/[pcre]/code/trunk/pcre_internal.h
ViewVC logotype

Diff of /code/trunk/pcre_internal.h

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

revision 286 by ph10, Mon Dec 17 14:46:11 2007 UTC revision 1128 by chpe, Thu Oct 18 18:35:08 2012 UTC
# Line 7  Line 7 
7  and semantics are as close as possible to those of the Perl 5 language.  and semantics are as close as possible to those of the Perl 5 language.
8    
9                         Written by Philip Hazel                         Written by Philip Hazel
10             Copyright (c) 1997-2007 University of Cambridge             Copyright (c) 1997-2012 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 40  POSSIBILITY OF SUCH DAMAGE. Line 40  POSSIBILITY OF SUCH DAMAGE.
40    
41  /* This header contains definitions that are shared between the different  /* This header contains definitions that are shared between the different
42  modules, but which are not relevant to the exported API. This includes some  modules, but which are not relevant to the exported API. This includes some
43  functions whose names all begin with "_pcre_". */  functions whose names all begin with "_pcre_", "_pcre16_" or "_pcre32_"
44    depending on the PRIV macro. */
45    
46  #ifndef PCRE_INTERNAL_H  #ifndef PCRE_INTERNAL_H
47  #define PCRE_INTERNAL_H  #define PCRE_INTERNAL_H
48    
49  /* Define DEBUG to get debugging output on stdout. */  /* Define PCRE_DEBUG to get debugging output on stdout. */
50    
51  #if 0  #if 0
52  #define DEBUG  #define PCRE_DEBUG
53    #endif
54    
55    /* PCRE is compiled as an 8 bit library if it is not requested otherwise. */
56    #if !defined COMPILE_PCRE16 && !defined COMPILE_PCRE32
57    #define COMPILE_PCRE8
58    #endif
59    
60    /* If SUPPORT_UCP is defined, SUPPORT_UTF must also be defined. The
61    "configure" script ensures this, but not everybody uses "configure". */
62    
63    #if defined SUPPORT_UCP && !(defined SUPPORT_UTF)
64    #define SUPPORT_UTF 1
65    #endif
66    
67    /* We define SUPPORT_UTF if SUPPORT_UTF8 is enabled for compatibility
68    reasons with existing code. */
69    
70    #if defined SUPPORT_UTF8 && !(defined SUPPORT_UTF)
71    #define SUPPORT_UTF 1
72    #endif
73    
74    /* Fixme: SUPPORT_UTF8 should be eventually disappear from the code.
75    Until then we define it if SUPPORT_UTF is defined. */
76    
77    #if defined SUPPORT_UTF && !(defined SUPPORT_UTF8)
78    #define SUPPORT_UTF8 1
79    #endif
80    
81    /* We do not support both EBCDIC and UTF-8/16/32 at the same time. The "configure"
82    script prevents both being selected, but not everybody uses "configure". */
83    
84    #if defined EBCDIC && defined SUPPORT_UTF
85    #error The use of both EBCDIC and SUPPORT_UTF is not supported.
86  #endif  #endif
87    
88  /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef  /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
# Line 60  It turns out that the Mac Debugging.h he Line 94  It turns out that the Mac Debugging.h he
94  be absolutely sure we get our version. */  be absolutely sure we get our version. */
95    
96  #undef DPRINTF  #undef DPRINTF
97  #ifdef DEBUG  #ifdef PCRE_DEBUG
98  #define DPRINTF(p) printf p  #define DPRINTF(p) printf p
99  #else  #else
100  #define DPRINTF(p) /* Nothing */  #define DPRINTF(p) /* Nothing */
# Line 72  setjmp and stdarg are used is when NO_RE Line 106  setjmp and stdarg are used is when NO_RE
106    
107  #include <ctype.h>  #include <ctype.h>
108  #include <limits.h>  #include <limits.h>
 #include <setjmp.h>  
 #include <stdarg.h>  
109  #include <stddef.h>  #include <stddef.h>
110  #include <stdio.h>  #include <stdio.h>
111  #include <stdlib.h>  #include <stdlib.h>
# Line 132  PCRE_EXP_DATA_DEFN only if they are not Line 164  PCRE_EXP_DATA_DEFN only if they are not
164  #  endif  #  endif
165  #endif  #endif
166    
167  /* We need to have types that specify unsigned 16-bit and 32-bit integers. We  /* When compiling with the MSVC compiler, it is sometimes necessary to include
168    a "calling convention" before exported function names. (This is secondhand
169    information; I know nothing about MSVC myself). For example, something like
170    
171      void __cdecl function(....)
172    
173    might be needed. In order so make this easy, all the exported functions have
174    PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
175    set, we ensure here that it has no effect. */
176    
177    #ifndef PCRE_CALL_CONVENTION
178    #define PCRE_CALL_CONVENTION
179    #endif
180    
181    /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
182  cannot determine these outside the compilation (e.g. by running a program as  cannot determine these outside the compilation (e.g. by running a program as
183  part of "configure") because PCRE is often cross-compiled for use on other  part of "configure") because PCRE is often cross-compiled for use on other
184  systems. Instead we make use of the maximum sizes that are available at  systems. Instead we make use of the maximum sizes that are available at
185  preprocessor time in standard C environments. */  preprocessor time in standard C environments. */
186    
187    typedef unsigned char pcre_uint8;
188    
189  #if USHRT_MAX == 65535  #if USHRT_MAX == 65535
190    typedef unsigned short pcre_uint16;    typedef unsigned short pcre_uint16;
191      typedef short pcre_int16;
192  #elif UINT_MAX == 65535  #elif UINT_MAX == 65535
193    typedef unsigned int pcre_uint16;    typedef unsigned int pcre_uint16;
194      typedef int pcre_int16;
195  #else  #else
196    #error Cannot determine a type for 16-bit unsigned integers    #error Cannot determine a type for 16-bit unsigned integers
197  #endif  #endif
198    
199  #if UINT_MAX == 4294967295  #if UINT_MAX == 4294967295
200    typedef unsigned int pcre_uint32;    typedef unsigned int pcre_uint32;
201      typedef int pcre_int32;
202  #elif ULONG_MAX == 4294967295  #elif ULONG_MAX == 4294967295
203    typedef unsigned long int pcre_uint32;    typedef unsigned long int pcre_uint32;
204      typedef long int pcre_int32;
205  #else  #else
206    #error Cannot determine a type for 32-bit unsigned integers    #error Cannot determine a type for 32-bit unsigned integers
207  #endif  #endif
208    
209    /* When checking for integer overflow in pcre_compile(), we need to handle
210    large integers. If a 64-bit integer type is available, we can use that.
211    Otherwise we have to cast to double, which of course requires floating point
212    arithmetic. Handle this by defining a macro for the appropriate type. If
213    stdint.h is available, include it; it may define INT64_MAX. Systems that do not
214    have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
215    by "configure". */
216    
217    #if HAVE_STDINT_H
218    #include <stdint.h>
219    #elif HAVE_INTTYPES_H
220    #include <inttypes.h>
221    #endif
222    
223    #if defined INT64_MAX || defined int64_t
224    #define INT64_OR_DOUBLE int64_t
225    #else
226    #define INT64_OR_DOUBLE double
227    #endif
228    
229  /* All character handling must be done as unsigned characters. Otherwise there  /* All character handling must be done as unsigned characters. Otherwise there
230  are problems with top-bit-set characters and functions such as isspace().  are problems with top-bit-set characters and functions such as isspace().
231  However, we leave the interface to the outside world as char *, because that  However, we leave the interface to the outside world as char * or short *,
232  should make things easier for callers. We define a short type for unsigned char  because that should make things easier for callers. This character type is
233  to save lots of typing. I tried "uchar", but it causes problems on Digital  called pcre_uchar.
234  Unix, where it is defined in sys/types, so use "uschar" instead. */  
235    The IN_UCHARS macro multiply its argument with the byte size of the current
236    pcre_uchar type. Useful for memcpy and such operations, whose require the
237    byte size of their input/output buffers.
238    
239    The MAX_255 macro checks whether its pcre_uchar input is less than 256.
240    
241    The TABLE_GET macro is designed for accessing elements of tables whose contain
242    exactly 256 items. When the character is able to contain more than 256
243    items, some check is needed before accessing these tables.
244    */
245    
246    #if defined COMPILE_PCRE8
247    
248    typedef unsigned char pcre_uchar;
249    #define IN_UCHARS(x) (x)
250    #define MAX_255(c) 1
251    #define TABLE_GET(c, table, default) ((table)[c])
252    
253    #elif defined COMPILE_PCRE16
254    
255    #if USHRT_MAX != 65535
256    /* This is a warning message. Change PCRE_UCHAR16 to a 16 bit data type in
257    pcre.h(.in) and disable (comment out) this message. */
258    #error Warning: PCRE_UCHAR16 is not a 16 bit data type.
