &&
901                (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
902                NLBLOCK->nltype == NLTYPE_FIXED &&
903                NLBLOCK->nllen == 2 &&
904                c == NLBLOCK->nl[0])
905              {
906              could_continue = partial_newline = TRUE;
907              }
908            else
909              {
910              ADD_NEW(state_offset + 1, 0);
911              }
912            }
913        break;        break;
914    
915        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
# Line 871  for (;;) Line 920  for (;;)
920    
921        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
922        case OP_EODN:        case OP_EODN:
923        if (clen == 0 || (IS_NEWLINE(ptr) && ptr == end_subject - md->nllen))        if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0)
924            could_continue = TRUE;
925          else if (clen == 0 || (IS_NEWLINE(ptr) && ptr == end_subject - md->nllen))
926          { ADD_ACTIVE(state_offset + 1, 0); }          { ADD_ACTIVE(state_offset + 1, 0); }
927        break;        break;
928    
# Line 879  for (;;) Line 930  for (;;)
930        case OP_DOLL:        case OP_DOLL:
931        if ((md->moptions & PCRE_NOTEOL) == 0)        if ((md->moptions & PCRE_NOTEOL) == 0)
932          {          {
933          if (clen == 0 ||          if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0)
934              could_continue = TRUE;
935            else if (clen == 0 ||
936              ((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr) &&              ((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr) &&
937                 ((ims & PCRE_MULTILINE) != 0 || ptr == end_subject - md->nllen)                 (ptr == end_subject - md->nllen)
938              ))              ))
939            { ADD_ACTIVE(state_offset + 1, 0); }            { ADD_ACTIVE(state_offset + 1, 0); }
940            else if (ptr + 1 >= md->end_subject &&
941                     (md->moptions & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) != 0 &&
942                     NLBLOCK->nltype == NLTYPE_FIXED &&
943                     NLBLOCK->nllen == 2 &&
944                     c == NLBLOCK->nl[0])
945              {
946              if ((md->moptions & PCRE_PARTIAL_HARD) != 0)
947                {
948                reset_could_continue = TRUE;
949                ADD_NEW_DATA(-(state_offset + 1), 0, 1);
950                }
951              else could_continue = partial_newline = TRUE;
952              }
953            }
954          break;
955    
956          /*-----------------------------------------------------------------*/
957          case OP_DOLLM:
958          if ((md->moptions & PCRE_NOTEOL) == 0)
959            {
960            if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0)
961              could_continue = TRUE;
962            else if (clen == 0 ||
963                ((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr)))
964              { ADD_ACTIVE(state_offset + 1, 0); }
965            else if (ptr + 1 >= md->end_subject &&
966                     (md->moptions & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) != 0 &&
967                     NLBLOCK->nltype == NLTYPE_FIXED &&
968                     NLBLOCK->nllen == 2 &&
969                     c == NLBLOCK->nl[0])
970              {
971              if ((md->moptions & PCRE_PARTIAL_HARD) != 0)
972                {
973                reset_could_continue = TRUE;
974                ADD_NEW_DATA(-(state_offset + 1), 0, 1);
975                }
976              else could_continue = partial_newline = TRUE;
977              }
978          }          }
979        else if ((ims & PCRE_MULTILINE) != 0 && IS_NEWLINE(ptr))        else if (IS_NEWLINE(ptr))
980          { ADD_ACTIVE(state_offset + 1, 0); }          { ADD_ACTIVE(state_offset + 1, 0); }
981        break;        break;
982    
# Line 916  for (;;) Line 1007  for (;;)
1007    
1008          if (ptr > start_subject)          if (ptr > start_subject)
1009            {            {
1010            const uschar *temp = ptr - 1;            const pcre_uchar *temp = ptr - 1;
1011            if (temp < md->start_used_ptr) md->start_used_ptr = temp;            if (temp < md->start_used_ptr) md->start_used_ptr = temp;
1012  #ifdef SUPPORT_UTF8  #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
1013            if (utf8) BACKCHAR(temp);            if (utf) { BACKCHAR(temp); }
1014  #endif  #endif
1015            GETCHARTEST(d, temp);            GETCHARTEST(d, temp);
1016  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
1017            if ((md->poptions & PCRE_UCP) != 0)            if ((md->poptions & PCRE_UCP) != 0)
1018              {              {
1019              if (d == '_') left_word = TRUE; else              if (d == '_') left_word = TRUE; else
1020                {                {
1021                int cat = UCD_CATEGORY(d);                int cat = UCD_CATEGORY(d);
1022                left_word = (cat == ucp_L || cat == ucp_N);                left_word = (cat == ucp_L || cat == ucp_N);
1023                }                }
1024              }              }
1025            else            else
1026  #endif  #endif
1027            left_word = d < 256 && (ctypes[d] & ctype_word) != 0;            left_word = d < 256 && (ctypes[d] & ctype_word) != 0;
1028            }            }
1029          else left_word = FALSE;          else left_word = FALSE;
1030    
1031          if (clen > 0)          if (clen > 0)
1032            {            {
1033  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
1034            if ((md->poptions & PCRE_UCP) != 0)            if ((md->poptions & PCRE_UCP) != 0)
1035              {              {
1036              if (c == '_') right_word = TRUE; else              if (c == '_') right_word = TRUE; else
1037                {                {
1038                int cat = UCD_CATEGORY(c);                int cat = UCD_CATEGORY(c);
1039                right_word = (cat == ucp_L || cat == ucp_N);                right_word = (cat == ucp_L || cat == ucp_N);
1040                }                }
1041              }              }
1042            else            else
1043  #endif  #endif
1044            right_word = c < 256 && (ctypes[c] & ctype_word) != 0;            right_word = c < 256 && (ctypes[c] & ctype_word) != 0;
1045            }            }
1046          else right_word = FALSE;          else right_word = FALSE;
1047    
1048          if ((left_word == right_word) == (codevalue == OP_NOT_WORD_BOUNDARY))          if ((left_word == right_word) == (codevalue == OP_NOT_WORD_BOUNDARY))
# Line 971  for (;;) Line 1062  for (;;)
1062        if (clen > 0)        if (clen > 0)
1063          {          {
1064          BOOL OK;          BOOL OK;
1065            const pcre_uint32 *cp;
1066          const ucd_record * prop = GET_UCD(c);          const ucd_record * prop = GET_UCD(c);
1067          switch(code[1])          switch(code[1])
1068            {            {
# Line 979  for (;;) Line 1071  for (;;)
1071            break;            break;
1072    
1073            case PT_LAMP:            case PT_LAMP:
1074            OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||            OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
1075                 prop->chartype == ucp_Lt;                 prop->chartype == ucp_Lt;
1076            break;            break;
1077    
1078            case PT_GC:            case PT_GC:
1079            OK = _pcre_ucp_gentype[prop->chartype] == code[2];            OK = PRIV(ucp_gentype)[prop->chartype] == code[2];
1080            break;            break;
1081    
1082            case PT_PC:            case PT_PC:
# Line 994  for (;;) Line 1086  for (;;)
1086            case PT_SC:            case PT_SC:
1087            OK = prop->script == code[2];            OK = prop->script == code[2];
1088            break;            break;
1089    
1090            /* These are specials for combination cases. */            /* These are specials for combination cases. */
1091    
1092            case PT_ALNUM:            case PT_ALNUM:
1093            OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
1094                 _pcre_ucp_gentype[prop->chartype] == ucp_N;                 PRIV(ucp_gentype)[prop->chartype] == ucp_N;
1095            break;            break;
1096    
1097            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
1098            OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
1099                 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;                 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
1100            break;            break;
1101    
1102            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
1103            OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
1104                 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||                 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
1105                 c == CHAR_FF || c == CHAR_CR;                 c == CHAR_FF || c == CHAR_CR;
1106            break;            break;
1107    
1108            case PT_WORD:            case PT_WORD:
1109            OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
1110                 _pcre_ucp_gentype[prop->chartype] == ucp_N ||                 PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
1111                 c == CHAR_UNDERSCORE;                 c == CHAR_UNDERSCORE;
1112            break;            break;
1113    
1114              case PT_CLIST:
1115              cp = PRIV(ucd_caseless_sets) + code[2];
1116              for (;;)
1117                {
1118                if (c < *cp) { OK = FALSE; break; }
1119                if (c == *cp++) { OK = TRUE; break; }
1120                }
1121              break;
1122    
1123              case PT_UCNC:
1124              OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
1125                   c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
1126                   c >= 0xe000;
1127              break;
1128    
1129            /* Should never occur, but keep compilers from grumbling. */            /* Should never occur, but keep compilers from grumbling. */
1130    
# Line 1046  for (;;) Line 1153  for (;;)
1153        if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }        if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }
1154        if (clen > 0)        if (clen > 0)
1155          {          {
1156          if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||          if (d == OP_ANY && ptr + 1 >= md->end_subject &&
1157                (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
1158                NLBLOCK->nltype == NLTYPE_FIXED &&
1159                NLBLOCK->nllen == 2 &&
1160                c == NLBLOCK->nl[0])
1161              {
1162              could_continue = partial_newline = TRUE;
1163              }
1164            else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
1165              (c < 256 &&              (c < 256 &&
1166                (d != OP_ANY || !IS_NEWLINE(ptr)) &&                (d != OP_ANY || !IS_NEWLINE(ptr)) &&
1167                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
# Line 1069  for (;;) Line 1184  for (;;)
1184        ADD_ACTIVE(state_offset + 2, 0);        ADD_ACTIVE(state_offset + 2, 0);
1185        if (clen > 0)        if (clen > 0)
1186          {          {
1187          if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||          if (d == OP_ANY && ptr + 1 >= md->end_subject &&
1188                (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
1189                NLBLOCK->nltype == NLTYPE_FIXED &&
1190                NLBLOCK->nllen == 2 &&
1191                c == NLBLOCK->nl[0])
1192              {
1193              could_continue = partial_newline = TRUE;
1194              }
1195            else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
1196              (c < 256 &&              (c < 256 &&
1197                (d != OP_ANY || !IS_NEWLINE(ptr)) &&                (d != OP_ANY || !IS_NEWLINE(ptr)) &&
1198                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
# Line 1091  for (;;) Line 1214  for (;;)
1214        ADD_ACTIVE(state_offset + 2, 0);        ADD_ACTIVE(state_offset + 2, 0);
1215        if (clen > 0)        if (clen > 0)
1216          {          {
1217          if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||          if (d == OP_ANY && ptr + 1 >= md->end_subject &&
1218                (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
1219                NLBLOCK->nltype == NLTYPE_FIXED &&
1220                NLBLOCK->nllen == 2 &&
1221                c == NLBLOCK->nl[0])
1222              {
1223              could_continue = partial_newline = TRUE;
1224              }
1225            else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
1226              (c < 256 &&              (c < 256 &&
1227                (d != OP_ANY || !IS_NEWLINE(ptr)) &&                (d != OP_ANY || !IS_NEWLINE(ptr)) &&
1228                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
# Line 1111  for (;;) Line 1242  for (;;)
1242        count = current_state->count;  /* Number already matched */        count = current_state->count;  /* Number already matched */
1243        if (clen > 0)        if (clen > 0)
1244          {          {
1245          if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||          if (d == OP_ANY && ptr + 1 >= md->end_subject &&
1246                (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
1247                NLBLOCK->nltype == NLTYPE_FIXED &&
1248                NLBLOCK->nllen == 2 &&
1249                c == NLBLOCK->nl[0])
1250              {
1251              could_continue = partial_newline = TRUE;
1252              }
1253            else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
1254              (c < 256 &&              (c < 256 &&
1255                (d != OP_ANY || !IS_NEWLINE(ptr)) &&                (d != OP_ANY || !IS_NEWLINE(ptr)) &&
1256                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
1257            {            {
1258            if (++count >= GET2(code, 1))            if (++count >= GET2(code, 1))
1259              { ADD_NEW(state_offset + 4, 0); }              { ADD_NEW(state_offset + 1 + IMM2_SIZE + 1, 0); }
1260            else            else
1261              { ADD_NEW(state_offset, count); }              { ADD_NEW(state_offset, count); }
1262            }            }
# Line 1128  for (;;) Line 1267  for (;;)
1267        case OP_TYPEUPTO:        case OP_TYPEUPTO:
1268        case OP_TYPEMINUPTO:        case OP_TYPEMINUPTO:
1269        case OP_TYPEPOSUPTO:        case OP_TYPEPOSUPTO:
1270        ADD_ACTIVE(state_offset + 4, 0);        ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0);
1271        count = current_state->count;  /* Number already matched */        count = current_state->count;  /* Number already matched */