259    #endif
260    
261    typedef pcre_uint16 pcre_uchar;
262    #define UCHAR_SHIFT (1)
263    #define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
264    #define MAX_255(c) ((c) <= 255u)
265    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
266    
267    #elif defined COMPILE_PCRE32
268    
269    typedef pcre_uint32 pcre_uchar;
270    #define UCHAR_SHIFT (2)
271    #define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
272    #define MAX_255(c) ((c) <= 255u)
273    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
274    
275  typedef unsigned char uschar;  /* Assert that pcre_uchar32 is a 32-bit type */
276    typedef int __assert_pcre_uchar32_size[sizeof(pcre_uchar) == 4 ? 1 : -1];
277    
278    #else
279    #error Unsupported compiling mode
280    #endif /* COMPILE_PCRE[8|16|32] */
281    
282  /* This is an unsigned int value that no character can ever have. UTF-8  /* This is an unsigned int value that no character can ever have. UTF-8
283  characters only go up to 0x7fffffff (though Unicode doesn't go beyond  characters only go up to 0x7fffffff (though Unicode doesn't go beyond
# Line 184  start/end of string field names are. */ Line 300  start/end of string field names are. */
300  #define IS_NEWLINE(p) \  #define IS_NEWLINE(p) \
301    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
302      ((p) < NLBLOCK->PSEND && \      ((p) < NLBLOCK->PSEND && \
303       _pcre_is_newline((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\       PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
304         utf8)) \         &(NLBLOCK->nllen), utf)) \
305      : \      : \
306      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
307       (p)[0] == NLBLOCK->nl[0] && \       RAWUCHARTEST(p) == NLBLOCK->nl[0] && \
308       (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \       (NLBLOCK->nllen == 1 || RAWUCHARTEST(p+1) == NLBLOCK->nl[1])       \
309      ) \      ) \
310    )    )
311    
# Line 198  start/end of string field names are. */ Line 314  start/end of string field names are. */
314  #define WAS_NEWLINE(p) \  #define WAS_NEWLINE(p) \
315    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
316      ((p) > NLBLOCK->PSSTART && \      ((p) > NLBLOCK->PSSTART && \
317       _pcre_was_newline((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \       PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
318         &(NLBLOCK->nllen), utf8)) \         &(NLBLOCK->nllen), utf)) \
319      : \      : \
320      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
321       (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \       RAWUCHARTEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] &&              \
322       (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \       (NLBLOCK->nllen == 1 || RAWUCHARTEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \
323      ) \      ) \
324    )    )
325    
# Line 217  used for the external interface and appe Line 333  used for the external interface and appe
333  must begin with PCRE_. */  must begin with PCRE_. */
334    
335  #ifdef CUSTOM_SUBJECT_PTR  #ifdef CUSTOM_SUBJECT_PTR
336  #define PCRE_SPTR CUSTOM_SUBJECT_PTR  #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
 #define USPTR CUSTOM_SUBJECT_PTR  
337  #else  #else
338  #define PCRE_SPTR const char *  #define PCRE_PUCHAR const pcre_uchar *
 #define USPTR const unsigned char *  
339  #endif  #endif
340    
   
   
341  /* Include the public PCRE header and the definitions of UCP character property  /* Include the public PCRE header and the definitions of UCP character property
342  values. */  values. */
343    
# Line 293  The macros are controlled by the value o Line 405  The macros are controlled by the value o
405  the config.h file, but can be overridden by using -D on the command line. This  the config.h file, but can be overridden by using -D on the command line. This
406  is automated on Unix systems via the "configure" command. */  is automated on Unix systems via the "configure" command. */
407    
408    #if defined COMPILE_PCRE8
409    
410  #if LINK_SIZE == 2  #if LINK_SIZE == 2
411    
412  #define PUT(a,n,d)   \  #define PUT(a,n,d)   \
# Line 329  is automated on Unix systems via the "co Line 443  is automated on Unix systems via the "co
443  #define GET(a,n) \  #define GET(a,n) \
444    (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])    (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
445    
446  #define MAX_PATTERN_SIZE (1 << 30)   /* Keep it positive */  /* Keep it positive */
447    #define MAX_PATTERN_SIZE (1 << 30)
448    
449    #else
450    #error LINK_SIZE must be either 2, 3, or 4
451    #endif
452    
453    #elif defined COMPILE_PCRE16
454    
455    #if LINK_SIZE == 2
456    
457    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
458    #undef LINK_SIZE
459    #define LINK_SIZE 1
460    
461    #define PUT(a,n,d)   \
462      (a[n] = (d))
463    
464    #define GET(a,n) \
465      (a[n])
466    
467    #define MAX_PATTERN_SIZE (1 << 16)
468    
469    #elif LINK_SIZE == 3 || LINK_SIZE == 4
470    
471    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
472    #undef LINK_SIZE
473    #define LINK_SIZE 2
474    
475    #define PUT(a,n,d)   \
476      (a[n] = (d) >> 16), \
477      (a[(n)+1] = (d) & 65535)
478    
479    #define GET(a,n) \
480      (((a)[n] << 16) | (a)[(n)+1])
481    
482    /* Keep it positive */
483    #define MAX_PATTERN_SIZE (1 << 30)
484    
485  #else  #else
486  #error LINK_SIZE must be either 2, 3, or 4  #error LINK_SIZE must be either 2, 3, or 4
487  #endif  #endif
488    
489    #elif defined COMPILE_PCRE32
490    
491    /* Only supported LINK_SIZE is 4 */
492    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
493    #undef LINK_SIZE
494    #define LINK_SIZE 1
495    
496    #define PUT(a,n,d)   \
497      (a[n] = (d))
498    
499    #define GET(a,n) \
500      (a[n])
501    
502    /* Keep it positive */
503    #define MAX_PATTERN_SIZE (1 << 30)
504    
505    #else
506    #error Unsupported compiling mode
507    #endif /* COMPILE_PCRE[8|16|32] */
508    
509  /* Convenience macro defined in terms of the others */  /* Convenience macro defined in terms of the others */
510    
# Line 346  is automated on Unix systems via the "co Line 515  is automated on Unix systems via the "co
515  offsets changes. There are used for repeat counts and for other things such as  offsets changes. There are used for repeat counts and for other things such as
516  capturing parenthesis numbers in back references. */  capturing parenthesis numbers in back references. */
517    
518    #if defined COMPILE_PCRE8
519    
520    #define IMM2_SIZE 2
521    
522  #define PUT2(a,n,d)   \  #define PUT2(a,n,d)   \
523    a[n] = (d) >> 8; \    a[n] = (d) >> 8; \
524    a[(n)+1] = (d) & 255    a[(n)+1] = (d) & 255
# Line 353  capturing parenthesis numbers in back re Line 526  capturing parenthesis numbers in back re
526  #define GET2(a,n) \  #define GET2(a,n) \
527    (((a)[n] << 8) | (a)[(n)+1])    (((a)[n] << 8) | (a)[(n)+1])
528    
529  #define PUT2INC(a,n,d)  PUT2(a,n,d), a += 2  #elif defined COMPILE_PCRE16
530    
531    #define IMM2_SIZE 1
532    
533  /* When UTF-8 encoding is being used, a character is no longer just a single  #define PUT2(a,n,d)   \
534  byte. The macros for character handling generate simple sequences when used in     a[n] = d
535  byte-mode, and more complicated ones for UTF-8 characters. BACKCHAR should  
536  never be called in byte mode. To make sure it can never even appear when UTF-8  #define GET2(a,n) \
537  support is omitted, we don't even define it. */     a[n]
538    
539    #elif defined COMPILE_PCRE32
540    
541    #define IMM2_SIZE 1
542    
543    #define PUT2(a,n,d)   \
544       a[n] = d
545    
546  #ifndef SUPPORT_UTF8  #define GET2(a,n) \
547       a[n]
548    
549    #else
550    #error Unsupported compiling mode
551    #endif /* COMPILE_PCRE[8|16|32] */
552    
553    #define PUT2INC(a,n,d)  PUT2(a,n,d), a += IMM2_SIZE
554    
555    /* The maximum length of a MARK name is currently one data unit; it may be
556    changed in future to be a fixed number of bytes or to depend on LINK_SIZE. */
557    
558    #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
559    #define MAX_MARK ((1u << 16) - 1)
560    #else
561    #define MAX_MARK ((1u << 8) - 1)
562    #endif
563    
564    /* When UTF encoding is being used, a character is no longer just a single
565    byte. The macros for character handling generate simple sequences when used in
566    character-mode, and more complicated ones for UTF characters. GETCHARLENTEST
567    and other macros are not used when UTF is not supported, so they are not
568    defined. To make sure they can never even appear when UTF support is omitted,
569    we don't even define them. */
570    
571    #ifndef SUPPORT_UTF
572    
573    /* #define MAX_VALUE_FOR_SINGLE_CHAR */
574    /* #define HAS_EXTRALEN(c) */
575    /* #define GET_EXTRALEN(c) */
576    /* #define NOT_FIRSTCHAR(c) */
577  #define GETCHAR(c, eptr) c = *eptr;  #define GETCHAR(c, eptr) c = *eptr;
578  #define GETCHARTEST(c, eptr) c = *eptr;  #define GETCHARTEST(c, eptr) c = *eptr;
579  #define GETCHARINC(c, eptr) c = *eptr++;  #define GETCHARINC(c, eptr) c = *eptr++;
580  #define GETCHARINCTEST(c, eptr) c = *eptr++;  #define GETCHARINCTEST(c, eptr) c = *eptr++;
581  #define GETCHARLEN(c, eptr, len) c = *eptr;  #define GETCHARLEN(c, eptr, len) c = *eptr;
582    #define RAWUCHAR(eptr) (*(eptr))
583    #define RAWUCHARINC(eptr) (*(eptr)++)
584    #define RAWUCHARTEST(eptr) (*(eptr))
585    #define RAWUCHARINCTEST(eptr) (*(eptr)++)
586    /* #define GETCHARLENTEST(c, eptr, len) */
587  /* #define BACKCHAR(eptr) */  /* #define BACKCHAR(eptr) */
588    /* #define FORWARDCHAR(eptr) */
589    /* #define ACROSSCHAR(condition, eptr, action) */
590    
591  #else   /* SUPPORT_UTF8 */  #else   /* SUPPORT_UTF */
592    
593  /* Get the next UTF-8 character, not advancing the pointer. This is called when  /* Tests whether the code point needs extra characters to decode. */
 we know we are in UTF-8 mode. */  
594    
595  #define GETCHAR(c, eptr) \  #define HASUTF8EXTRALEN(c) ((c) >= 0xc0)
596    c = *eptr; \  
597    if (c >= 0xc0) \  /* Base macro to pick up the remaining bytes of a UTF-8 character, not
598    advancing the pointer. */
599    
600    #define GETUTF8(c, eptr) \
601      { \      { \
602      int gcii; \      if ((c & 0x20) == 0) \
603      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \        c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
604      int gcss = 6*gcaa; \      else if ((c & 0x10) == 0) \
605      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
606      for (gcii = 1; gcii <= gcaa; gcii++) \      else if ((c & 0x08) == 0) \
607          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
608          ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
609        else if ((c & 0x04) == 0) \
610          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
611              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
612              (eptr[4] & 0x3f); \
613        else \
614          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
615              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
616              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
617        }
618    
619    /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
620    the pointer. */
621    
622    #define GETUTF8INC(c, eptr) \
623        { \
624        if ((c & 0x20) == 0) \
625          c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
626        else if ((c & 0x10) == 0) \
627        { \        { \
628        gcss -= 6; \        c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
629        c |= (eptr[gcii] & 0x3f) << gcss; \        eptr += 2; \
630          } \
631        else if ((c & 0x08) == 0) \
632          { \
633          c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
634              ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
635          eptr += 3; \
636          } \
637        else if ((c & 0x04) == 0) \
638          { \
639          c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
640              ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
641              (eptr[3] & 0x3f); \
642          eptr += 4; \
643          } \
644        else \
645          { \
646          c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
647              ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
648              ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
649          eptr += 5; \
650        } \        } \
651      }      }
652    
653    #if defined COMPILE_PCRE8
654    
655    /* These macros were originally written in the form of loops that used data
656    from the tables whose names start with PRIV(utf8_table). They were rewritten by
657    a user so as not to use loops, because in some environments this gives a
658    significant performance advantage, and it seems never to do any harm. */
659    
660    /* Tells the biggest code point which can be encoded as a single character. */
661    
662    #define MAX_VALUE_FOR_SINGLE_CHAR 127
663    
664    /* Tests whether the code point needs extra characters to decode. */
665    
666    #define HAS_EXTRALEN(c) ((c) >= 0xc0)
667    
668    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
669    Otherwise it has an undefined behaviour. */
670    
671    #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
672    
673    /* Returns TRUE, if the given character is not the first character
674    of a UTF sequence. */
675    
676    #define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80)
677    
678    /* Get the next UTF-8 character, not advancing the pointer. This is called when
679    we know we are in UTF-8 mode. */
680    
681    #define GETCHAR(c, eptr) \
682      c = *eptr; \
683      if (c >= 0xc0) GETUTF8(c, eptr);
684    
685  /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the  /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
686  pointer. */  pointer. */
687    
688  #define GETCHARTEST(c, eptr) \  #define GETCHARTEST(c, eptr) \
689    c = *eptr; \    c = *eptr; \
690    if (utf8 && c >= 0xc0) \    if (utf && c >= 0xc0) GETUTF8(c, eptr);
     { \  
     int gcii; \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     for (gcii = 1; gcii <= gcaa; gcii++) \  
       { \  
       gcss -= 6; \  
       c |= (eptr[gcii] & 0x3f) << gcss; \  
       } \  
     }  
691    
692  /* Get the next UTF-8 character, advancing the pointer. This is called when we  /* Get the next UTF-8 character, advancing the pointer. This is called when we
693  know we are in UTF-8 mode. */  know we are in UTF-8 mode. */
694    
695  #define GETCHARINC(c, eptr) \  #define GETCHARINC(c, eptr) \
696    c = *eptr++; \    c = *eptr++; \
697    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8INC(c, eptr);
     { \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     while (gcaa-- > 0) \  
       { \  
       gcss -= 6; \  
       c |= (*eptr++ & 0x3f) << gcss; \  
       } \  
     }  
698    
699  /* Get the next character, testing for UTF-8 mode, and advancing the pointer */  /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
700    This is called when we don't know if we are in UTF-8 mode. */
701    
702  #define GETCHARINCTEST(c, eptr) \  #define GETCHARINCTEST(c, eptr) \
703    c = *eptr++; \    c = *eptr++; \
704    if (utf8 && c >= 0xc0) \    if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
705    
706    /* Base macro to pick up the remaining bytes of a UTF-8 character, not
707    advancing the pointer, incrementing the length. */
708    
709    #define GETUTF8LEN(c, eptr, len) \
710      { \      { \
711      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \      if ((c & 0x20) == 0) \
712      int gcss = 6*gcaa; \        { \
713      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
714      while (gcaa-- > 0) \        len++; \
715          } \
716        else if ((c & 0x10)  == 0) \
717          { \
718          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
719          len += 2; \
720          } \
721        else if ((c & 0x08)  == 0) \
722          {\
723          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
724              ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
725          len += 3; \
726          } \
727        else if ((c & 0x04)  == 0) \
728        { \        { \
729        gcss -= 6; \        c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
730        c |= (*eptr++ & 0x3f) << gcss; \            ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
731              (eptr[4] & 0x3f); \
732          len += 4; \
733          } \
734        else \
735          {\
736          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
737              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
738              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
739          len += 5; \
740        } \        } \
741      }      }
742    
# Line 446  if there are extra bytes. This is called Line 745  if there are extra bytes. This is called
745    
746  #define GETCHARLEN(c, eptr, len) \  #define GETCHARLEN(c, eptr, len) \
747    c = *eptr; \    c = *eptr; \
748    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
749      { \  
750      int gcii; \  /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
751      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  pointer, incrementing length if there are extra bytes. This is called when we
752      int gcss = 6*gcaa; \  do not know if we are in UTF-8 mode. */
753      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
754      for (gcii = 1; gcii <= gcaa; gcii++) \  #define GETCHARLENTEST(c, eptr, len) \
755        { \    c = *eptr; \
756        gcss -= 6; \    if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
757        c |= (eptr[gcii] & 0x3f) << gcss; \  
758        } \  /* Returns the next uchar, not advancing the pointer. This is called when
759      len += gcaa; \  we know we are in UTF mode. */
760      }  
761    #define RAWUCHAR(eptr) \
762      (*(eptr))
763    
764    /* Returns the next uchar, advancing the pointer. This is called when
765    we know we are in UTF mode. */
766    
767    #define RAWUCHARINC(eptr) \
768      (*(eptr)++)
769    
770    /* Returns the next uchar, testing for UTF mode, and not advancing the
771    pointer. */
772    
773    #define RAWUCHARTEST(eptr) \
774      (*(eptr))
775    
776    /* Returns the next uchar, testing for UTF mode, advancing the
777    pointer. */
778    
779    #define RAWUCHARINCTEST(eptr) \
780      (*(eptr)++)
781    
782  /* If the pointer is not at the start of a character, move it back until  /* If the pointer is not at the start of a character, move it back until