1272        if (clen > 0)        if (clen > 0)
1273          {          {
1274          if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||          if (d == OP_ANY && ptr + 1 >= md->end_subject &&
1275                (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
1276                NLBLOCK->nltype == NLTYPE_FIXED &&
1277                NLBLOCK->nllen == 2 &&
1278                c == NLBLOCK->nl[0])
1279              {
1280              could_continue = partial_newline = TRUE;
1281              }
1282            else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
1283              (c < 256 &&              (c < 256 &&
1284                (d != OP_ANY || !IS_NEWLINE(ptr)) &&                (d != OP_ANY || !IS_NEWLINE(ptr)) &&
1285                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))                ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
# Line 1143  for (;;) Line 1290  for (;;)
1290              next_active_state--;              next_active_state--;
1291              }              }
1292            if (++count >= GET2(code, 1))            if (++count >= GET2(code, 1))
1293              { ADD_NEW(state_offset + 4, 0); }              { ADD_NEW(state_offset + 2 + IMM2_SIZE, 0); }
1294            else            else
1295              { ADD_NEW(state_offset, count); }              { ADD_NEW(state_offset, count); }
1296            }            }
# Line 1165  for (;;) Line 1312  for (;;)
1312        if (clen > 0)        if (clen > 0)
1313          {          {
1314          BOOL OK;          BOOL OK;
1315            const pcre_uint32 *cp;
1316          const ucd_record * prop = GET_UCD(c);          const ucd_record * prop = GET_UCD(c);
1317          switch(code[2])          switch(code[2])
1318            {            {
# Line 1173  for (;;) Line 1321  for (;;)
1321            break;            break;
1322    
1323            case PT_LAMP:            case PT_LAMP:
1324            OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||            OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
1325              prop->chartype == ucp_Lt;              prop->chartype == ucp_Lt;
1326            break;            break;
1327    
1328            case PT_GC:            case PT_GC:
1329            OK = _pcre_ucp_gentype[prop->chartype] == code[3];            OK = PRIV(ucp_gentype)[prop->chartype] == code[3];
1330            break;            break;
1331    
1332            case PT_PC:            case PT_PC:
# Line 1190  for (;;) Line 1338  for (;;)
1338            break;            break;
1339    
1340            /* These are specials for combination cases. */            /* These are specials for combination cases. */
1341    
1342            case PT_ALNUM:            case PT_ALNUM:
1343            OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
1344                 _pcre_ucp_gentype[prop->chartype] == ucp_N;                 PRIV(ucp_gentype)[prop->chartype] == ucp_N;
1345            break;            break;
1346    
1347            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
1348            OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
1349                 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;                 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
1350            break;            break;
1351    
1352            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
1353            OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
1354                 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||                 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
1355                 c == CHAR_FF || c == CHAR_CR;                 c == CHAR_FF || c == CHAR_CR;
1356            break;            break;
1357    
1358            case PT_WORD:            case PT_WORD:
1359            OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
1360                 _pcre_ucp_gentype[prop->chartype] == ucp_N ||                 PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
1361                 c == CHAR_UNDERSCORE;                 c == CHAR_UNDERSCORE;
1362            break;            break;
1363    
1364              case PT_CLIST:
1365              cp = PRIV(ucd_caseless_sets) + code[3];
1366              for (;;)
1367                {
1368                if (c < *cp) { OK = FALSE; break; }
1369                if (c == *cp++) { OK = TRUE; break; }
1370                }
1371              break;
1372    
1373              case PT_UCNC:
1374              OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
1375                   c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
1376                   c >= 0xe000;
1377              break;
1378    
1379            /* Should never occur, but keep compilers from grumbling. */            /* Should never occur, but keep compilers from grumbling. */
1380    
# Line 1239  for (;;) Line 1402  for (;;)
1402        case OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS:        case OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS:
1403        count = current_state->count;  /* Already matched */        count = current_state->count;  /* Already matched */
1404        if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }        if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }
1405        if (clen > 0 && UCD_CATEGORY(c) != ucp_M)        if (clen > 0)
1406          {          {
1407          const uschar *nptr = ptr + clen;          int lgb, rgb;
1408            const pcre_uchar *nptr = ptr + clen;
1409          int ncount = 0;          int ncount = 0;
1410          if (count > 0 && codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS)          if (count > 0 && codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS)
1411            {            {
1412            active_count--;           /* Remove non-match possibility */            active_count--;           /* Remove non-match possibility */
1413            next_active_state--;            next_active_state--;
1414            }            }
1415            lgb = UCD_GRAPHBREAK(c);
1416          while (nptr < end_subject)          while (nptr < end_subject)
1417            {            {
1418            int nd;            dlen = 1;
1419            int ndlen = 1;            if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); }
1420            GETCHARLEN(nd, nptr, ndlen);            rgb = UCD_GRAPHBREAK(d);
1421            if (UCD_CATEGORY(nd) != ucp_M) break;            if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
1422            ncount++;            ncount++;
1423            nptr += ndlen;            lgb = rgb;
1424              nptr += dlen;
1425            }            }
1426          count++;          count++;
1427          ADD_NEW_DATA(-state_offset, count, ncount);          ADD_NEW_DATA(-state_offset, count, ncount);
# Line 1274  for (;;) Line 1440  for (;;)
1440          int ncount = 0;          int ncount = 0;
1441          switch (c)          switch (c)
1442            {            {
1443            case 0x000b:            case CHAR_VT:
1444            case 0x000c:            case CHAR_FF:
1445            case 0x0085:            case CHAR_NEL:
1446    #ifndef EBCDIC
1447            case 0x2028:            case 0x2028:
1448            case 0x2029:            case 0x2029:
1449    #endif  /* Not EBCDIC */
1450            if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;            if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;
1451            goto ANYNL01;            goto ANYNL01;
1452    
1453            case 0x000d:            case CHAR_CR:
1454            if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1;            if (ptr + 1 < end_subject && RAWUCHARTEST(ptr + 1) == CHAR_LF) ncount = 1;
1455            /* Fall through */            /* Fall through */
1456    
1457            ANYNL01:            ANYNL01:
1458            case 0x000a:            case CHAR_LF:
1459            if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS)            if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS)
1460              {              {
1461              active_count--;           /* Remove non-match possibility */              active_count--;           /* Remove non-match possibility */
# Line 1314  for (;;) Line 1482  for (;;)
1482          BOOL OK;          BOOL OK;
1483          switch (c)          switch (c)
1484            {            {
1485            case 0x000a:            VSPACE_CASES:
           case 0x000b:  
           case 0x000c:  
           case 0x000d:  
           case 0x0085:  
           case 0x2028:  
           case 0x2029:  
1486            OK = TRUE;            OK = TRUE;
1487            break;            break;
1488    
# Line 1353  for (;;) Line 1515  for (;;)
1515          BOOL OK;          BOOL OK;
1516          switch (c)          switch (c)
1517            {            {
1518            case 0x09:      /* HT */            HSPACE_CASES:
           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 */  
1519            OK = TRUE;            OK = TRUE;
1520            break;            break;
1521    
# Line 1412  for (;;) Line 1556  for (;;)
1556        if (clen > 0)        if (clen > 0)
1557          {          {
1558          BOOL OK;          BOOL OK;
1559            const pcre_uint32 *cp;
1560          const ucd_record * prop = GET_UCD(c);          const ucd_record * prop = GET_UCD(c);
1561          switch(code[2])          switch(code[2])
1562            {            {
# Line 1420  for (;;) Line 1565  for (;;)
1565            break;            break;
1566    
1567            case PT_LAMP:            case PT_LAMP:
1568            OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||            OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
1569              prop->chartype == ucp_Lt;              prop->chartype == ucp_Lt;
1570            break;            break;
1571    
1572            case PT_GC:            case PT_GC:
1573            OK = _pcre_ucp_gentype[prop->chartype] == code[3];            OK = PRIV(ucp_gentype)[prop->chartype] == code[3];
1574            break;            break;
1575    
1576            case PT_PC:            case PT_PC:
# Line 1435  for (;;) Line 1580  for (;;)
1580            case PT_SC:            case PT_SC:
1581            OK = prop->script == code[3];            OK = prop->script == code[3];
1582            break;            break;
1583    
1584            /* These are specials for combination cases. */            /* These are specials for combination cases. */
1585    
1586            case PT_ALNUM:            case PT_ALNUM:
1587            OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
1588                 _pcre_ucp_gentype[prop->chartype] == ucp_N;                 PRIV(ucp_gentype)[prop->chartype] == ucp_N;
1589            break;            break;
1590    
1591            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
1592            OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
1593                 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;                 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
1594            break;            break;
1595    
1596            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
1597            OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
1598                 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||                 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
1599                 c == CHAR_FF || c == CHAR_CR;                 c == CHAR_FF || c == CHAR_CR;
1600            break;            break;
1601    
1602            case PT_WORD:            case PT_WORD:
1603            OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
1604                 _pcre_ucp_gentype[prop->chartype] == ucp_N ||                 PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
1605                 c == CHAR_UNDERSCORE;                 c == CHAR_UNDERSCORE;
1606            break;            break;
1607    
1608              case PT_CLIST:
1609              cp = PRIV(ucd_caseless_sets) + code[3];
1610              for (;;)
1611                {
1612                if (c < *cp) { OK = FALSE; break; }
1613                if (c == *cp++) { OK = TRUE; break; }
1614                }
1615              break;
1616    
1617              case PT_UCNC:
1618              OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
1619                   c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
1620                   c >= 0xe000;
1621              break;
1622    
1623            /* Should never occur, but keep compilers from grumbling. */            /* Should never occur, but keep compilers from grumbling. */
1624    
# Line 1495  for (;;) Line 1655  for (;;)
1655        QS2:        QS2:
1656    
1657        ADD_ACTIVE(state_offset + 2, 0);        ADD_ACTIVE(state_offset + 2, 0);
1658        if (clen > 0 && UCD_CATEGORY(c) != ucp_M)        if (clen > 0)
1659          {          {
1660          const uschar *nptr = ptr + clen;          int lgb, rgb;
1661            const pcre_uchar *nptr = ptr + clen;
1662          int ncount = 0;          int ncount = 0;
1663          if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR ||          if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR ||
1664              codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY)              codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY)
# Line 1505  for (;;) Line 1666  for (;;)
1666            active_count--;           /* Remove non-match possibility */            active_count--;           /* Remove non-match possibility */
1667            next_active_state--;            next_active_state--;
1668            }            }
1669            lgb = UCD_GRAPHBREAK(c);
1670          while (nptr < end_subject)          while (nptr < end_subject)
1671            {            {
1672            int nd;            dlen = 1;
1673            int ndlen = 1;            if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); }
1674            GETCHARLEN(nd, nptr, ndlen);            rgb = UCD_GRAPHBREAK(d);
1675            if (UCD_CATEGORY(nd) != ucp_M) break;            if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
1676            ncount++;            ncount++;
1677            nptr += ndlen;            lgb = rgb;
1678              nptr += dlen;
1679            }            }
1680          ADD_NEW_DATA(-(state_offset + count), 0, ncount);          ADD_NEW_DATA(-(state_offset + count), 0, ncount);
1681          }          }
# Line 1538  for (;;) Line 1701  for (;;)
1701          int ncount = 0;          int ncount = 0;
1702          switch (c)          switch (c)
1703            {            {
1704            case 0x000b:            case CHAR_VT:
1705            case 0x000c:            case CHAR_FF:
1706            case 0x0085:            case CHAR_NEL:
1707    #ifndef EBCDIC
1708            case 0x2028:            case 0x2028:
1709            case 0x2029:            case 0x2029:
1710    #endif  /* Not EBCDIC */
1711            if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;            if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;
1712            goto ANYNL02;            goto ANYNL02;
1713    
1714            case 0x000d:            case CHAR_CR:
1715            if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1;            if (ptr + 1 < end_subject && RAWUCHARTEST(ptr + 1) == CHAR_LF) ncount = 1;
1716            /* Fall through */            /* Fall through */
1717    
1718            ANYNL02:            ANYNL02:
1719            case 0x000a:            case CHAR_LF:
1720            if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR ||            if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR ||
1721                codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY)                codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY)
1722              {              {
1723              active_count--;           /* Remove non-match possibility */              active_count--;           /* Remove non-match possibility */
1724              next_active_state--;              next_active_state--;
1725              }              }
1726            ADD_NEW_DATA(-(state_offset + count), 0, ncount);            ADD_NEW_DATA(-(state_offset + (int)count), 0, ncount);
1727            break;            break;
1728    
1729            default:            default:
# Line 1586  for (;;) Line 1751  for (;;)
1751          BOOL OK;          BOOL OK;
1752          switch (c)          switch (c)
1753            {            {
1754            case 0x000a:            VSPACE_CASES:
           case 0x000b:  
           case 0x000c:  
           case 0x000d:  
           case 0x0085:  
           case 0x2028:  
           case 0x2029:  
1755            OK = TRUE;            OK = TRUE;
1756            break;            break;
1757    
# Line 1608  for (;;) Line 1767  for (;;)
1767              active_count--;           /* Remove non-match possibility */              active_count--;           /* Remove non-match possibility */
1768              next_active_state--;              next_active_state--;
1769              }              }
1770            ADD_NEW_DATA(-(state_offset + count), 0, 0);            ADD_NEW_DATA(-(state_offset + (int)count), 0, 0);
1771            }            }
1772          }          }
1773        break;        break;
# Line 1632  for (;;) Line 1791  for (;;)
1791          BOOL OK;          BOOL OK;
1792          switch (c)          switch (c)
1793            {            {
1794            case 0x09:      /* HT */            HSPACE_CASES:
           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 */  
1795            OK = TRUE;            OK = TRUE;
1796            break;            break;
1797    
# Line 1667  for (;;) Line 1808  for (;;)
1808              active_count--;           /* Remove non-match possibility */              active_count--;           /* Remove non-match possibility */
1809              next_active_state--;              next_active_state--;
1810              }              }
1811            ADD_NEW_DATA(-(state_offset + count), 0, 0);            ADD_NEW_DATA(-(state_offset + (int)count), 0, 0);
1812            }            }
1813          }          }
1814        break;        break;
# Line 1679  for (;;) Line 1820  for (;;)
1820        case OP_PROP_EXTRA + OP_TYPEMINUPTO:        case OP_PROP_EXTRA + OP_TYPEMINUPTO:
1821        case OP_PROP_EXTRA + OP_TYPEPOSUPTO:        case OP_PROP_EXTRA + OP_TYPEPOSUPTO:
1822        if (codevalue != OP_PROP_EXTRA + OP_TYPEEXACT)        if (codevalue != OP_PROP_EXTRA + OP_TYPEEXACT)
1823          { ADD_ACTIVE(state_offset + 6, 0); }          { ADD_ACTIVE(state_offset + 1 + IMM2_SIZE + 3, 0); }
1824        count = current_state->count;  /* Number already matched */        count = current_state->count;  /* Number already matched */
1825        if (clen > 0)        if (clen > 0)
1826          {          {
1827          BOOL OK;          BOOL OK;
1828            const pcre_uint32 *cp;
1829          const ucd_record * prop = GET_UCD(c);          const ucd_record * prop = GET_UCD(c);
1830          switch(code[4])          switch(code[1 + IMM2_SIZE + 1])
1831            {            {
1832            case PT_ANY:            case PT_ANY:
1833            OK = TRUE;            OK = TRUE;
1834            break;            break;
1835    
1836            case PT_LAMP:            case PT_LAMP:
1837            OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||            OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
1838              prop->chartype == ucp_Lt;              prop->chartype == ucp_Lt;
1839            break;            break;
1840    
1841            case PT_GC:            case PT_GC:
1842            OK = _pcre_ucp_gentype[prop->chartype] == code[5];            OK = PRIV(ucp_gentype)[prop->chartype] == code[1 + IMM2_SIZE + 2];
1843            break;            break;
1844    
1845            case PT_PC:            case PT_PC:
1846            OK = prop->chartype == code[5];            OK = prop->chartype == code[1 + IMM2_SIZE + 2];
1847            break;            break;
1848    
1849            case PT_SC:            case PT_SC:
1850            OK = prop->script == code[5];            OK = prop->script == code[1 + IMM2_SIZE + 2];
1851            break;            break;
1852    
1853            /* These are specials for combination cases. */            /* These are specials for combination cases. */
1854    
1855            case PT_ALNUM:            case PT_ALNUM:
1856            OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
1857                 _pcre_ucp_gentype[prop->chartype] == ucp_N;                 PRIV(ucp_gentype)[prop->chartype] == ucp_N;
1858            break;            break;
1859    
1860            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
1861            OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
1862                 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;                 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
1863            break;            break;
1864    
1865            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
1866            OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
1867                 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||                 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
1868                 c == CHAR_FF || c == CHAR_CR;                 c == CHAR_FF || c == CHAR_CR;
1869            break;            break;
1870    
1871            case PT_WORD:            case PT_WORD:
1872            OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||            OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
1873                 _pcre_ucp_gentype[prop->chartype] == ucp_N ||                 PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
1874                 c == CHAR_UNDERSCORE;                 c == CHAR_UNDERSCORE;
1875            break;            break;
1876    
1877              case PT_CLIST:
1878              cp = PRIV(ucd_caseless_sets) + code[1 + IMM2_SIZE + 2];
1879              for (;;)
1880                {
1881                if (c < *cp) { OK = FALSE; break; }
1882                if (c == *cp++) { OK = TRUE; break; }
1883                }
1884              break;
1885    
1886              case PT_UCNC:
1887              OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
1888                   c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
1889                   c >= 0xe000;
1890              break;
1891    
1892            /* Should never occur, but keep compilers from grumbling. */            /* Should never occur, but keep compilers from grumbling. */
1893    
# Line 1747  for (;;) Line 1904  for (;;)
1904              next_active_state--;              next_active_state--;
1905              }              }
1906            if (++count >= GET2(code, 1))            if (++count >= GET2(code, 1))
1907              { ADD_NEW(state_offset + 6, 0); }              { ADD_NEW(state_offset + 1 + IMM2_SIZE + 3, 0); }
1908            else            else
1909              { ADD_NEW(state_offset, count); }              { ADD_NEW(state_offset, count); }
1910            }            }
# Line 1760  for (;;) Line 1917  for (;;)
1917        case OP_EXTUNI_EXTRA + OP_TYPEMINUPTO:        case OP_EXTUNI_EXTRA + OP_TYPEMINUPTO:
1918        case OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO:        case OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO:
1919        if (codevalue != OP_EXTUNI_EXTRA + OP_TYPEEXACT)        if (codevalue != OP_EXTUNI_EXTRA + OP_TYPEEXACT)
1920          { ADD_ACTIVE(state_offset + 4, 0); }          { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); }
1921        count = current_state->count;  /* Number already matched */        count = current_state->count;  /* Number already matched */
1922        if (clen > 0 && UCD_CATEGORY(c) != ucp_M)        if (clen > 0)
1923          {          {
1924          const uschar *nptr = ptr + clen;          int lgb, rgb;
1925            const pcre_uchar *nptr = ptr + clen;
1926          int ncount = 0;          int ncount = 0;
1927          if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO)          if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO)
1928            {            {
1929            active_count--;           /* Remove non-match possibility */            active_count--;           /* Remove non-match possibility */
1930            next_active_state--;            next_active_state--;
1931            }            }
1932            lgb = UCD_GRAPHBREAK(c);
1933          while (nptr < end_subject)          while (nptr < end_subject)
1934            {            {
1935            int nd;            dlen = 1;
1936            int ndlen = 1;            if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); }
1937            GETCHARLEN(nd, nptr, ndlen);            rgb = UCD_GRAPHBREAK(d);
1938            if (UCD_CATEGORY(nd) != ucp_M) break;            if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
1939            ncount++;            ncount++;
1940            nptr += ndlen;            lgb = rgb;
1941              nptr += dlen;
1942            }            }
1943            if (nptr >= end_subject && (md->moptions & PCRE_PARTIAL_HARD) != 0)
1944                reset_could_continue = TRUE;
1945          if (++count >= GET2(code, 1))          if (++count >= GET2(code, 1))
1946            { ADD_NEW_DATA(-(state_offset + 4), 0, ncount); }            { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, ncount); }
1947          else          else
1948            { ADD_NEW_DATA(-state_offset, count, ncount); }            { ADD_NEW_DATA(-state_offset, count, ncount); }
1949          }          }
# Line 1794  for (;;) Line 1956  for (;;)
1956        case OP_ANYNL_EXTRA + OP_TYPEMINUPTO:        case OP_ANYNL_EXTRA + OP_TYPEMINUPTO:
1957        case OP_ANYNL_EXTRA + OP_TYPEPOSUPTO:        case OP_ANYNL_EXTRA + OP_TYPEPOSUPTO:
1958        if (codevalue != OP_ANYNL_EXTRA + OP_TYPEEXACT)        if (codevalue != OP_ANYNL_EXTRA + OP_TYPEEXACT)
1959          { ADD_ACTIVE(state_offset + 4, 0); }          { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); }
1960        count = current_state->count;  /* Number already matched */        count = current_state->count;  /* Number already matched */
1961        if (clen > 0)        if (clen > 0)
1962          {          {
1963          int ncount = 0;          int ncount = 0;
1964          switch (c)          switch (c)
1965            {            {
1966            case 0x000b:            case CHAR_VT:
1967            case 0x000c:            case CHAR_FF:
1968            case 0x0085:            case CHAR_NEL:
1969    #ifndef EBCDIC
1970            case 0x2028:            case 0x2028:
1971            case 0x2029:            case 0x2029:
1972    #endif  /* Not EBCDIC */
1973            if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;            if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;
1974            goto ANYNL03;            goto ANYNL03;
1975    
1976            case 0x000d:            case CHAR_CR:
1977            if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1;            if (ptr + 1 < end_subject && RAWUCHARTEST(ptr + 1) == CHAR_LF) ncount = 1;
1978            /* Fall through */            /* Fall through */
1979    
1980            ANYNL03:            ANYNL03:
1981            case 0x000a:            case CHAR_LF:
1982            if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO)            if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO)
1983              {              {
1984              active_count--;           /* Remove non-match possibility */              active_count--;           /* Remove non-match possibility */
1985              next_active_state--;              next_active_state--;
1986              }              }
1987            if (++count >= GET2(code, 1))            if (++count >= GET2(code, 1))
1988              { ADD_NEW_DATA(-(state_offset + 4), 0, ncount); }              { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, ncount); }
1989            else            else
1990              { ADD_NEW_DATA(-state_offset, count, ncount); }              { ADD_NEW_DATA(-state_offset, count, ncount); }
1991            break;            break;
# Line 1838  for (;;) Line 2002  for (;;)
2002        case OP_VSPACE_EXTRA + OP_TYPEMINUPTO:        case OP_VSPACE_EXTRA + OP_TYPEMINUPTO:
2003        case OP_VSPACE_EXTRA + OP_TYPEPOSUPTO:        case OP_VSPACE_EXTRA + OP_TYPEPOSUPTO:
2004        if (codevalue != OP_VSPACE_EXTRA + OP_TYPEEXACT)        if (codevalue != OP_VSPACE_EXTRA + OP_TYPEEXACT)
2005          { ADD_ACTIVE(state_offset + 4, 0); }          { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); }
2006        count = current_state->count;  /* Number already matched */        count = current_state->count;  /* Number already matched */
2007        if (clen > 0)        if (clen > 0)
2008          {          {
2009          BOOL OK;          BOOL OK;
2010          switch (c)          switch (c)
2011            {            {
2012            case 0x000a:            VSPACE_CASES:
           case 0x000b:  
           case 0x000c:  
           case 0x000d:  
           case 0x0085:  
           case 0x2028:  
           case 0x2029:  
2013            OK = TRUE;            OK = TRUE;
2014            break;            break;
2015    
# Line 1867  for (;;) Line 2025  for (;;)
2025              next_active_state--;              next_active_state--;
2026              }              }
2027            if (++count >= GET2(code, 1))            if (++count >= GET2(code, 1))
2028              { ADD_NEW_DATA(-(state_offset + 4), 0, 0); }              { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, 0); }
2029            else            else
2030              { ADD_NEW_DATA(-state_offset, count, 0); }              { ADD_NEW_DATA(-state_offset, count, 0); }