783  it is. This is called only in UTF-8 mode - we don't put a test within the macro  it is. This is called only in UTF-8 mode - we don't put a test within the macro
# Line 466  because almost all calls are already wit Line 785  because almost all calls are already wit
785    
786  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
787    
788  #endif  /* Same as above, just in the other direction. */
789    #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
790    
791    /* Same as above, but it allows a fully customizable form. */
792    #define ACROSSCHAR(condition, eptr, action) \
793      while((condition) && ((eptr) & 0xc0) == 0x80) action
794    
795  /* In case there is no definition of offsetof() provided - though any proper  #elif defined COMPILE_PCRE16
 Standard C system should have one. */  
796    
797  #ifndef offsetof  /* Tells the biggest code point which can be encoded as a single character. */
798  #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))  
799  #endif  #define MAX_VALUE_FOR_SINGLE_CHAR 65535
800    
801    /* Tests whether the code point needs extra characters to decode. */
802    
803    #define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
804    
805    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
806    Otherwise it has an undefined behaviour. */
807    
808    #define GET_EXTRALEN(c) 1
809    
810    /* Returns TRUE, if the given character is not the first character
811    of a UTF sequence. */
812    
813    #define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00)
814    
815    /* Base macro to pick up the low surrogate of a UTF-16 character, not
816    advancing the pointer. */
817    
818    #define GETUTF16(c, eptr) \
819       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
820    
821    /* Get the next UTF-16 character, not advancing the pointer. This is called when
822    we know we are in UTF-16 mode. */
823    
824    #define GETCHAR(c, eptr) \
825      c = *eptr; \
826      if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
827    
828    /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
829    pointer. */
830    
831    #define GETCHARTEST(c, eptr) \
832      c = *eptr; \
833      if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
834    
835    /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
836    the pointer. */
837    
838    #define GETUTF16INC(c, eptr) \
839       { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
840    
841    /* Get the next UTF-16 character, advancing the pointer. This is called when we
842    know we are in UTF-16 mode. */
843    
844    #define GETCHARINC(c, eptr) \
845      c = *eptr++; \
846      if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
847    
848    /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
849    This is called when we don't know if we are in UTF-16 mode. */
850    
851    #define GETCHARINCTEST(c, eptr) \
852      c = *eptr++; \
853      if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
854    
855    /* Base macro to pick up the low surrogate of a UTF-16 character, not
856    advancing the pointer, incrementing the length. */
857    
858    #define GETUTF16LEN(c, eptr, len) \
859       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
860    
861    /* Get the next UTF-16 character, not advancing the pointer, incrementing
862    length if there is a low surrogate. This is called when we know we are in
863    UTF-16 mode. */
864    
865    #define GETCHARLEN(c, eptr, len) \
866      c = *eptr; \
867      if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
868    
869    /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
870    pointer, incrementing length if there is a low surrogate. This is called when
871    we do not know if we are in UTF-16 mode. */
872    
873    #define GETCHARLENTEST(c, eptr, len) \
874      c = *eptr; \
875      if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
876    
877    /* Returns the next uchar, not advancing the pointer. This is called when
878    we know we are in UTF mode. */
879    
880    #define RAWUCHAR(eptr) \
881      (*(eptr))
882    
883    /* Returns the next uchar, advancing the pointer. This is called when
884    we know we are in UTF mode. */
885    
886    #define RAWUCHARINC(eptr) \
887      (*(eptr)++)
888    
889    /* Returns the next uchar, testing for UTF mode, and not advancing the
890    pointer. */
891    
892    #define RAWUCHARTEST(eptr) \
893      (*(eptr))
894    
895    /* Returns the next uchar, testing for UTF mode, advancing the
896    pointer. */
897    
898    #define RAWUCHARINCTEST(eptr) \
899      (*(eptr)++)
900    
901    /* If the pointer is not at the start of a character, move it back until
902    it is. This is called only in UTF-16 mode - we don't put a test within the
903    macro because almost all calls are already within a block of UTF-16 only
904    code. */
905    
906    #define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
907    
908    /* Same as above, just in the other direction. */
909    #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
910    
911    /* Same as above, but it allows a fully customizable form. */
912    #define ACROSSCHAR(condition, eptr, action) \
913      if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
914    
915    #elif defined COMPILE_PCRE32
916    
917    /* These are trivial for the 32-bit library, since all UTF-32 characters fit
918    into one pcre_uchar unit. */
919    #define MAX_VALUE_FOR_SINGLE_CHAR (0x10ffffu)
920    #define HAS_EXTRALEN(c) (0)
921    #define GET_EXTRALEN(c) (0)
922    #define NOT_FIRSTCHAR(c) (0)
923    
924    #define UTF32_MASK (0x1fffffu)
925    
926    /* Get the next UTF-32 character, not advancing the pointer. This is called when
927    we know we are in UTF-32 mode. */
928    
929    #define GETCHAR(c, eptr) \
930      c = (*eptr) & UTF32_MASK;
931    
932    /* Get the next UTF-32 character, testing for UTF-32 mode, and not advancing the
933    pointer. */
934    
935    #define GETCHARTEST(c, eptr) \
936      c = *eptr; \
937      if (utf) c &= UTF32_MASK;
938    
939    /* Get the next UTF-32 character, advancing the pointer. This is called when we
940    know we are in UTF-32 mode. */
941    
942    #define GETCHARINC(c, eptr) \
943      c = (*eptr++) & UTF32_MASK;
944    
945    /* Get the next character, testing for UTF-32 mode, and advancing the pointer.
946    This is called when we don't know if we are in UTF-32 mode. */
947    
948    #define GETCHARINCTEST(c, eptr) \
949      c = *eptr++; \
950      if (utf) c &= UTF32_MASK;
951    
952    /* Get the next UTF-32 character, not advancing the pointer, not incrementing
953    length (since all UTF-32 is of length 1). This is called when we know we are in
954    UTF-32 mode. */
955    
956    #define GETCHARLEN(c, eptr, len) \
957      GETCHAR(c, eptr)
958    
959    /* Get the next UTF-32character, testing for UTF-32 mode, not advancing the
960    pointer, not incrementing the length (since all UTF-32 is of length 1).
961    This is called when we do not know if we are in UTF-32 mode. */
962    
963    #define GETCHARLENTEST(c, eptr, len) \
964      GETCHARTEST(c, eptr)
965    
966    /* Returns the next uchar, not advancing the pointer. This is called when
967    we know we are in UTF mode. */
968    
969    #define RAWUCHAR(eptr) \
970      (*(eptr) & UTF32_MASK)
971    
972    /* Returns the next uchar, advancing the pointer. This is called when
973    we know we are in UTF mode. */
974    
975    #define RAWUCHARINC(eptr) \
976      (*(eptr)++ & UTF32_MASK)
977    
978    /* Returns the next uchar, testing for UTF mode, and not advancing the
979    pointer. */
980    
981    #define RAWUCHARTEST(eptr) \
982      (utf ? (*(eptr) & UTF32_MASK) : *(eptr))
983    
984    /* Returns the next uchar, testing for UTF mode, advancing the
985    pointer. */
986    
987    #define RAWUCHARINCTEST(eptr) \
988      (utf ? (*(eptr)++ & UTF32_MASK) : *(eptr)++)
989    
990    /* If the pointer is not at the start of a character, move it back until
991    it is. This is called only in UTF-32 mode - we don't put a test within the
992    macro because almost all calls are already within a block of UTF-32 only
993    code.
994    These are all no-ops since all UTF-32 characters fit into one pcre_uchar. */
995    
996    #define BACKCHAR(eptr) do { } while (0)
997    
998    /* Same as above, just in the other direction. */
999    #define FORWARDCHAR(eptr) do { } while (0)
1000    
1001    /* Same as above, but it allows a fully customizable form. */
1002    #define ACROSSCHAR(condition, eptr, action) do { } while (0)
1003    
1004    #else
1005    #error Unsupported compiling mode
1006    #endif /* COMPILE_PCRE[8|16|32] */
1007    
1008    #endif  /* SUPPORT_UTF */
1009    
1010    /* Tests for Unicode horizontal and vertical whitespace characters must check a
1011    number of different values. Using a switch statement for this generates the
1012    fastest code (no loop, no memory access), and there are several places in the
1013    interpreter code where this happens. In order to ensure that all the case lists
1014    remain in step, we use macros so that there is only one place where the lists
1015    are defined.
1016    
1017    These values are also required as lists in pcre_compile.c when processing \h,
1018    \H, \v and \V in a character class. The lists are defined in pcre_tables.c, but