2031            }            }
# Line 1880  for (;;) Line 2038  for (;;)
2038        case OP_HSPACE_EXTRA + OP_TYPEMINUPTO:        case OP_HSPACE_EXTRA + OP_TYPEMINUPTO:
2039        case OP_HSPACE_EXTRA + OP_TYPEPOSUPTO:        case OP_HSPACE_EXTRA + OP_TYPEPOSUPTO:
2040        if (codevalue != OP_HSPACE_EXTRA + OP_TYPEEXACT)        if (codevalue != OP_HSPACE_EXTRA + OP_TYPEEXACT)
2041          { ADD_ACTIVE(state_offset + 4, 0); }          { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); }
2042        count = current_state->count;  /* Number already matched */        count = current_state->count;  /* Number already matched */
2043        if (clen > 0)        if (clen > 0)
2044          {          {
2045          BOOL OK;          BOOL OK;
2046          switch (c)          switch (c)
2047            {            {
2048            case 0x09:      /* HT */            HSPACE_CASES:
           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 */  
2049            OK = TRUE;            OK = TRUE;
2050            break;            break;
2051    
# Line 1922  for (;;) Line 2062  for (;;)
2062              next_active_state--;              next_active_state--;
2063              }              }
2064            if (++count >= GET2(code, 1))            if (++count >= GET2(code, 1))
2065              { ADD_NEW_DATA(-(state_offset + 4), 0, 0); }              { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, 0); }
2066            else            else
2067              { ADD_NEW_DATA(-state_offset, count, 0); }              { ADD_NEW_DATA(-state_offset, count, 0); }
2068            }            }
# Line 1941  for (;;) Line 2081  for (;;)
2081        break;        break;
2082    
2083        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
2084        case OP_CHARNC:        case OP_CHARI:
2085        if (clen == 0) break;        if (clen == 0) break;
2086    
2087  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2088        if (utf8)        if (utf)
2089          {          {
2090          if (c == d) { ADD_NEW(state_offset + dlen + 1, 0); } else          if (c == d) { ADD_NEW(state_offset + dlen + 1, 0); } else
2091            {            {
2092            unsigned int othercase;            unsigned int othercase;
2093            if (c < 128) othercase = fcc[c]; else            if (c < 128)
2094                othercase = fcc[c];
2095            /* If we have Unicode property support, we can use it to test the            else
2096            other case of the character. */              /* If we have Unicode property support, we can use it to test the
2097                other case of the character. */
2098  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2099            othercase = UCD_OTHERCASE(c);              othercase = UCD_OTHERCASE(c);
2100  #else  #else
2101            othercase = NOTACHAR;              othercase = NOTACHAR;
2102  #endif  #endif
2103    
2104            if (d == othercase) { ADD_NEW(state_offset + dlen + 1, 0); }            if (d == othercase) { ADD_NEW(state_offset + dlen + 1, 0); }
2105            }            }
2106          }          }
2107        else        else
2108  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
2109          /* Not UTF mode */
       /* Non-UTF-8 mode */  
2110          {          {
2111          if (lcc[c] == lcc[d]) { ADD_NEW(state_offset + 2, 0); }          if (TABLE_GET(c, lcc, c) == TABLE_GET(d, lcc, d))
2112              { ADD_NEW(state_offset + 2, 0); }
2113          }          }
2114        break;        break;
2115    
# Line 1981  for (;;) Line 2121  for (;;)
2121        to wait for them to pass before continuing. */        to wait for them to pass before continuing. */
2122    
2123        case OP_EXTUNI:        case OP_EXTUNI:
2124        if (clen > 0 && UCD_CATEGORY(c) != ucp_M)        if (clen > 0)
2125          {          {
2126          const uschar *nptr = ptr + clen;          int lgb, rgb;
2127            const pcre_uchar *nptr = ptr + clen;
2128          int ncount = 0;          int ncount = 0;
2129            lgb = UCD_GRAPHBREAK(c);
2130          while (nptr < end_subject)          while (nptr < end_subject)
2131            {            {
2132            int nclen = 1;            dlen = 1;
2133            GETCHARLEN(c, nptr, nclen);            if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); }
2134            if (UCD_CATEGORY(c) != ucp_M) break;            rgb = UCD_GRAPHBREAK(d);
2135              if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
2136            ncount++;            ncount++;
2137            nptr += nclen;            lgb = rgb;
2138              nptr += dlen;
2139            }            }
2140            if (nptr >= end_subject && (md->moptions & PCRE_PARTIAL_HARD) != 0)
2141                reset_could_continue = TRUE;
2142          ADD_NEW_DATA(-(state_offset + 1), 0, ncount);          ADD_NEW_DATA(-(state_offset + 1), 0, ncount);
2143          }          }
2144        break;        break;
# Line 2006  for (;;) Line 2152  for (;;)
2152        case OP_ANYNL:        case OP_ANYNL:
2153        if (clen > 0) switch(c)        if (clen > 0) switch(c)
2154          {          {
2155          case 0x000b:          case CHAR_VT:
2156          case 0x000c:          case CHAR_FF:
2157          case 0x0085:          case CHAR_NEL:
2158    #ifndef EBCDIC
2159          case 0x2028:          case 0x2028:
2160          case 0x2029:          case 0x2029:
2161    #endif  /* Not EBCDIC */
2162          if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;          if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;
2163    
2164          case 0x000a:          case CHAR_LF:
2165          ADD_NEW(state_offset + 1, 0);          ADD_NEW(state_offset + 1, 0);
2166          break;          break;
2167    
2168          case 0x000d:          case CHAR_CR:
2169          if (ptr + 1 < end_subject && ptr[1] == 0x0a)          if (ptr + 1 >= end_subject)
2170              {
2171              ADD_NEW(state_offset + 1, 0);
2172              if ((md->moptions & PCRE_PARTIAL_HARD) != 0)
2173                reset_could_continue = TRUE;
2174              }
2175            else if (RAWUCHARTEST(ptr + 1) == CHAR_LF)
2176            {            {
2177            ADD_NEW_DATA(-(state_offset + 1), 0, 1);            ADD_NEW_DATA(-(state_offset + 1), 0, 1);
2178            }            }
# Line 2034  for (;;) Line 2188  for (;;)
2188        case OP_NOT_VSPACE:        case OP_NOT_VSPACE:
2189        if (clen > 0) switch(c)        if (clen > 0) switch(c)
2190          {          {
2191          case 0x000a:          VSPACE_CASES:
         case 0x000b:  
         case 0x000c:  
         case 0x000d:  
         case 0x0085:  
         case 0x2028:  
         case 0x2029:  
2192          break;          break;
2193    
2194          default:          default:
# Line 2053  for (;;) Line 2201  for (;;)
2201        case OP_VSPACE:        case OP_VSPACE:
2202        if (clen > 0) switch(c)        if (clen > 0) switch(c)
2203          {          {
2204          case 0x000a:          VSPACE_CASES:
         case 0x000b:  
         case 0x000c:  
         case 0x000d:  
         case 0x0085:  
         case 0x2028:  
         case 0x2029:  
2205          ADD_NEW(state_offset + 1, 0);          ADD_NEW(state_offset + 1, 0);
2206          break;          break;
2207    
2208          default: break;          default:
2209            break;
2210          }          }
2211        break;        break;
2212    
# Line 2071  for (;;) Line 2214  for (;;)
2214        case OP_NOT_HSPACE:        case OP_NOT_HSPACE:
2215        if (clen > 0) switch(c)        if (clen > 0) switch(c)
2216          {          {
2217          case 0x09:      /* HT */          HSPACE_CASES:
         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 */  
2218          break;          break;
2219    
2220          default:          default:
# Line 2102  for (;;) Line 2227  for (;;)
2227        case OP_HSPACE:        case OP_HSPACE:
2228        if (clen > 0) switch(c)        if (clen > 0) switch(c)
2229          {          {
2230          case 0x09:      /* HT */          HSPACE_CASES:
         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 */  
2231          ADD_NEW(state_offset + 1, 0);          ADD_NEW(state_offset + 1, 0);
2232          break;          break;
2233    
2234            default:
2235            break;
2236          }          }
2237        break;        break;
2238    
2239        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
2240        /* Match a negated single character. This is only used for one-byte        /* Match a negated single character casefully. */
       characters, that is, we know that d < 256. The character we are  
       checking (c) can be multibyte. */  
2241    
2242        case OP_NOT:        case OP_NOT:
2243          if (clen > 0 && c != d) { ADD_NEW(state_offset + dlen + 1, 0); }
2244          break;
2245    
2246          /*-----------------------------------------------------------------*/
2247          /* Match a negated single character caselessly. */
2248    
2249          case OP_NOTI:
2250        if (clen > 0)        if (clen > 0)
2251          {          {
2252          unsigned int otherd = ((ims & PCRE_CASELESS) != 0)? fcc[d] : d;          unsigned int otherd;
2253          if (c != d && c != otherd) { ADD_NEW(state_offset + dlen + 1, 0); }  #ifdef SUPPORT_UTF
2254            if (utf && d >= 128)
2255              {
2256    #ifdef SUPPORT_UCP
2257              otherd = UCD_OTHERCASE(d);
2258    #endif  /* SUPPORT_UCP */
2259              }
2260            else
2261    #endif  /* SUPPORT_UTF */
2262            otherd = TABLE_GET(d, fcc, d);
2263            if (c != d && c != otherd)
2264              { ADD_NEW(state_offset + dlen + 1, 0); }
2265          }          }
2266        break;        break;
2267    
2268        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
2269          case OP_PLUSI:
2270          case OP_MINPLUSI:
2271          case OP_POSPLUSI:
2272          case OP_NOTPLUSI:
2273          case OP_NOTMINPLUSI:
2274          case OP_NOTPOSPLUSI:
2275          caseless = TRUE;
2276          codevalue -= OP_STARI - OP_STAR;
2277    
2278          /* Fall through */
2279        case OP_PLUS:        case OP_PLUS:
2280        case OP_MINPLUS:        case OP_MINPLUS:
2281        case OP_POSPLUS:        case OP_POSPLUS:
# Line 2150  for (;;) Line 2286  for (;;)
2286        if (count > 0) { ADD_ACTIVE(state_offset + dlen + 1, 0); }        if (count > 0) { ADD_ACTIVE(state_offset + dlen + 1, 0); }
2287        if (clen > 0)        if (clen > 0)
2288          {          {
2289          unsigned int otherd = NOTACHAR;          pcre_uint32 otherd = NOTACHAR;
2290          if ((ims & PCRE_CASELESS) != 0)          if (caseless)
2291            {            {
2292  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2293            if (utf8 && d >= 128)            if (utf && d >= 128)
2294              {              {
2295  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2296              otherd = UCD_OTHERCASE(d);              otherd = UCD_OTHERCASE(d);
2297  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2298              }              }
2299            else            else
2300  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
2301            otherd = fcc[d];            otherd = TABLE_GET(d, fcc, d);
2302            }            }
2303          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
2304            {            {
# Line 2179  for (;;) Line 2315  for (;;)
2315        break;        break;
2316    
2317        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
2318          case OP_QUERYI:
2319          case OP_MINQUERYI:
2320          case OP_POSQUERYI:
2321          case OP_NOTQUERYI:
2322          case OP_NOTMINQUERYI:
2323          case OP_NOTPOSQUERYI:
2324          caseless = TRUE;
2325          codevalue -= OP_STARI - OP_STAR;
2326          /* Fall through */
2327        case OP_QUERY:        case OP_QUERY:
2328        case OP_MINQUERY:        case OP_MINQUERY:
2329        case OP_POSQUERY:        case OP_POSQUERY:
# Line 2188  for (;;) Line 2333  for (;;)
2333        ADD_ACTIVE(state_offset + dlen + 1, 0);        ADD_ACTIVE(state_offset + dlen + 1, 0);
2334        if (clen > 0)        if (clen > 0)
2335          {          {
2336          unsigned int otherd = NOTACHAR;          pcre_uint32 otherd = NOTACHAR;
2337          if ((ims & PCRE_CASELESS) != 0)          if (caseless)
2338            {            {
2339  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2340            if (utf8 && d >= 128)            if (utf && d >= 128)
2341              {              {
2342  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2343              otherd = UCD_OTHERCASE(d);              otherd = UCD_OTHERCASE(d);
2344  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2345              }              }
2346            else            else
2347  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
2348            otherd = fcc[d];            otherd = TABLE_GET(d, fcc, d);
2349            }            }
2350          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
2351            {            {
# Line 2215  for (;;) Line 2360  for (;;)
2360        break;        break;
2361    
2362        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
2363          case OP_STARI:
2364          case OP_MINSTARI:
2365          case OP_POSSTARI:
2366          case OP_NOTSTARI:
2367          case OP_NOTMINSTARI:
2368          case OP_NOTPOSSTARI:
2369          caseless = TRUE;
2370          codevalue -= OP_STARI - OP_STAR;
2371          /* Fall through */
2372        case OP_STAR:        case OP_STAR:
2373        case OP_MINSTAR:        case OP_MINSTAR:
2374        case OP_POSSTAR:        case OP_POSSTAR:
# Line 2224  for (;;) Line 2378  for (;;)
2378        ADD_ACTIVE(state_offset + dlen + 1, 0);        ADD_ACTIVE(state_offset + dlen + 1, 0);
2379        if (clen > 0)        if (clen > 0)
2380          {          {
2381          unsigned int otherd = NOTACHAR;          pcre_uint32 otherd = NOTACHAR;
2382          if ((ims & PCRE_CASELESS) != 0)          if (caseless)
2383            {            {
2384  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2385            if (utf8 && d >= 128)            if (utf && d >= 128)
2386              {              {
2387  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2388              otherd = UCD_OTHERCASE(d);              otherd = UCD_OTHERCASE(d);
2389  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2390              }              }
2391            else            else
2392  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
2393            otherd = fcc[d];            otherd = TABLE_GET(d, fcc, d);
2394            }            }
2395          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
2396            {            {
# Line 2251  for (;;) Line 2405  for (;;)
2405        break;        break;
2406    
2407        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
2408          case OP_EXACTI:
2409          case OP_NOTEXACTI:
2410          caseless = TRUE;
2411          codevalue -= OP_STARI - OP_STAR;
2412          /* Fall through */
2413        case OP_EXACT:        case OP_EXACT:
2414        case OP_NOTEXACT:        case OP_NOTEXACT:
2415        count = current_state->count;  /* Number already matched */        count = current_state->count;  /* Number already matched */
2416        if (clen > 0)        if (clen > 0)
2417          {          {
2418          unsigned int otherd = NOTACHAR;          pcre_uint32 otherd = NOTACHAR;
2419          if ((ims & PCRE_CASELESS) != 0)          if (caseless)
2420            {            {
2421  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2422            if (utf8 && d >= 128)            if (utf && d >= 128)
2423              {              {
2424  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2425              otherd = UCD_OTHERCASE(d);              otherd = UCD_OTHERCASE(d);
2426  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2427              }              }
2428            else            else
2429  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
2430            otherd = fcc[d];            otherd = TABLE_GET(d, fcc, d);
2431            }            }
2432          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
2433            {            {
2434            if (++count >= GET2(code, 1))            if (++count >= GET2(code, 1))
2435              { ADD_NEW(state_offset + dlen + 3, 0); }              { ADD_NEW(state_offset + dlen + 1 + IMM2_SIZE, 0); }
2436            else            else
2437              { ADD_NEW(state_offset, count); }              { ADD_NEW(state_offset, count); }
2438            }            }
# Line 2281  for (;;) Line 2440  for (;;)
2440        break;        break;
2441    
2442        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
2443          case OP_UPTOI:
2444          case OP_MINUPTOI:
2445          case OP_POSUPTOI:
2446          case OP_NOTUPTOI:
2447          case OP_NOTMINUPTOI:
2448          case OP_NOTPOSUPTOI:
2449          caseless = TRUE;
2450          codevalue -= OP_STARI - OP_STAR;
2451          /* Fall through */
2452        case OP_UPTO:        case OP_UPTO:
2453        case OP_MINUPTO:        case OP_MINUPTO:
2454        case OP_POSUPTO:        case OP_POSUPTO:
2455        case OP_NOTUPTO:        case OP_NOTUPTO:
2456        case OP_NOTMINUPTO:        case OP_NOTMINUPTO:
2457        case OP_NOTPOSUPTO:        case OP_NOTPOSUPTO:
2458        ADD_ACTIVE(state_offset + dlen + 3, 0);        ADD_ACTIVE(state_offset + dlen + 1 + IMM2_SIZE, 0);
2459        count = current_state->count;  /* Number already matched */        count = current_state->count;  /* Number already matched */
2460        if (clen > 0)        if (clen > 0)
2461          {          {
2462          unsigned int otherd = NOTACHAR;          pcre_uint32 otherd = NOTACHAR;
2463          if ((ims & PCRE_CASELESS) != 0)          if (caseless)
2464            {            {
2465  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
2466            if (utf8 && d >= 128)            if (utf && d >= 128)
2467              {              {
2468  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
2469              otherd = UCD_OTHERCASE(d);              otherd = UCD_OTHERCASE(d);
2470  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
2471              }              }
2472            else            else
2473  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF */
2474            otherd = fcc[d];            otherd = TABLE_GET(d, fcc, d);
2475            }            }
2476          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))          if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
2477            {            {
# Line 2313  for (;;) Line 2481  for (;;)
2481              next_active_state--;              next_active_state--;
2482              }              }
2483            if (++count >= GET2(code, 1))            if (++count >= GET2(code, 1))
2484              { ADD_NEW(state_offset + dlen + 3, 0); }              { ADD_NEW(state_offset + dlen + 1 + IMM2_SIZE, 0); }
2485            else            else
2486              { ADD_NEW(state_offset, count); }              { ADD_NEW(state_offset, count); }
2487            }            }
# Line 2330  for (;;) Line 2498  for (;;)
2498          {          {
2499          BOOL isinclass = FALSE;          BOOL isinclass = FALSE;
2500          int next_state_offset;          int next_state_offset;
2501          const uschar *ecode;          const pcre_uchar *ecode;
2502    
2503          /* For a simple class, there is always just a 32-byte table, and we          /* For a simple class, there is always just a 32-byte table, and we
2504          can set isinclass from it. */          can set isinclass from it. */
2505    
2506          if (codevalue != OP_XCLASS)          if (codevalue != OP_XCLASS)
2507            {            {
2508            ecode = code + 33;            ecode = code + 1 + (32 / sizeof(pcre_uchar));
2509            if (clen > 0)            if (clen > 0)
2510              {              {
2511              isinclass = (c > 255)? (codevalue == OP_NCLASS) :              isinclass = (c > 255)? (codevalue == OP_NCLASS) :
2512                ((code[1 + c/8] & (1 << (c&7))) != 0);                ((((pcre_uint8 *)(code + 1))[c/8] & (1 << (c&7))) != 0);
2513              }              }
2514            }            }
2515    
# Line 2352  for (;;) Line 2520  for (;;)
2520          else          else
2521           {           {
2522           ecode = code + GET(code, 1);           ecode = code + GET(code, 1);
2523           if (clen > 0) isinclass = _pcre_xclass(c, code + 1 + LINK_SIZE);           if (clen > 0) isinclass = PRIV(xclass)(c, code + 1 + LINK_SIZE, utf);
2524           }           }
2525    
2526          /* At this point, isinclass is set for all kinds of class, and ecode          /* At this point, isinclass is set for all kinds of class, and ecode
# Line 2386  for (;;) Line 2554  for (;;)
2554            case OP_CRMINRANGE:            case OP_CRMINRANGE:
2555            count = current_state->count;  /* Already matched */            count = current_state->count;  /* Already matched */
2556            if (count >= GET2(ecode, 1))            if (count >= GET2(ecode, 1))
2557              { ADD_ACTIVE(next_state_offset + 5, 0); }              { ADD_ACTIVE(next_state_offset + 1 + 2 * IMM2_SIZE, 0); }
2558            if (isinclass)            if (isinclass)
2559              {              {
2560              int max = GET2(ecode, 3);              unsigned int max = GET2(ecode, 1 + IMM2_SIZE);
2561              if (++count >= max && max != 0)   /* Max 0 => no limit */              if (++count >= max && max != 0)   /* Max 0 => no limit */
2562                { ADD_NEW(next_state_offset + 5, 0); }                { ADD_NEW(next_state_offset + 1 + 2 * IMM2_SIZE, 0); }
2563              else              else
2564                { ADD_NEW(state_offset, count); }                { ADD_NEW(state_offset, count); }
2565              }              }
# Line 2422  for (;;) Line 2590  for (;;)
2590          int rc;          int rc;
2591          int local_offsets[2];          int local_offsets[2];
2592          int local_workspace[1000];          int local_workspace[1000];
2593          const uschar *endasscode = code + GET(code, 1);          const pcre_uchar *endasscode = code + GET(code, 1);
2594    
2595          while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1);          while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1);
2596    
# Line 2435  for (;;) Line 2603  for (;;)
2603            sizeof(local_offsets)/sizeof(int),    /* size of same */            sizeof(local_offsets)/sizeof(int),    /* size of same */
2604            local_workspace,                      /* workspace vector */            local_workspace,                      /* workspace vector */
2605            sizeof(local_workspace)/sizeof(int),  /* size of same */            sizeof(local_workspace)/sizeof(int),  /* size of same */
2606            ims,                                  /* the current ims flags */            rlevel);                              /* function recursion level */
           rlevel,                               /* function recursion level */  
           recursing);                           /* pass on regex recursion */  
2607    
2608          if (rc == PCRE_ERROR_DFA_UITEM) return rc;          if (rc == PCRE_ERROR_DFA_UITEM) return rc;
2609          if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK))          if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK))
# Line 2461  for (;;) Line 2627  for (;;)
2627          if (code[LINK_SIZE+1] == OP_CALLOUT)          if (code[LINK_SIZE+1] == OP_CALLOUT)
2628            {            {
2629            rrc = 0;            rrc = 0;
2630            if (pcre_callout != NULL)            if (PUBL(callout) != NULL)
2631              {              {
2632              pcre_callout_block cb;              PUBL(callout_block) cb;
2633              cb.version          = 1;   /* Version 1 of the callout block */              cb.version          = 1;   /* Version 1 of the callout block */
2634              cb.callout_number   = code[LINK_SIZE+2];              cb.callout_number   = code[LINK_SIZE+2];
2635              cb.offset_vector    = offsets;              cb.offset_vector    = offsets;
2636    #if defined COMPILE_PCRE8
2637              cb.subject          = (PCRE_SPTR)start_subject;              cb.subject          = (PCRE_SPTR)start_subject;
2638    #elif defined COMPILE_PCRE16
2639                cb.subject          = (PCRE_SPTR16)start_subject;
2640    #elif defined COMPILE_PCRE32
2641                cb.subject          = (PCRE_SPTR32)start_subject;
2642    #endif
2643              cb.subject_length   = (int)(end_subject - start_subject);              cb.subject_length   = (int)(end_subject - start_subject);
2644              cb.start_match      = (int)(current_subject - start_subject);              cb.start_match      = (int)(current_subject - start_subject);
2645              cb.current_position = (int)(ptr - start_subject);              cb.current_position = (int)(ptr - start_subject);
# Line 2476  for (;;) Line 2648  for (;;)
2648              cb.capture_top      = 1;              cb.capture_top      = 1;
2649              cb.capture_last     = -1;              cb.capture_last     = -1;
2650              cb.callout_data     = md->callout_data;              cb.callout_data     = md->callout_data;
2651              if ((rrc = (*pcre_callout)(&cb)) < 0) return rrc;   /* Abandon */              cb.mark             = NULL;   /* No (*MARK) support */
2652                if ((rrc = (*PUBL(callout))(&cb)) < 0) return rrc;   /* Abandon */
2653              }              }
2654            if (rrc > 0) break;                      /* Fail this thread */            if (rrc > 0) break;                      /* Fail this thread */
2655            code += _pcre_OP_lengths[OP_CALLOUT];    /* Skip callout data */            code += PRIV(OP_lengths)[OP_CALLOUT];    /* Skip callout data */
2656            }            }
2657    
2658          condcode = code[LINK_SIZE+1];          condcode = code[LINK_SIZE+1];
# Line 2500  for (;;) Line 2673  for (;;)
2673    
2674          else if (condcode == OP_RREF || condcode == OP_NRREF)          else if (condcode == OP_RREF || condcode == OP_NRREF)
2675            {            {
2676            int value = GET2(code, LINK_SIZE+2);            int value = GET2(code, LINK_SIZE + 2);
2677            if (value != RREF_ANY) return PCRE_ERROR_DFA_UCOND;            if (value != RREF_ANY) return PCRE_ERROR_DFA_UCOND;
2678            if (recursing > 0)            if (md->recursive != NULL)
2679              { ADD_ACTIVE(state_offset + LINK_SIZE + 4, 0); }              { ADD_ACTIVE(state_offset + LINK_SIZE + 2 + IMM2_SIZE, 0); }
2680            else { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); }            else { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); }
2681            }            }
2682    
# Line 2512  for (;;) Line 2685  for (;;)
2685          else          else
2686            {            {
2687            int rc;            int rc;
2688            const uschar *asscode = code + LINK_SIZE + 1;            const pcre_uchar *asscode = code + LINK_SIZE + 1;
2689            const uschar *endasscode = asscode + GET(asscode, 1);            const pcre_uchar *endasscode = asscode + GET(asscode, 1);
2690    
2691            while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1);            while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1);
2692    
# Line 2526  for (;;) Line 2699  for (;;)
2699              sizeof(local_offsets)/sizeof(int),    /* size of same */              sizeof(local_offsets)/sizeof(int),    /* size of same */
2700              local_workspace,                      /* workspace vector */              local_workspace,                      /* workspace vector */
2701              sizeof(local_workspace)/sizeof(int),  /* size of same */              sizeof(local_workspace)/sizeof(int),  /* size of same */
2702              ims,                                  /* the current ims flags */              rlevel);                              /* function recursion level */
             rlevel,                               /* function recursion level */  
             recursing);                           /* pass on regex recursion */  
2703    
2704            if (rc == PCRE_ERROR_DFA_UITEM) return rc;            if (rc == PCRE_ERROR_DFA_UITEM) return rc;
2705            if ((rc >= 0) ==            if ((rc >= 0) ==
# Line 2543  for (;;) Line 2714  for (;;)
2714        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
2715        case OP_RECURSE:        case OP_RECURSE:
2716          {          {
2717            dfa_recursion_info *ri;
2718          int local_offsets[1000];          int local_offsets[1000];
2719          int local_workspace[1000];          int local_workspace[1000];
2720            const pcre_uchar *callpat = start_code + GET(code, 1);
2721            int recno = (callpat == md->start_code)? 0 :
2722              GET2(callpat, 1 + LINK_SIZE);
2723          int rc;          int rc;
2724    
2725          DPRINTF(("%.*sStarting regex recursion %d\n", rlevel*2-2, SP,          DPRINTF(("%.*sStarting regex recursion\n", rlevel*2-2, SP));
2726            recursing + 1));  
2727            /* Check for repeating a recursion without advancing the subject
2728            pointer. This should catch convoluted mutual recursions. (Some simple
2729            cases are caught at compile time.) */
2730    
2731            for (ri = md->recursive; ri != NULL; ri = ri->prevrec)
2732              if (recno == ri->group_num && ptr == ri->subject_position)
2733                return PCRE_ERROR_RECURSELOOP;