1019    macros that define the values are here so that all the definitions are
1020    together. The lists must be in ascending character order, terminated by
1021    NOTACHAR (which is 0xffffffff).
1022    
1023    Any changes should ensure that the various macros are kept in step with each
1024    other. NOTE: The values also appear in pcre_jit_compile.c. */
1025    
1026    /* ------ ASCII/Unicode environments ------ */
1027    
1028    #ifndef EBCDIC
1029    
1030    #define HSPACE_LIST \
1031      CHAR_HT, CHAR_SPACE, 0xa0, \
1032      0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \
1033      0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \
1034      NOTACHAR
1035    
1036    #define HSPACE_MULTIBYTE_CASES \
1037      case 0x1680:  /* OGHAM SPACE MARK */ \
1038      case 0x180e:  /* MONGOLIAN VOWEL SEPARATOR */ \
1039      case 0x2000:  /* EN QUAD */ \
1040      case 0x2001:  /* EM QUAD */ \
1041      case 0x2002:  /* EN SPACE */ \
1042      case 0x2003:  /* EM SPACE */ \
1043      case 0x2004:  /* THREE-PER-EM SPACE */ \
1044      case 0x2005:  /* FOUR-PER-EM SPACE */ \
1045      case 0x2006:  /* SIX-PER-EM SPACE */ \
1046      case 0x2007:  /* FIGURE SPACE */ \
1047      case 0x2008:  /* PUNCTUATION SPACE */ \
1048      case 0x2009:  /* THIN SPACE */ \
1049      case 0x200A:  /* HAIR SPACE */ \
1050      case 0x202f:  /* NARROW NO-BREAK SPACE */ \
1051      case 0x205f:  /* MEDIUM MATHEMATICAL SPACE */ \
1052      case 0x3000   /* IDEOGRAPHIC SPACE */
1053    
1054    #define HSPACE_BYTE_CASES \
1055      case CHAR_HT: \
1056      case CHAR_SPACE: \
1057      case 0xa0     /* NBSP */
1058    
1059    #define HSPACE_CASES \
1060      HSPACE_BYTE_CASES: \
1061      HSPACE_MULTIBYTE_CASES
1062    
1063    #define VSPACE_LIST \
1064      CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR
1065    
1066    #define VSPACE_MULTIBYTE_CASES \
1067      case 0x2028:    /* LINE SEPARATOR */ \
1068      case 0x2029     /* PARAGRAPH SEPARATOR */
1069    
1070    #define VSPACE_BYTE_CASES \
1071      case CHAR_LF: \
1072      case CHAR_VT: \
1073      case CHAR_FF: \
1074      case CHAR_CR: \
1075      case CHAR_NEL
1076    
1077    #define VSPACE_CASES \
1078      VSPACE_BYTE_CASES: \
1079      VSPACE_MULTIBYTE_CASES
1080    
1081    /* ------ EBCDIC environments ------ */
1082    
1083    #else
1084    #define HSPACE_LIST CHAR_HT, CHAR_SPACE
1085    
1086    #define HSPACE_BYTE_CASES \
1087      case CHAR_HT: \
1088      case CHAR_SPACE
1089    
1090    #define HSPACE_CASES HSPACE_BYTE_CASES
1091    
1092    #ifdef EBCDIC_NL25
1093    #define VSPACE_LIST \
1094      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR
1095    #else
1096    #define VSPACE_LIST \
1097      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR
1098    #endif
1099    
1100    #define VSPACE_BYTE_CASES \
1101      case CHAR_LF: \
1102      case CHAR_VT: \
1103      case CHAR_FF: \
1104      case CHAR_CR: \
1105      case CHAR_NEL
1106    
1107    #define VSPACE_CASES VSPACE_BYTE_CASES
1108    #endif  /* EBCDIC */
1109    
1110    /* ------ End of whitespace macros ------ */
1111    
 /* These are the public options that can change during matching. */  
1112    
 #define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)  
1113    
1114  /* Private flags containing information about the compiled regex. They used to  /* Private flags containing information about the compiled regex. They used to
1115  live at the top end of the options word, but that got almost full, so now they  live at the top end of the options word, but that got almost full, so now they
1116  are in a 16-bit flags word. */  are in a 16-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as
1117    the restrictions on partial matching have been lifted. It remains for backwards
1118  #define PCRE_NOPARTIAL     0x0001  /* can't use partial with this regex */  compatibility. */
1119  #define PCRE_FIRSTSET      0x0002  /* first_byte is set */  
1120  #define PCRE_REQCHSET      0x0004  /* req_byte is set */  #define PCRE_MODE8         0x0001  /* compiled in 8 bit mode */
1121  #define PCRE_STARTLINE     0x0008  /* start after \n for multiline */  #define PCRE_MODE16        0x0002  /* compiled in 16 bit mode */
1122  #define PCRE_JCHANGED      0x0010  /* j option used in regex */  #define PCRE_MODE32        0x0004  /* compiled in 32 bit mode */
1123  #define PCRE_HASCRORLF     0x0020  /* explicit \r or \n in pattern */  #define PCRE_FIRSTSET      0x0010  /* first_char is set */
1124    #define PCRE_FCH_CASELESS  0x0020  /* caseless first char */
1125    #define PCRE_REQCHSET      0x0040  /* req_byte is set */
1126    #define PCRE_RCH_CASELESS  0x0080  /* caseless requested char */
1127    #define PCRE_STARTLINE     0x0100  /* start after \n for multiline */
1128    #define PCRE_NOPARTIAL     0x0200  /* can't use partial with this regex */
1129    #define PCRE_JCHANGED      0x0400  /* j option used in regex */
1130    #define PCRE_HASCRORLF     0x0800  /* explicit \r or \n in pattern */
1131    #define PCRE_HASTHEN       0x1000  /* pattern contains (*THEN) */
1132    
1133    #if defined COMPILE_PCRE8
1134    #define PCRE_MODE          PCRE_MODE8
1135    #elif defined COMPILE_PCRE16
1136    #define PCRE_MODE          PCRE_MODE16
1137    #elif defined COMPILE_PCRE32
1138    #define PCRE_MODE          PCRE_MODE32
1139    #endif
1140    #define PCRE_MODE_MASK     (PCRE_MODE8 | PCRE_MODE16 | PCRE_MODE32)
1141    
1142  /* Options for the "extra" block produced by pcre_study(). */  /* Flags for the "extra" block produced by pcre_study(). */
1143    
1144  #define PCRE_STUDY_MAPPED   0x01     /* a map of starting chars exists */  #define PCRE_STUDY_MAPPED  0x0001  /* a map of starting chars exists */
1145    #define PCRE_STUDY_MINLEN  0x0002  /* a minimum length field exists */
1146    
1147  /* Masks for identifying the public options that are permitted at compile  /* Masks for identifying the public options that are permitted at compile
1148  time, run time, or study time, respectively. */  time, run time, or study time, respectively. */
# Line 502  time, run time, or study time, respectiv Line 1150  time, run time, or study time, respectiv
1150  #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \  #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
1151                             PCRE_NEWLINE_ANYCRLF)                             PCRE_NEWLINE_ANYCRLF)
1152    
1153  #define PUBLIC_OPTIONS \  #define PUBLIC_COMPILE_OPTIONS \
1154    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
1155     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
1156     PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \     PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
1157     PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)     PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1158       PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE)
1159    
1160  #define PUBLIC_EXEC_OPTIONS \  #define PUBLIC_EXEC_OPTIONS \
1161    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1162     PCRE_PARTIAL|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
1163       PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
1164    
1165  #define PUBLIC_DFA_EXEC_OPTIONS \  #define PUBLIC_DFA_EXEC_OPTIONS \
1166    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1167     PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_BITS| \     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
1168     PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)     PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1169       PCRE_NO_START_OPTIMIZE)
1170    
1171    #define PUBLIC_STUDY_OPTIONS \
1172       (PCRE_STUDY_JIT_COMPILE|PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE| \
1173        PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE|PCRE_STUDY_EXTRA_NEEDED)
1174    
1175  #define PUBLIC_STUDY_OPTIONS 0   /* None defined */  /* Magic number to provide a small check against being handed junk. */
   
 /* Magic number to provide a small check against being handed junk. Also used  
 to detect whether a pattern was compiled on a host of different endianness. */  
1176    
1177  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */
1178    
1179  /* Negative values for the firstchar and reqchar variables */  /* This variable is used to detect a loaded regular expression
1180    in different endianness. */
1181    
1182  #define REQ_UNSET (-2)  #define REVERSED_MAGIC_NUMBER  0x45524350UL   /* 'ERCP' */
 #define REQ_NONE  (-1)  
1183    
1184  /* The maximum remaining length of subject we are prepared to search for a  /* The maximum remaining length of subject we are prepared to search for a
1185  req_byte match. */  req_byte match. */
1186    
1187  #define REQ_BYTE_MAX 1000  #define REQ_BYTE_MAX 1000
1188    
1189  /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a  /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
1190  variable-length repeat, or a anything other than literal characters. */  environments where these macros are defined elsewhere. Unfortunately, there
1191    is no way to do the same for the typedef. */
 #define REQ_CASELESS 0x0100    /* indicates caselessness */  
 #define REQ_VARY     0x0200    /* reqbyte followed non-literal item */  
   
 /* Miscellaneous definitions */  
1192    
1193  typedef int BOOL;  typedef int BOOL;
1194    
1195    #ifndef FALSE
1196  #define FALSE   0  #define FALSE   0
1197  #define TRUE    1  #define TRUE    1
1198    #endif
1199    
1200    /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
1201    character constants like '*' because the compiler would emit their EBCDIC code,
1202    which is different from their ASCII/UTF-8 code. Instead we define macros for
1203    the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
1204    is enabled. When UTF-8 support is not enabled, the definitions use character
1205    literals. Both character and string versions of each character are needed, and
1206    there are some longer strings as well.
1207    
1208    This means that, on EBCDIC platforms, the PCRE library can handle either
1209    EBCDIC, or UTF-8, but not both. To support both in the same compiled library
1210    would need different lookups depending on whether PCRE_UTF8 was set or not.
1211    This would make it impossible to use characters in switch/case statements,
1212    which would reduce performance. For a theoretical use (which nobody has asked
1213    for) in a minority area (EBCDIC platforms), this is not sensible. Any
1214    application that did need both could compile two versions of the library, using
1215    macros to give the functions distinct names. */
1216    
1217    #ifndef SUPPORT_UTF
1218    
1219    /* UTF-8 support is not enabled; use the platform-dependent character literals
1220    so that PCRE works in both ASCII and EBCDIC environments, but only in non-UTF
1221    mode. Newline characters are problematic in EBCDIC. Though it has CR and LF
1222    characters, a common practice has been to use its NL (0x15) character as the
1223    line terminator in C-like processing environments. However, sometimes the LF
1224    (0x25) character is used instead, according to this Unicode document:
1225    
1226    http://unicode.org/standard/reports/tr13/tr13-5.html
1227    
1228    PCRE defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25
1229    instead. Whichever is *not* chosen is defined as NEL.
1230    
1231    In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the
1232    same code point. */
1233    
1234    #ifdef EBCDIC
1235    
1236    #ifndef EBCDIC_NL25
1237    #define CHAR_NL                     '\x15'
1238    #define CHAR_NEL                    '\x25'
1239    #define STR_NL                      "\x15"
1240    #define STR_NEL                     "\x25"
1241    #else
1242    #define CHAR_NL                     '\x25'
1243    #define CHAR_NEL                    '\x15'
1244    #define STR_NL                      "\x25"
1245    #define STR_NEL                     "\x15"
1246    #endif
1247    
1248    #define CHAR_LF                     CHAR_NL
1249    #define STR_LF                      STR_NL
1250    
1251    #define CHAR_ESC                    '\047'
1252    #define CHAR_DEL                    '\007'
1253    #define STR_ESC                     "\047"
1254    #define STR_DEL                     "\007"
1255    
1256    #else  /* Not EBCDIC */
1257    
1258    /* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for
1259    compatibility. NEL is the Unicode newline character; make sure it is
1260    a positive value. */
1261    
1262    #define CHAR_LF                     '\n'
1263    #define CHAR_NL                     CHAR_LF
1264    #define CHAR_NEL                    ((unsigned char)'\x85')
1265    #define CHAR_ESC                    '\033'
1266    #define CHAR_DEL                    '\177'
1267    
1268    #define STR_LF                      "\n"
1269    #define STR_NL                      STR_LF
1270    #define STR_NEL                     "\x85"
1271    #define STR_ESC                     "\033"
1272    #define STR_DEL                     "\177"
1273    
1274    #endif  /* EBCDIC */
1275    
1276    /* The remaining definitions work in both environments. */
1277    
1278    #define CHAR_HT                     '\t'
1279    #define CHAR_VT                     '\v'
1280    #define CHAR_FF                     '\f'
1281    #define CHAR_CR                     '\r'
1282    #define CHAR_BS                     '\b'
1283    #define CHAR_BEL                    '\a'
1284    
1285    #define CHAR_SPACE                  ' '
1286    #define CHAR_EXCLAMATION_MARK       '!'
1287    #define CHAR_QUOTATION_MARK         '"'
1288    #define CHAR_NUMBER_SIGN            '#'
1289    #define CHAR_DOLLAR_SIGN            '$'
1290    #define CHAR_PERCENT_SIGN           '%'
1291    #define CHAR_AMPERSAND              '&'
1292    #define CHAR_APOSTROPHE             '\''
1293    #define CHAR_LEFT_PARENTHESIS       '('
1294    #define CHAR_RIGHT_PARENTHESIS      ')'
1295    #define CHAR_ASTERISK               '*'
1296    #define CHAR_PLUS                   '+'
1297    #define CHAR_COMMA                  ','
1298    #define CHAR_MINUS                  '-'
1299    #define CHAR_DOT                    '.'
1300    #define CHAR_SLASH                  '/'
1301    #define CHAR_0                      '0'
1302    #define CHAR_1                      '1'
1303    #define CHAR_2                      '2'
1304    #define CHAR_3                      '3'
1305    #define CHAR_4                      '4'
1306    #define CHAR_5                      '5'
1307    #define CHAR_6                      '6'
1308    #define CHAR_7                      '7'
1309    #define CHAR_8                      '8'
1310    #define CHAR_9                      '9'
1311    #define CHAR_COLON                  ':'
1312    #define CHAR_SEMICOLON              ';'
1313    #define CHAR_LESS_THAN_SIGN         '<'
1314    #define CHAR_EQUALS_SIGN            '='
1315    #define CHAR_GREATER_THAN_SIGN      '>'
1316    #define CHAR_QUESTION_MARK          '?'
1317    #define CHAR_COMMERCIAL_AT          '@'
1318    #define CHAR_A                      'A'
1319    #define CHAR_B                      'B'
1320    #define CHAR_C                      'C'
1321    #define CHAR_D                      'D'
1322    #define CHAR_E                      'E'
1323    #define CHAR_F                      'F'
1324    #define CHAR_G                      'G'
1325    #define CHAR_H                      'H'
1326    #define CHAR_I                      'I'
1327    #define CHAR_J                      'J'
1328    #define CHAR_K                      'K'
1329    #define CHAR_L                      'L'
1330    #define CHAR_M                      'M'
1331    #define CHAR_N                      'N'
1332    #define CHAR_O                      'O'
1333    #define CHAR_P                      'P'
1334    #define CHAR_Q                      'Q'
1335    #define CHAR_R                      'R'
1336    #define CHAR_S                      'S'
1337    #define CHAR_T                      'T'
1338    #define CHAR_U                      'U'
1339    #define CHAR_V                      'V'
1340    #define CHAR_W                      'W'
1341    #define CHAR_X                      'X'
1342    #define CHAR_Y                      'Y'
1343    #define CHAR_Z                      'Z'
1344    #define CHAR_LEFT_SQUARE_BRACKET    '['
1345    #define CHAR_BACKSLASH              '\\'
1346    #define CHAR_RIGHT_SQUARE_BRACKET   ']'
1347    #define CHAR_CIRCUMFLEX_ACCENT      '^'
1348    #define CHAR_UNDERSCORE             '_'
1349    #define CHAR_GRAVE_ACCENT           '`'
1350    #define CHAR_a                      'a'
1351    #define CHAR_b                      'b'
1352    #define CHAR_c                      'c'
1353    #define CHAR_d                      'd'
1354    #define CHAR_e                      'e'
1355    #define CHAR_f                      'f'
1356    #define CHAR_g                      'g'
1357    #define CHAR_h                      'h'
1358    #define CHAR_i                      'i'
1359    #define CHAR_j                      'j'
1360    #define CHAR_k                      'k'
1361    #define CHAR_l                      'l'
1362    #define CHAR_m                      'm'
1363    #define CHAR_n                      'n'
1364    #define CHAR_o                      'o'
1365    #define CHAR_p                      'p'
1366    #define CHAR_q                      'q'
1367    #define CHAR_r                      'r'
1368    #define CHAR_s                      's'
1369    #define CHAR_t                      't'
1370    #define CHAR_u                      'u'
1371    #define CHAR_v                      'v'
1372    #define CHAR_w                      'w'
1373    #define CHAR_x                      'x'
1374    #define CHAR_y                      'y'
1375    #define CHAR_z                      'z'
1376    #define CHAR_LEFT_CURLY_BRACKET     '{'
1377    #define CHAR_VERTICAL_LINE          '|'
1378    #define CHAR_RIGHT_CURLY_BRACKET    '}'
1379    #define CHAR_TILDE                  '~'
1380    
1381    #define STR_HT                      "\t"
1382    #define STR_VT                      "\v"
1383    #define STR_FF                      "\f"
1384    #define STR_CR                      "\r"
1385    #define STR_BS                      "\b"
1386    #define STR_BEL                     "\a"
1387    
1388    #define STR_SPACE                   " "
1389    #define STR_EXCLAMATION_MARK        "!"
1390    #define STR_QUOTATION_MARK          "\""
1391    #define STR_NUMBER_SIGN             "#"
1392    #define STR_DOLLAR_SIGN             "$"
1393    #define STR_PERCENT_SIGN            "%"
1394    #define STR_AMPERSAND               "&"
1395    #define STR_APOSTROPHE              "'"
1396    #define STR_LEFT_PARENTHESIS        "("
1397    #define STR_RIGHT_PARENTHESIS       ")"
1398    #define STR_ASTERISK                "*"
1399    #define STR_PLUS                    "+"
1400    #define STR_COMMA                   ","
1401    #define STR_MINUS                   "-"
1402    #define STR_DOT                     "."
1403    #define STR_SLASH                   "/"
1404    #define STR_0                       "0"
1405    #define STR_1                       "1"
1406    #define STR_2                       "2"
1407    #define STR_3                       "3"
1408    #define STR_4                       "4"
1409    #define STR_5                       "5"
1410    #define STR_6                       "6"
1411    #define STR_7                       "7"
1412    #define STR_8                       "8"
1413    #define STR_9                       "9"
1414    #define STR_COLON                   ":"
1415    #define STR_SEMICOLON               ";"
1416    #define STR_LESS_THAN_SIGN          "<"
1417    #define STR_EQUALS_SIGN             "="
1418    #define STR_GREATER_THAN_SIGN       ">"
1419    #define STR_QUESTION_MARK           "?"