2734    
2735            /* Remember this recursion and where we started it so as to
2736            catch infinite loops. */
2737    
2738            new_recursive.group_num = recno;
2739            new_recursive.subject_position = ptr;
2740            new_recursive.prevrec = md->recursive;
2741            md->recursive = &new_recursive;
2742    
2743          rc = internal_dfa_exec(          rc = internal_dfa_exec(
2744            md,                                   /* fixed match data */            md,                                   /* fixed match data */
2745            start_code + GET(code, 1),            /* this subexpression's code */            callpat,                              /* this subexpression's code */
2746            ptr,                                  /* where we currently are */            ptr,                                  /* where we currently are */
2747            (int)(ptr - start_subject),           /* start offset */            (int)(ptr - start_subject),           /* start offset */
2748            local_offsets,                        /* offset vector */            local_offsets,                        /* offset vector */
2749            sizeof(local_offsets)/sizeof(int),    /* size of same */            sizeof(local_offsets)/sizeof(int),    /* size of same */
2750            local_workspace,                      /* workspace vector */            local_workspace,                      /* workspace vector */
2751            sizeof(local_workspace)/sizeof(int),  /* size of same */            sizeof(local_workspace)/sizeof(int),  /* size of same */
2752            ims,                                  /* the current ims flags */            rlevel);                              /* function recursion level */
2753            rlevel,                               /* function recursion level */  
2754            recursing + 1);                       /* regex recurse level */          md->recursive = new_recursive.prevrec;  /* Done this recursion */
2755    
2756          DPRINTF(("%.*sReturn from regex recursion %d: rc=%d\n", rlevel*2-2, SP,          DPRINTF(("%.*sReturn from regex recursion: rc=%d\n", rlevel*2-2, SP,
2757            recursing + 1, rc));            rc));
2758    
2759          /* Ran out of internal offsets */          /* Ran out of internal offsets */
2760    
# Line 2578  for (;;) Line 2768  for (;;)
2768            {            {
2769            for (rc = rc*2 - 2; rc >= 0; rc -= 2)            for (rc = rc*2 - 2; rc >= 0; rc -= 2)
2770              {              {
             const uschar *p = start_subject + local_offsets[rc];  
             const uschar *pp = start_subject + local_offsets[rc+1];  
2771              int charcount = local_offsets[rc+1] - local_offsets[rc];              int charcount = local_offsets[rc+1] - local_offsets[rc];
2772              while (p < pp) if ((*p++ & 0xc0) == 0x80) charcount--;  #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
2773                if (utf)
2774                  {
2775                  const pcre_uchar *p = start_subject + local_offsets[rc];
2776                  const pcre_uchar *pp = start_subject + local_offsets[rc+1];
2777                  while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--;
2778                  }
2779    #endif
2780              if (charcount > 0)              if (charcount > 0)
2781                {                {
2782                ADD_NEW_DATA(-(state_offset + LINK_SIZE + 1), 0, (charcount - 1));                ADD_NEW_DATA(-(state_offset + LINK_SIZE + 1), 0, (charcount - 1));
# Line 2597  for (;;) Line 2792  for (;;)
2792        break;        break;
2793    
2794        /*-----------------------------------------------------------------*/        /*-----------------------------------------------------------------*/
2795          case OP_BRAPOS:
2796          case OP_SBRAPOS:
2797          case OP_CBRAPOS:
2798          case OP_SCBRAPOS:
2799          case OP_BRAPOSZERO:
2800            {
2801            int charcount, matched_count;
2802            const pcre_uchar *local_ptr = ptr;
2803            BOOL allow_zero;
2804    
2805            if (codevalue == OP_BRAPOSZERO)
2806              {
2807              allow_zero = TRUE;
2808              codevalue = *(++code);  /* Codevalue will be one of above BRAs */
2809              }
2810            else allow_zero = FALSE;
2811    
2812            /* Loop to match the subpattern as many times as possible as if it were
2813            a complete pattern. */
2814    
2815            for (matched_count = 0;; matched_count++)
2816              {
2817              int local_offsets[2];
2818              int local_workspace[1000];
2819    
2820              int rc = internal_dfa_exec(
2821                md,                                   /* fixed match data */
2822                code,                                 /* this subexpression's code */
2823                local_ptr,                            /* where we currently are */
2824                (int)(ptr - start_subject),           /* start offset */
2825                local_offsets,                        /* offset vector */
2826                sizeof(local_offsets)/sizeof(int),    /* size of same */
2827                local_workspace,                      /* workspace vector */
2828                sizeof(local_workspace)/sizeof(int),  /* size of same */
2829                rlevel);                              /* function recursion level */
2830    
2831              /* Failed to match */
2832    
2833              if (rc < 0)
2834                {
2835                if (rc != PCRE_ERROR_NOMATCH) return rc;
2836                break;
2837                }
2838    
2839              /* Matched: break the loop if zero characters matched. */
2840    
2841              charcount = local_offsets[1] - local_offsets[0];
2842              if (charcount == 0) break;
2843              local_ptr += charcount;    /* Advance temporary position ptr */
2844              }
2845    
2846            /* At this point we have matched the subpattern matched_count
2847            times, and local_ptr is pointing to the character after the end of the
2848            last match. */
2849    
2850            if (matched_count > 0 || allow_zero)
2851              {
2852              const pcre_uchar *end_subpattern = code;
2853              int next_state_offset;
2854    
2855              do { end_subpattern += GET(end_subpattern, 1); }
2856                while (*end_subpattern == OP_ALT);
2857              next_state_offset =
2858                (int)(end_subpattern - start_code + LINK_SIZE + 1);
2859    
2860              /* Optimization: if there are no more active states, and there
2861              are no new states yet set up, then skip over the subject string
2862              right here, to save looping. Otherwise, set up the new state to swing
2863              into action when the end of the matched substring is reached. */
2864    
2865              if (i + 1 >= active_count && new_count == 0)
2866                {
2867                ptr = local_ptr;
2868                clen = 0;
2869                ADD_NEW(next_state_offset, 0);
2870                }
2871              else
2872                {
2873                const pcre_uchar *p = ptr;
2874                const pcre_uchar *pp = local_ptr;
2875                charcount = (int)(pp - p);
2876    #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
2877                if (utf) while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--;
2878    #endif
2879                ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1));
2880                }
2881              }
2882            }
2883          break;
2884    
2885          /*-----------------------------------------------------------------*/
2886        case OP_ONCE:        case OP_ONCE:
2887          case OP_ONCE_NC:
2888          {          {
2889          int local_offsets[2];          int local_offsets[2];
2890          int local_workspace[1000];          int local_workspace[1000];
# Line 2611  for (;;) Line 2898  for (;;)
2898            sizeof(local_offsets)/sizeof(int),    /* size of same */            sizeof(local_offsets)/sizeof(int),    /* size of same */
2899            local_workspace,                      /* workspace vector */            local_workspace,                      /* workspace vector */
2900            sizeof(local_workspace)/sizeof(int),  /* size of same */            sizeof(local_workspace)/sizeof(int),  /* size of same */
2901            ims,                                  /* the current ims flags */            rlevel);                              /* function recursion level */
           rlevel,                               /* function recursion level */  
           recursing);                           /* pass on regex recursion */  
2902    
2903          if (rc >= 0)          if (rc >= 0)
2904            {            {
2905            const uschar *end_subpattern = code;            const pcre_uchar *end_subpattern = code;
2906            int charcount = local_offsets[1] - local_offsets[0];            int charcount = local_offsets[1] - local_offsets[0];
2907            int next_state_offset, repeat_state_offset;            int next_state_offset, repeat_state_offset;
2908    
2909            do { end_subpattern += GET(end_subpattern, 1); }            do { end_subpattern += GET(end_subpattern, 1); }
2910              while (*end_subpattern == OP_ALT);              while (*end_subpattern == OP_ALT);
2911            next_state_offset =            next_state_offset =
2912              (int)(end_subpattern - start_code + LINK_SIZE + 1);              (int)(end_subpattern - start_code + LINK_SIZE + 1);
2913    
2914            /* If the end of this subpattern is KETRMAX or KETRMIN, we must            /* If the end of this subpattern is KETRMAX or KETRMIN, we must
# Line 2647  for (;;) Line 2932  for (;;)
2932            /* Optimization: if there are no more active states, and there            /* Optimization: if there are no more active states, and there
2933            are no new states yet set up, then skip over the subject string            are no new states yet set up, then skip over the subject string
2934            right here, to save looping. Otherwise, set up the new state to swing            right here, to save looping. Otherwise, set up the new state to swing
2935            into action when the end of the substring is reached. */            into action when the end of the matched substring is reached. */
2936    
2937            else if (i + 1 >= active_count && new_count == 0)            else if (i + 1 >= active_count && new_count == 0)
2938              {              {
# Line 2670  for (;;) Line 2955  for (;;)
2955              }              }
2956            else            else
2957              {              {
2958              const uschar *p = start_subject + local_offsets[0];  #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
2959              const uschar *pp = start_subject + local_offsets[1];              if (utf)
2960              while (p < pp) if ((*p++ & 0xc0) == 0x80) charcount--;                {
2961                  const pcre_uchar *p = start_subject + local_offsets[0];
2962                  const pcre_uchar *pp = start_subject + local_offsets[1];
2963                  while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--;
2964                  }
2965    #endif
2966              ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1));              ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1));
2967              if (repeat_state_offset >= 0)              if (repeat_state_offset >= 0)
2968                { ADD_NEW_DATA(-repeat_state_offset, 0, (charcount - 1)); }                { ADD_NEW_DATA(-repeat_state_offset, 0, (charcount - 1)); }
2969              }              }
   
2970            }            }
2971          else if (rc != PCRE_ERROR_NOMATCH) return rc;          else if (rc != PCRE_ERROR_NOMATCH) return rc;
2972          }          }
# Line 2689  for (;;) Line 2978  for (;;)
2978    
2979        case OP_CALLOUT:        case OP_CALLOUT:
2980        rrc = 0;        rrc = 0;
2981        if (pcre_callout != NULL)        if (PUBL(callout) != NULL)
2982          {          {
2983          pcre_callout_block cb;          PUBL(callout_block) cb;
2984          cb.version          = 1;   /* Version 1 of the callout block */          cb.version          = 1;   /* Version 1 of the callout block */
2985          cb.callout_number   = code[1];          cb.callout_number   = code[1];
2986          cb.offset_vector    = offsets;          cb.offset_vector    = offsets;
2987    #if defined COMPILE_PCRE8
2988          cb.subject          = (PCRE_SPTR)start_subject;          cb.subject          = (PCRE_SPTR)start_subject;
2989    #elif defined COMPILE_PCRE16
2990            cb.subject          = (PCRE_SPTR16)start_subject;
2991    #elif defined COMPILE_PCRE32
2992            cb.subject          = (PCRE_SPTR32)start_subject;
2993    #endif
2994          cb.subject_length   = (int)(end_subject - start_subject);          cb.subject_length   = (int)(end_subject - start_subject);
2995          cb.start_match      = (int)(current_subject - start_subject);          cb.start_match      = (int)(current_subject - start_subject);
2996          cb.current_position = (int)(ptr - start_subject);          cb.current_position = (int)(ptr - start_subject);
# Line 2704  for (;;) Line 2999  for (;;)
2999          cb.capture_top      = 1;          cb.capture_top      = 1;
3000          cb.capture_last     = -1;          cb.capture_last     = -1;
3001          cb.callout_data     = md->callout_data;          cb.callout_data     = md->callout_data;
3002          if ((rrc = (*pcre_callout)(&cb)) < 0) return rrc;   /* Abandon */          cb.mark             = NULL;   /* No (*MARK) support */
3003            if ((rrc = (*PUBL(callout))(&cb)) < 0) return rrc;   /* Abandon */
3004          }          }
3005        if (rrc == 0)        if (rrc == 0)
3006          { ADD_ACTIVE(state_offset + _pcre_OP_lengths[OP_CALLOUT], 0); }          { ADD_ACTIVE(state_offset + PRIV(OP_lengths)[OP_CALLOUT], 0); }
3007        break;        break;
3008    
3009    
# Line 2736  for (;;) Line 3032  for (;;)
3032    if (new_count <= 0)    if (new_count <= 0)
3033      {      {
3034      if (rlevel == 1 &&                               /* Top level, and */      if (rlevel == 1 &&                               /* Top level, and */
3035          could_continue &&                            /* Some could go on */          could_continue &&                            /* Some could go on, and */
3036          forced_fail != workspace[1] &&               /* Not all forced fail & */          forced_fail != workspace[1] &&               /* Not all forced fail & */
3037          (                                            /* either... */          (                                            /* either... */