1420    #define STR_COMMERCIAL_AT           "@"
1421    #define STR_A                       "A"
1422    #define STR_B                       "B"
1423    #define STR_C                       "C"
1424    #define STR_D                       "D"
1425    #define STR_E                       "E"
1426    #define STR_F                       "F"
1427    #define STR_G                       "G"
1428    #define STR_H                       "H"
1429    #define STR_I                       "I"
1430    #define STR_J                       "J"
1431    #define STR_K                       "K"
1432    #define STR_L                       "L"
1433    #define STR_M                       "M"
1434    #define STR_N                       "N"
1435    #define STR_O                       "O"
1436    #define STR_P                       "P"
1437    #define STR_Q                       "Q"
1438    #define STR_R                       "R"
1439    #define STR_S                       "S"
1440    #define STR_T                       "T"
1441    #define STR_U                       "U"
1442    #define STR_V                       "V"
1443    #define STR_W                       "W"
1444    #define STR_X                       "X"
1445    #define STR_Y                       "Y"
1446    #define STR_Z                       "Z"
1447    #define STR_LEFT_SQUARE_BRACKET     "["
1448    #define STR_BACKSLASH               "\\"
1449    #define STR_RIGHT_SQUARE_BRACKET    "]"
1450    #define STR_CIRCUMFLEX_ACCENT       "^"
1451    #define STR_UNDERSCORE              "_"
1452    #define STR_GRAVE_ACCENT            "`"
1453    #define STR_a                       "a"
1454    #define STR_b                       "b"
1455    #define STR_c                       "c"
1456    #define STR_d                       "d"
1457    #define STR_e                       "e"
1458    #define STR_f                       "f"
1459    #define STR_g                       "g"
1460    #define STR_h                       "h"
1461    #define STR_i                       "i"
1462    #define STR_j                       "j"
1463    #define STR_k                       "k"
1464    #define STR_l                       "l"
1465    #define STR_m                       "m"
1466    #define STR_n                       "n"
1467    #define STR_o                       "o"
1468    #define STR_p                       "p"
1469    #define STR_q                       "q"
1470    #define STR_r                       "r"
1471    #define STR_s                       "s"
1472    #define STR_t                       "t"
1473    #define STR_u                       "u"
1474    #define STR_v                       "v"
1475    #define STR_w                       "w"
1476    #define STR_x                       "x"
1477    #define STR_y                       "y"
1478    #define STR_z                       "z"
1479    #define STR_LEFT_CURLY_BRACKET      "{"
1480    #define STR_VERTICAL_LINE           "|"
1481    #define STR_RIGHT_CURLY_BRACKET     "}"
1482    #define STR_TILDE                   "~"
1483    
1484    #define STRING_ACCEPT0              "ACCEPT\0"
1485    #define STRING_COMMIT0              "COMMIT\0"
1486    #define STRING_F0                   "F\0"
1487    #define STRING_FAIL0                "FAIL\0"
1488    #define STRING_MARK0                "MARK\0"
1489    #define STRING_PRUNE0               "PRUNE\0"
1490    #define STRING_SKIP0                "SKIP\0"
1491    #define STRING_THEN                 "THEN"
1492    
1493    #define STRING_alpha0               "alpha\0"
1494    #define STRING_lower0               "lower\0"
1495    #define STRING_upper0               "upper\0"
1496    #define STRING_alnum0               "alnum\0"
1497    #define STRING_ascii0               "ascii\0"
1498    #define STRING_blank0               "blank\0"
1499    #define STRING_cntrl0               "cntrl\0"
1500    #define STRING_digit0               "digit\0"
1501    #define STRING_graph0               "graph\0"
1502    #define STRING_print0               "print\0"
1503    #define STRING_punct0               "punct\0"
1504    #define STRING_space0               "space\0"
1505    #define STRING_word0                "word\0"
1506    #define STRING_xdigit               "xdigit"
1507    
1508    #define STRING_DEFINE               "DEFINE"
1509    
1510    #define STRING_CR_RIGHTPAR             "CR)"
1511    #define STRING_LF_RIGHTPAR             "LF)"
1512    #define STRING_CRLF_RIGHTPAR           "CRLF)"
1513    #define STRING_ANY_RIGHTPAR            "ANY)"
1514    #define STRING_ANYCRLF_RIGHTPAR        "ANYCRLF)"
1515    #define STRING_BSR_ANYCRLF_RIGHTPAR    "BSR_ANYCRLF)"
1516    #define STRING_BSR_UNICODE_RIGHTPAR    "BSR_UNICODE)"
1517    #ifdef COMPILE_PCRE8
1518    #define STRING_UTF_RIGHTPAR            "UTF8)"
1519    #endif
1520    #ifdef COMPILE_PCRE16
1521    #define STRING_UTF_RIGHTPAR            "UTF16)"
1522    #endif
1523    #ifdef COMPILE_PCRE32
1524    #define STRING_UTF_RIGHTPAR            "UTF32)"
1525    #endif
1526    #define STRING_UCP_RIGHTPAR            "UCP)"
1527    #define STRING_NO_START_OPT_RIGHTPAR   "NO_START_OPT)"
1528    
1529    #else  /* SUPPORT_UTF */
1530    
1531    /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1532    works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1533    only. */
1534    
1535    #define CHAR_HT                     '\011'
1536    #define CHAR_VT                     '\013'
1537    #define CHAR_FF                     '\014'
1538    #define CHAR_CR                     '\015'
1539    #define CHAR_LF                     '\012'
1540    #define CHAR_NL                     CHAR_LF
1541    #define CHAR_NEL                    ((unsigned char)'\x85')
1542    #define CHAR_BS                     '\010'
1543    #define CHAR_BEL                    '\007'
1544    #define CHAR_ESC                    '\033'
1545    #define CHAR_DEL                    '\177'
1546    
1547    #define CHAR_SPACE                  '\040'
1548    #define CHAR_EXCLAMATION_MARK       '\041'
1549    #define CHAR_QUOTATION_MARK         '\042'
1550    #define CHAR_NUMBER_SIGN            '\043'
1551    #define CHAR_DOLLAR_SIGN            '\044'
1552    #define CHAR_PERCENT_SIGN           '\045'
1553    #define CHAR_AMPERSAND              '\046'
1554    #define CHAR_APOSTROPHE             '\047'
1555    #define CHAR_LEFT_PARENTHESIS       '\050'
1556    #define CHAR_RIGHT_PARENTHESIS      '\051'
1557    #define CHAR_ASTERISK               '\052'
1558    #define CHAR_PLUS                   '\053'
1559    #define CHAR_COMMA                  '\054'
1560    #define CHAR_MINUS                  '\055'
1561    #define CHAR_DOT                    '\056'
1562    #define CHAR_SLASH                  '\057'
1563    #define CHAR_0                      '\060'
1564    #define CHAR_1                      '\061'
1565    #define CHAR_2                      '\062'
1566    #define CHAR_3                      '\063'
1567    #define CHAR_4                      '\064'
1568    #define CHAR_5                      '\065'
1569    #define CHAR_6                      '\066'
1570    #define CHAR_7                      '\067'
1571    #define CHAR_8                      '\070'
1572    #define CHAR_9                      '\071'
1573    #define CHAR_COLON                  '\072'
1574    #define CHAR_SEMICOLON              '\073'
1575    #define CHAR_LESS_THAN_SIGN         '\074'
1576    #define CHAR_EQUALS_SIGN            '\075'
1577    #define CHAR_GREATER_THAN_SIGN      '\076'
1578    #define CHAR_QUESTION_MARK          '\077'
1579    #define CHAR_COMMERCIAL_AT          '\100'
1580    #define CHAR_A                      '\101'
1581    #define CHAR_B                      '\102'
1582    #define CHAR_C                      '\103'
1583    #define CHAR_D                      '\104'
1584    #define CHAR_E                      '\105'
1585    #define CHAR_F                      '\106'
1586    #define CHAR_G                      '\107'
1587    #define CHAR_H                      '\110'
1588    #define CHAR_I                      '\111'
1589    #define CHAR_J                      '\112'
1590    #define CHAR_K                      '\113'
1591    #define CHAR_L                      '\114'
1592    #define CHAR_M                      '\115'
1593    #define CHAR_N                      '\116'
1594    #define CHAR_O                      '\117'
1595    #define CHAR_P                      '\120'
1596    #define CHAR_Q                      '\121'
1597    #define CHAR_R                      '\122'
1598    #define CHAR_S                      '\123'
1599    #define CHAR_T                      '\124'
1600    #define CHAR_U                      '\125'
1601    #define CHAR_V                      '\126'
1602    #define CHAR_W                      '\127'
1603    #define CHAR_X                      '\130'
1604    #define CHAR_Y                      '\131'
1605    #define CHAR_Z                      '\132'
1606    #define CHAR_LEFT_SQUARE_BRACKET    '\133'
1607    #define CHAR_BACKSLASH              '\134'
1608    #define CHAR_RIGHT_SQUARE_BRACKET   '\135'
1609    #define CHAR_CIRCUMFLEX_ACCENT      '\136'
1610    #define CHAR_UNDERSCORE             '\137'
1611    #define CHAR_GRAVE_ACCENT           '\140'
1612    #define CHAR_a                      '\141'
1613    #define CHAR_b                      '\142'
1614    #define CHAR_c                      '\143'
1615    #define CHAR_d                      '\144'
1616    #define CHAR_e                      '\145'
1617    #define CHAR_f                      '\146'
1618    #define CHAR_g                      '\147'
1619    #define CHAR_h                      '\150'
1620    #define CHAR_i                      '\151'
1621    #define CHAR_j                      '\152'
1622    #define CHAR_k                      '\153'
1623    #define CHAR_l                      '\154'
1624    #define CHAR_m                      '\155'
1625    #define CHAR_n                      '\156'
1626    #define CHAR_o                      '\157'
1627    #define CHAR_p                      '\160'
1628    #define CHAR_q                      '\161'
1629    #define CHAR_r                      '\162'
1630    #define CHAR_s                      '\163'
1631    #define CHAR_t                      '\164'
1632    #define CHAR_u                      '\165'
1633    #define CHAR_v                      '\166'
1634    #define CHAR_w                      '\167'
1635    #define CHAR_x                      '\170'