3038          (md->moptions & PCRE_PARTIAL_HARD) != 0      /* Hard partial */          (md->moptions & PCRE_PARTIAL_HARD) != 0      /* Hard partial */
# Line 2744  for (;;) Line 3040  for (;;)
3040          ((md->moptions & PCRE_PARTIAL_SOFT) != 0 &&  /* Soft partial and */          ((md->moptions & PCRE_PARTIAL_SOFT) != 0 &&  /* Soft partial and */
3041           match_count < 0)                            /* no matches */           match_count < 0)                            /* no matches */
3042          ) &&                                         /* And... */          ) &&                                         /* And... */
3043          ptr >= end_subject &&                     /* Reached end of subject */          (
3044          ptr > current_subject)                    /* Matched non-empty string */          partial_newline ||                           /* Either partial NL */
3045        {            (                                          /* or ... */
3046        if (offsetcount >= 2)            ptr >= end_subject &&                /* End of subject and */
3047          {            ptr > md->start_used_ptr)            /* Inspected non-empty string */
3048          offsets[0] = (int)(md->start_used_ptr - start_subject);            )
3049          offsets[1] = (int)(end_subject - start_subject);          )
         }  
3050        match_count = PCRE_ERROR_PARTIAL;        match_count = PCRE_ERROR_PARTIAL;
       }  
   
3051      DPRINTF(("%.*sEnd of internal_dfa_exec %d: returning %d\n"      DPRINTF(("%.*sEnd of internal_dfa_exec %d: returning %d\n"
3052        "%.*s---------------------\n\n", rlevel*2-2, SP, rlevel, match_count,        "%.*s---------------------\n\n", rlevel*2-2, SP, rlevel, match_count,
3053        rlevel*2-2, SP));        rlevel*2-2, SP));
# Line 2804  Returns:          > 0 => number of match Line 3097  Returns:          > 0 => number of match
3097                   < -1 => some kind of unexpected problem                   < -1 => some kind of unexpected problem
3098  */  */
3099    
3100    #if defined COMPILE_PCRE8
3101  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION  PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
3102  pcre_dfa_exec(const pcre *argument_re, const pcre_extra *extra_data,  pcre_dfa_exec(const pcre *argument_re, const pcre_extra *extra_data,
3103    const char *subject, int length, int start_offset, int options, int *offsets,    const char *subject, int length, int start_offset, int options, int *offsets,
3104    int offsetcount, int *workspace, int wscount)    int offsetcount, int *workspace, int wscount)
3105    #elif defined COMPILE_PCRE16
3106    PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
3107    pcre16_dfa_exec(const pcre16 *argument_re, const pcre16_extra *extra_data,
3108      PCRE_SPTR16 subject, int length, int start_offset, int options, int *offsets,
3109      int offsetcount, int *workspace, int wscount)
3110    #elif defined COMPILE_PCRE32
3111    PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
3112    pcre32_dfa_exec(const pcre32 *argument_re, const pcre32_extra *extra_data,
3113      PCRE_SPTR32 subject, int length, int start_offset, int options, int *offsets,
3114      int offsetcount, int *workspace, int wscount)
3115    #endif
3116  {  {
3117  real_pcre *re = (real_pcre *)argument_re;  REAL_PCRE *re = (REAL_PCRE *)argument_re;
3118  dfa_match_data match_block;  dfa_match_data match_block;
3119  dfa_match_data *md = &match_block;  dfa_match_data *md = &match_block;
3120  BOOL utf8, anchored, startline, firstline;  BOOL utf, anchored, startline, firstline;
3121  const uschar *current_subject, *end_subject, *lcc;  const pcre_uchar *current_subject, *end_subject;
   
 pcre_study_data internal_study;  
3122  const pcre_study_data *study = NULL;  const pcre_study_data *study = NULL;
 real_pcre internal_re;  
3123    
3124  const uschar *req_byte_ptr;  const pcre_uchar *req_char_ptr;
3125  const uschar *start_bits = NULL;  const pcre_uint8 *start_bits = NULL;
3126  BOOL first_byte_caseless = FALSE;  BOOL has_first_char = FALSE;
3127  BOOL req_byte_caseless = FALSE;  BOOL has_req_char = FALSE;
3128  int first_byte = -1;  pcre_uchar first_char = 0;
3129  int req_byte = -1;  pcre_uchar first_char2 = 0;
3130  int req_byte2 = -1;  pcre_uchar req_char = 0;
3131    pcre_uchar req_char2 = 0;
3132  int newline;  int newline;
3133    
3134  /* Plausibility checks */  /* Plausibility checks */
# Line 2835  if (re == NULL || subject == NULL || wor Line 3138  if (re == NULL || subject == NULL || wor
3138     (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL;     (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL;
3139  if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;  if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;
3140  if (wscount < 20) return PCRE_ERROR_DFA_WSSIZE;  if (wscount < 20) return PCRE_ERROR_DFA_WSSIZE;
3141    if (length < 0) return PCRE_ERROR_BADLENGTH;
3142    if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET;
3143    
3144  /* We need to find the pointer to any study data before we test for byte  /* Check that the first field in the block is the magic number. If it is not,
3145  flipping, so we scan the extra_data block first. This may set two fields in the  return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to
3146  match block, so we must initialize them beforehand. However, the other fields  REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which
3147  in the match block must not be set until after the byte flipping. */  means that the pattern is likely compiled with different endianness. */
3148    
3149    if (re->magic_number != MAGIC_NUMBER)
3150      return re->magic_number == REVERSED_MAGIC_NUMBER?
3151        PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC;
3152    if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
3153    
3154    /* If restarting after a partial match, do some sanity checks on the contents
3155    of the workspace. */
3156    
3157    if ((options & PCRE_DFA_RESTART) != 0)
3158      {
3159      if ((workspace[0] & (-2)) != 0 || workspace[1] < 1 ||
3160        workspace[1] > (wscount - 2)/INTS_PER_STATEBLOCK)
3161          return PCRE_ERROR_DFA_BADRESTART;
3162      }
3163    
3164    /* Set up study, callout, and table data */
3165    
3166  md->tables = re->tables;  md->tables = re->tables;
3167  md->callout_data = NULL;  md->callout_data = NULL;
# Line 2858  if (extra_data != NULL) Line 3180  if (extra_data != NULL)
3180      md->tables = extra_data->tables;      md->tables = extra_data->tables;
3181    }    }
3182    
 /* Check that the first field in the block is the magic number. If it is not,  
 test for a regex that was compiled on a host of opposite endianness. If this is  
 the case, flipped values are put in internal_re and internal_study if there was  
 study data too. */  
   
 if (re->magic_number != MAGIC_NUMBER)  
   {  
   re = _pcre_try_flipped(re, &internal_re, study, &internal_study);  
   if (re == NULL) return PCRE_ERROR_BADMAGIC;  
   if (study != NULL) study = &internal_study;  
   }  
   
3183  /* Set some local values */  /* Set some local values */
3184    
3185  current_subject = (const unsigned char *)subject + start_offset;  current_subject = (const pcre_uchar *)subject + start_offset;
3186  end_subject = (const unsigned char *)subject + length;  end_subject = (const pcre_uchar *)subject + length;
3187  req_byte_ptr = current_subject - 1;  req_char_ptr = current_subject - 1;
3188    
3189  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3190  utf8 = (re->options & PCRE_UTF8) != 0;  /* PCRE_UTF(16|32) have the same value as PCRE_UTF8. */
3191    utf = (re->options & PCRE_UTF8) != 0;
3192  #else  #else
3193  utf8 = FALSE;  utf = FALSE;
3194  #endif  #endif
3195    
3196  anchored = (options & (PCRE_ANCHORED|PCRE_DFA_RESTART)) != 0 ||  anchored = (options & (PCRE_ANCHORED|PCRE_DFA_RESTART)) != 0 ||
# Line 2887  anchored = (options & (PCRE_ANCHORED|PCR Line 3198  anchored = (options & (PCRE_ANCHORED|PCR
3198    
3199  /* The remaining fixed data for passing around. */  /* The remaining fixed data for passing around. */
3200    
3201  md->start_code = (const uschar *)argument_re +  md->start_code = (const pcre_uchar *)argument_re +
3202      re->name_table_offset + re->name_count * re->name_entry_size;      re->name_table_offset + re->name_count * re->name_entry_size;
3203  md->start_subject = (const unsigned char *)subject;  md->start_subject = (const pcre_uchar *)subject;
3204  md->end_subject = end_subject;  md->end_subject = end_subject;
3205  md->start_offset = start_offset;  md->start_offset = start_offset;
3206  md->moptions = options;  md->moptions = options;
# Line 2950  else Line 3261  else
3261  /* Check a UTF-8 string if required. Unfortunately there's no way of passing  /* Check a UTF-8 string if required. Unfortunately there's no way of passing
3262  back the character offset. */  back the character offset. */
3263    
3264  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3265  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0)  if (utf && (options & PCRE_NO_UTF8_CHECK) == 0)
3266    {    {
3267    if (_pcre_valid_utf8((uschar *)subject, length) >= 0)    int erroroffset;
3268      return PCRE_ERROR_BADUTF8;    int errorcode = PRIV(valid_utf)((pcre_uchar *)subject, length, &erroroffset);
3269    if (start_offset > 0 && start_offset < length)    if (errorcode != 0)
3270      {      {
3271      int tb = ((uschar *)subject)[start_offset];      if (offsetcount >= 2)
     if (tb > 127)  
3272        {        {
3273        tb &= 0xc0;        offsets[0] = erroroffset;
3274        if (tb != 0 && tb != 0xc0) return PCRE_ERROR_BADUTF8_OFFSET;        offsets[1] = errorcode;
3275        }        }
3276    #if defined COMPILE_PCRE8
3277        return (errorcode <= PCRE_UTF8_ERR5 && (options & PCRE_PARTIAL_HARD) != 0) ?
3278          PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8;
3279    #elif defined COMPILE_PCRE16
3280        return (errorcode <= PCRE_UTF16_ERR1 && (options & PCRE_PARTIAL_HARD) != 0) ?
3281          PCRE_ERROR_SHORTUTF16 : PCRE_ERROR_BADUTF16;
3282    #elif defined COMPILE_PCRE32
3283        return PCRE_ERROR_BADUTF32;
3284    #endif
3285      }      }
3286    #if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
3287      if (start_offset > 0 && start_offset < length &&
3288            NOT_FIRSTCHAR(((PCRE_PUCHAR)subject)[start_offset]))
3289        return PCRE_ERROR_BADUTF8_OFFSET;
3290    #endif
3291    }    }
3292  #endif  #endif
3293    
# Line 2971  if (utf8 && (options & PCRE_NO_UTF8_CHEC Line 3295  if (utf8 && (options & PCRE_NO_UTF8_CHEC
3295  is a feature that makes it possible to save compiled regex and re-use them  is a feature that makes it possible to save compiled regex and re-use them
3296  in other programs later. */  in other programs later. */
3297    
3298  if (md->tables == NULL) md->tables = _pcre_default_tables;  if (md->tables == NULL) md->tables = PRIV(default_tables);
3299    
3300  /* The lower casing table and the "must be at the start of a line" flag are  /* The "must be at the start of a line" flags are used in a loop when finding
3301  used in a loop when finding where to start. */  where to start. */
3302    
 lcc = md->tables + lcc_offset;  
3303  startline = (re->flags & PCRE_STARTLINE) != 0;  startline = (re->flags & PCRE_STARTLINE) != 0;
3304  firstline = (re->options & PCRE_FIRSTLINE) != 0;  firstline = (re->options & PCRE_FIRSTLINE) != 0;
3305    
# Line 2990  if (!anchored) Line 3313  if (!anchored)
3313    {    {
3314    if ((re->flags & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
3315      {      {
3316      first_byte = re->first_byte & 255;      has_first_char = TRUE;
3317      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)      first_char = first_char2 = (pcre_uchar)(re->first_char);
3318        first_byte = lcc[first_byte];      if ((re->flags & PCRE_FCH_CASELESS) != 0)
3319          {
3320          first_char2 = TABLE_GET(first_char, md->tables + fcc_offset, first_char);
3321    #if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
3322          if (utf && first_char > 127)
3323            first_char2 = UCD_OTHERCASE(first_char);
3324    #endif
3325          }
3326      }      }
3327    else    else
3328      {      {
# Line 3007  character" set. */ Line 3337  character" set. */
3337    
3338  if ((re->flags & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
3339    {    {
3340    req_byte = re->req_byte & 255;    has_req_char = TRUE;
3341    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;    req_char = req_char2 = (pcre_uchar)(re->req_char);
3342    req_byte2 = (md->tables + fcc_offset)[req_byte];  /* case flipped */    if ((re->flags & PCRE_RCH_CASELESS) != 0)
3343        {
3344        req_char2 = TABLE_GET(req_char, md->tables + fcc_offset, req_char);
3345    #if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
3346        if (utf && req_char > 127)
3347          req_char2 = UCD_OTHERCASE(req_char);
3348    #endif
3349        }
3350    }    }
3351    
3352  /* Call the main matching function, looping for a non-anchored regex after a  /* Call the main matching function, looping for a non-anchored regex after a
# Line 3022  for (;;) Line 3359  for (;;)
3359    
3360    if ((options & PCRE_DFA_RESTART) == 0)    if ((options & PCRE_DFA_RESTART) == 0)
3361      {      {
3362      const uschar *save_end_subject = end_subject;      const pcre_uchar *save_end_subject = end_subject;
3363    
3364      /* If firstline is TRUE, the start of the match is constrained to the first      /* If firstline is TRUE, the start of the match is constrained to the first
3365      line of a multiline string. Implement this by temporarily adjusting      line of a multiline string. Implement this by temporarily adjusting
# Line 3031  for (;;) Line 3368  for (;;)
3368    
3369      if (firstline)      if (firstline)
3370        {        {
3371        USPTR t = current_subject;        PCRE_PUCHAR t = current_subject;
3372  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3373        if (utf8)        if (utf)
3374          {          {
3375          while (t < md->end_subject && !IS_NEWLINE(t))          while (t < md->end_subject && !IS_NEWLINE(t))
3376            {            {
3377            t++;            t++;
3378            while (t < end_subject && (*t & 0xc0) == 0x80) t++;            ACROSSCHAR(t < end_subject, *t, t++);
3379            }            }
3380          }          }
3381        else        else
# Line 3049  for (;;) Line 3386  for (;;)
3386    
3387      /* There are some optimizations that avoid running the match if a known      /* There are some optimizations that avoid running the match if a known
3388      starting point is not found. However, there is an option that disables      starting point is not found. However, there is an option that disables
3389      these, for testing and for ensuring that all callouts do actually occur. */      these, for testing and for ensuring that all callouts do actually occur.
3390        The option can be set in the regex by (*NO_START_OPT) or passed in
3391        match-time options. */
3392    
3393      if ((options & PCRE_NO_START_OPTIMIZE) == 0)      if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0)
3394        {        {
3395        /* Advance to a known first byte. */        /* Advance to a known first char. */
3396    
3397        if (first_byte >= 0)        if (has_first_char)
3398          {          {
3399          if (first_byte_caseless)          if (first_char != first_char2)
3400              {
3401              pcre_uchar csc;
3402            while (current_subject < end_subject &&            while (current_subject < end_subject &&
3403                   lcc[*current_subject] != first_byte)                   (csc = RAWUCHARTEST(current_subject)) != first_char && csc != first_char2)
3404              current_subject++;              current_subject++;
3405              }
3406          else          else
3407            while (current_subject < end_subject &&            while (current_subject < end_subject &&
3408                   *current_subject != first_byte)                   RAWUCHARTEST(current_subject) != first_char)
3409              current_subject++;              current_subject++;
3410          }          }
3411    
# Line 3073  for (;;) Line 3415  for (;;)
3415          {          {
3416          if (current_subject > md->start_subject + start_offset)          if (current_subject > md->start_subject + start_offset)
3417            {            {
3418  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF
3419            if (utf8)            if (utf)
3420              {              {
3421              while (current_subject < end_subject &&              while (current_subject < end_subject &&
3422                     !WAS_NEWLINE(current_subject))                     !WAS_NEWLINE(current_subject))
3423                {                {
3424                current_subject++;                current_subject++;
3425                while(current_subject < end_subject &&                ACROSSCHAR(current_subject < end_subject, *current_subject,
3426                      (*current_subject & 0xc0) == 0x80)                  current_subject++);
                 current_subject++;  
3427                }                }
3428              }              }
3429            else            else
# Line 3094  for (;;) Line 3435  for (;;)
3435            ANYCRLF, and we are now at a LF, advance the match position by one            ANYCRLF, and we are now at a LF, advance the match position by one
3436            more character. */            more character. */
3437    
3438            if (current_subject[-1] == CHAR_CR &&            if (RAWUCHARTEST(current_subject - 1) == CHAR_CR &&
3439                 (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&                 (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
3440                 current_subject < end_subject &&                 current_subject < end_subject &&
3441                 *current_subject == CHAR_NL)                 RAWUCHARTEST(current_subject) == CHAR_NL)
3442              current_subject++;              current_subject++;
3443            }            }
3444          }          }
# Line 3108  for (;;) Line 3449  for (;;)
3449          {          {
3450          while (current_subject < end_subject)          while (current_subject < end_subject)
3451            {            {
3452            register unsigned int c = *current_subject;            register pcre_uint32 c = RAWUCHARTEST(current_subject);
3453            if ((start_bits[c/8] & (1 << (c&7))) == 0) current_subject++;  #ifndef COMPILE_PCRE8
3454              else break;            if (c > 255) c = 255;
3455    #endif
3456              if ((start_bits[c/8] & (1 << (c&7))) == 0)
3457                {
3458                current_subject++;
3459    #if defined SUPPORT_UTF && defined COMPILE_PCRE8
3460                /* In non 8-bit mode, the iteration will stop for
3461                characters > 255 at the beginning or not stop at all. */
3462                if (utf)
3463                  ACROSSCHAR(current_subject < end_subject, *current_subject,
3464                    current_subject++);
3465    #endif
3466                }
3467              else break;
3468            }            }
3469          }          }
3470        }        }
# Line 3123  for (;;) Line 3477  for (;;)
3477      disabling is explicitly requested (and of course, by the test above, this      disabling is explicitly requested (and of course, by the test above, this
3478      code is not obeyed when restarting after a partial match). */      code is not obeyed when restarting after a partial match). */
3479    
3480      if ((options & PCRE_NO_START_OPTIMIZE) == 0 &&      if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0 &&
3481          (options & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) == 0)          (options & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) == 0)
3482        {        {
3483        /* If the pattern was studied, a minimum subject length may be set. This        /* If the pattern was studied, a minimum subject length may be set. This
# Line 3135  for (;;) Line 3489  for (;;)
3489            (pcre_uint32)(end_subject - current_subject) < study->minlength)            (pcre_uint32)(end_subject - current_subject) < study->minlength)
3490          return PCRE_ERROR_NOMATCH;          return PCRE_ERROR_NOMATCH;
3491    
3492        /* If req_byte is set, we know that that character must appear in the        /* If req_char is set, we know that that character must appear in the
3493        subject for the match to succeed. If the first character is set, req_byte        subject for the match to succeed. If the first character is set, req_char
3494        must be later in the subject; otherwise the test starts at the match        must be later in the subject; otherwise the test starts at the match
3495        point. This optimization can save a huge amount of work in patterns with        point. This optimization can save a huge amount of work in patterns with
3496        nested unlimited repeats that aren't going to match. Writing separate        nested unlimited repeats that aren't going to match. Writing separate
# Line 3148  for (;;) Line 3502  for (;;)
3502        patterns. This showed up when somebody was matching /^C/ on a 32-megabyte        patterns. This showed up when somebody was matching /^C/ on a 32-megabyte
3503        string... so we don't do this when the string is sufficiently long. */        string... so we don't do this when the string is sufficiently long. */
3504    
3505        if (req_byte >= 0 && end_subject - current_subject < REQ_BYTE_MAX)        if (has_req_char && end_subject - current_subject < REQ_BYTE_MAX)
3506          {          {
3507          register const uschar *p = current_subject + ((first_byte >= 0)? 1 : 0);          register PCRE_PUCHAR p = current_subject + (has_first_char? 1:0);
3508    
3509          /* We don't need to repeat the search if we haven't yet reached the          /* We don't need to repeat the search if we haven't yet reached the
3510          place we found it at last time. */          place we found it at last time. */
3511    
3512          if (p > req_byte_ptr)          if (p > req_char_ptr)
3513            {            {
3514            if (req_byte_caseless)            if (req_char != req_char2)
3515              {              {
3516              while (p < end_subject)              while (p < end_subject)
3517                {                {
3518                register int pp = *p++;                register pcre_uint32 pp = RAWUCHARINCTEST(p);
3519                if (pp == req_byte || pp == req_byte2) { p--; break; }                if (pp == req_char || pp == req_char2) { p--; break; }
3520                }                }
3521              }              }
3522            else            else
3523              {              {
3524              while (p < end_subject)              while (p < end_subject)
3525                {                {
3526                if (*p++ == req_byte) { p--; break; }                if (RAWUCHARINCTEST(p) == req_char) { p--; break; }
3527                }                }
3528              }              }
3529    
# Line 3182  for (;;) Line 3536  for (;;)
3536            found it, so that we don't search again next time round the loop if            found it, so that we don't search again next time round the loop if
3537            the start hasn't passed this character yet. */            the start hasn't passed this character yet. */
3538    
3539            req_byte_ptr = p;            req_char_ptr = p;
3540            }            }
3541          }          }
3542        }        }
# Line 3191  for (;;) Line 3545  for (;;)
3545    /* OK, now we can do the business */    /* OK, now we can do the business */
3546    
3547    md->start_used_ptr = current_subject;    md->start_used_ptr = current_subject;
3548      md->recursive = NULL;
3549    
3550    rc = internal_dfa_exec(    rc = internal_dfa_exec(
3551      md,                                /* fixed match data */      md,                                /* fixed match data */
# Line 3201  for (;;) Line 3556  for (;;)
3556      offsetcount,                       /* size of same */      offsetcount,                       /* size of same */
3557      workspace,                         /* workspace vector */      workspace,                         /* workspace vector */
3558      wscount,                           /* size of same */      wscount,                           /* size of same */
3559      re->options & (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL), /* ims flags */      0);                                /* function recurse level */
     0,                                 /* function recurse level */  
     0);                                /* regex recurse level */  
3560    
3561    /* Anything other than "no match" means we are done, always; otherwise, carry    /* Anything other than "no match" means we are done, always; otherwise, carry
3562    on only if not anchored. */    on only if not anchored. */
3563    
3564    if (rc != PCRE_ERROR_NOMATCH || anchored) return rc;    if (rc != PCRE_ERROR_NOMATCH || anchored)
3565        {
3566        if (rc == PCRE_ERROR_PARTIAL && offsetcount >= 2)
3567          {
3568          offsets[0] = (int)(md->start_used_ptr - (PCRE_PUCHAR)subject);
3569          offsets[1] = (int)(end_subject - (PCRE_PUCHAR)subject);
3570          if (offsetcount > 2)
3571            offsets[2] = (int)(current_subject - (PCRE_PUCHAR)subject);
3572          }
3573        return rc;
3574        }
3575    
3576    /* Advance to the next subject character unless we are at the end of a line    /* Advance to the next subject character unless we are at the end of a line
3577    and firstline is set. */    and firstline is set. */
3578    
3579    if (firstline && IS_NEWLINE(current_subject)) break;    if (firstline && IS_NEWLINE(current_subject)) break;
3580    current_subject++;    current_subject++;
3581    if (utf8)  #ifdef SUPPORT_UTF
3582      if (utf)
3583      {      {
3584      while (current_subject < end_subject && (*current_subject & 0xc0) == 0x80)      ACROSSCHAR(current_subject < end_subject, *current_subject,
3585        current_subject++;        current_subject++);
3586      }      }
3587    #endif
3588    if (current_subject > end_subject) break;    if (current_subject > end_subject) break;
3589    
3590    /* If we have just passed a CR and we are now at a LF, and the pattern does    /* If we have just passed a CR and we are now at a LF, and the pattern does
3591    not contain any explicit matches for \r or \n, and the newline option is CRLF    not contain any explicit matches for \r or \n, and the newline option is CRLF
3592    or ANY or ANYCRLF, advance the match position by one more character. */    or ANY or ANYCRLF, advance the match position by one more character. */
3593    
3594    if (current_subject[-1] == CHAR_CR &&    if (RAWUCHARTEST(current_subject - 1) == CHAR_CR &&
3595        current_subject < end_subject &&        current_subject < end_subject &&
3596        *current_subject == CHAR_NL &&        RAWUCHARTEST(current_subject) == CHAR_NL &&
3597        (re->flags & PCRE_HASCRORLF) == 0 &&        (re->flags & PCRE_HASCRORLF) == 0 &&
3598          (md->nltype == NLTYPE_ANY ||          (md->nltype == NLTYPE_ANY ||
3599           md->nltype == NLTYPE_ANYCRLF ||           md->nltype == NLTYPE_ANYCRLF ||

Legend:
Removed from v.530  
changed lines
  Added in v.1260

  ViewVC Help
Powered by ViewVC 1.1.5