1636    #define CHAR_y                      '\171'
1637    #define CHAR_z                      '\172'
1638    #define CHAR_LEFT_CURLY_BRACKET     '\173'
1639    #define CHAR_VERTICAL_LINE          '\174'
1640    #define CHAR_RIGHT_CURLY_BRACKET    '\175'
1641    #define CHAR_TILDE                  '\176'
1642    
1643    #define STR_HT                      "\011"
1644    #define STR_VT                      "\013"
1645    #define STR_FF                      "\014"
1646    #define STR_CR                      "\015"
1647    #define STR_NL                      "\012"
1648    #define STR_BS                      "\010"
1649    #define STR_BEL                     "\007"
1650    #define STR_ESC                     "\033"
1651    #define STR_DEL                     "\177"
1652    
1653    #define STR_SPACE                   "\040"
1654    #define STR_EXCLAMATION_MARK        "\041"
1655    #define STR_QUOTATION_MARK          "\042"
1656    #define STR_NUMBER_SIGN             "\043"
1657    #define STR_DOLLAR_SIGN             "\044"
1658    #define STR_PERCENT_SIGN            "\045"
1659    #define STR_AMPERSAND               "\046"
1660    #define STR_APOSTROPHE              "\047"
1661    #define STR_LEFT_PARENTHESIS        "\050"
1662    #define STR_RIGHT_PARENTHESIS       "\051"
1663    #define STR_ASTERISK                "\052"
1664    #define STR_PLUS                    "\053"
1665    #define STR_COMMA                   "\054"
1666    #define STR_MINUS                   "\055"
1667    #define STR_DOT                     "\056"
1668    #define STR_SLASH                   "\057"
1669    #define STR_0                       "\060"
1670    #define STR_1                       "\061"
1671    #define STR_2                       "\062"
1672    #define STR_3                       "\063"
1673    #define STR_4                       "\064"
1674    #define STR_5                       "\065"
1675    #define STR_6                       "\066"
1676    #define STR_7                       "\067"
1677    #define STR_8                       "\070"
1678    #define STR_9                       "\071"
1679    #define STR_COLON                   "\072"
1680    #define STR_SEMICOLON               "\073"
1681    #define STR_LESS_THAN_SIGN          "\074"
1682    #define STR_EQUALS_SIGN             "\075"
1683    #define STR_GREATER_THAN_SIGN       "\076"
1684    #define STR_QUESTION_MARK           "\077"
1685    #define STR_COMMERCIAL_AT           "\100"
1686    #define STR_A                       "\101"
1687    #define STR_B                       "\102"
1688    #define STR_C                       "\103"
1689    #define STR_D                       "\104"
1690    #define STR_E                       "\105"
1691    #define STR_F                       "\106"
1692    #define STR_G                       "\107"
1693    #define STR_H                       "\110"
1694    #define STR_I                       "\111"
1695    #define STR_J                       "\112"
1696    #define STR_K                       "\113"
1697    #define STR_L                       "\114"
1698    #define STR_M                       "\115"
1699    #define STR_N                       "\116"
1700    #define STR_O                       "\117"
1701    #define STR_P                       "\120"
1702    #define STR_Q                       "\121"
1703    #define STR_R                       "\122"
1704    #define STR_S                       "\123"
1705    #define STR_T                       "\124"
1706    #define STR_U                       "\125"
1707    #define STR_V                       "\126"
1708    #define STR_W                       "\127"
1709    #define STR_X                       "\130"
1710    #define STR_Y                       "\131"
1711    #define STR_Z                       "\132"
1712    #define STR_LEFT_SQUARE_BRACKET     "\133"
1713    #define STR_BACKSLASH               "\134"
1714    #define STR_RIGHT_SQUARE_BRACKET    "\135"
1715    #define STR_CIRCUMFLEX_ACCENT       "\136"
1716    #define STR_UNDERSCORE              "\137"
1717    #define STR_GRAVE_ACCENT            "\140"
1718    #define STR_a                       "\141"
1719    #define STR_b                       "\142"
1720    #define STR_c                       "\143"
1721    #define STR_d                       "\144"
1722    #define STR_e                       "\145"
1723    #define STR_f                       "\146"
1724    #define STR_g                       "\147"
1725    #define STR_h                       "\150"
1726    #define STR_i                       "\151"
1727    #define STR_j                       "\152"
1728    #define STR_k                       "\153"
1729    #define STR_l                       "\154"
1730    #define STR_m                       "\155"
1731    #define STR_n                       "\156"
1732    #define STR_o                       "\157"
1733    #define STR_p                       "\160"
1734    #define STR_q                       "\161"
1735    #define STR_r                       "\162"
1736    #define STR_s                       "\163"
1737    #define STR_t                       "\164"
1738    #define STR_u                       "\165"
1739    #define STR_v                       "\166"
1740    #define STR_w                       "\167"
1741    #define STR_x                       "\170"
1742    #define STR_y                       "\171"
1743    #define STR_z                       "\172"
1744    #define STR_LEFT_CURLY_BRACKET      "\173"
1745    #define STR_VERTICAL_LINE           "\174"
1746    #define STR_RIGHT_CURLY_BRACKET     "\175"
1747    #define STR_TILDE                   "\176"
1748    
1749    #define STRING_ACCEPT0              STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1750    #define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1751    #define STRING_F0                   STR_F "\0"
1752    #define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"
1753    #define STRING_MARK0                STR_M STR_A STR_R STR_K "\0"
1754    #define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"
1755    #define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"
1756    #define STRING_THEN                 STR_T STR_H STR_E STR_N
1757    
1758    #define STRING_alpha0               STR_a STR_l STR_p STR_h STR_a "\0"
1759    #define STRING_lower0               STR_l STR_o STR_w STR_e STR_r "\0"
1760    #define STRING_upper0               STR_u STR_p STR_p STR_e STR_r "\0"
1761    #define STRING_alnum0               STR_a STR_l STR_n STR_u STR_m "\0"
1762    #define STRING_ascii0               STR_a STR_s STR_c STR_i STR_i "\0"
1763    #define STRING_blank0               STR_b STR_l STR_a STR_n STR_k "\0"
1764    #define STRING_cntrl0               STR_c STR_n STR_t STR_r STR_l "\0"
1765    #define STRING_digit0               STR_d STR_i STR_g STR_i STR_t "\0"
1766    #define STRING_graph0               STR_g STR_r STR_a STR_p STR_h "\0"
1767    #define STRING_print0               STR_p STR_r STR_i STR_n STR_t "\0"
1768    #define STRING_punct0               STR_p STR_u STR_n STR_c STR_t "\0"
1769    #define STRING_space0               STR_s STR_p STR_a STR_c STR_e "\0"
1770    #define STRING_word0                STR_w STR_o STR_r STR_d       "\0"
1771    #define STRING_xdigit               STR_x STR_d STR_i STR_g STR_i STR_t
1772    
1773    #define STRING_DEFINE               STR_D STR_E STR_F STR_I STR_N STR_E
1774    
1775    #define STRING_CR_RIGHTPAR             STR_C STR_R STR_RIGHT_PARENTHESIS
1776    #define STRING_LF_RIGHTPAR             STR_L STR_F STR_RIGHT_PARENTHESIS
1777    #define STRING_CRLF_RIGHTPAR           STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1778    #define STRING_ANY_RIGHTPAR            STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1779    #define STRING_ANYCRLF_RIGHTPAR        STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1780    #define STRING_BSR_ANYCRLF_RIGHTPAR    STR_B STR_S STR_R STR_UNDERSCORE STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1781    #define STRING_BSR_UNICODE_RIGHTPAR    STR_B STR_S STR_R STR_UNDERSCORE STR_U STR_N STR_I STR_C STR_O STR_D STR_E STR_RIGHT_PARENTHESIS
1782    #ifdef COMPILE_PCRE8
1783    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1784    #endif
1785    #ifdef COMPILE_PCRE16
1786    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
1787    #endif
1788    #ifdef COMPILE_PCRE32
1789    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS
1790    #endif
1791    #define STRING_UCP_RIGHTPAR            STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1792    #define STRING_NO_START_OPT_RIGHTPAR   STR_N STR_O STR_UNDERSCORE STR_S STR_T STR_A STR_R STR_T STR_UNDERSCORE STR_O STR_P STR_T STR_RIGHT_PARENTHESIS
1793    
1794    #endif  /* SUPPORT_UTF */
1795    
1796  /* Escape items that are just an encoding of a particular data value. */  /* Escape items that are just an encoding of a particular data value. */
1797    
1798  #ifndef ESC_e  #ifndef ESC_e
1799  #define ESC_e 27  #define ESC_e CHAR_ESC
1800  #endif  #endif
1801    
1802  #ifndef ESC_f  #ifndef ESC_f
1803  #define ESC_f '\f'  #define ESC_f CHAR_FF
1804  #endif  #endif
1805    
1806  #ifndef ESC_n  #ifndef ESC_n
1807  #define ESC_n '\n'  #define ESC_n CHAR_LF
1808  #endif  #endif
1809    
1810  #ifndef ESC_r  #ifndef ESC_r
1811  #define ESC_r '\r'  #define ESC_r CHAR_CR
1812  #endif  #endif
1813    
1814  /* We can't officially use ESC_t because it is a POSIX reserved identifier  /* We can't officially use ESC_t because it is a POSIX reserved identifier
1815  (presumably because of all the others like size_t). */  (presumably because of all the others like size_t). */
1816    
1817  #ifndef ESC_tee  #ifndef ESC_tee
1818  #define ESC_tee '\t'  #define ESC_tee CHAR_HT
1819  #endif  #endif
1820    
1821  /* Codes for different types of Unicode property */  /* Codes for different types of Unicode property */
1822    
1823  #define PT_ANY        0    /* Any property - matches all chars */  #define PT_ANY        0    /* Any property - matches all chars */
1824  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */
1825  #define PT_GC         2    /* General characteristic (e.g. L) */  #define PT_GC         2    /* Specified general characteristic (e.g. L) */
1826  #define PT_PC         3    /* Particular characteristic (e.g. Lu) */  #define PT_PC         3    /* Specified particular characteristic (e.g. Lu) */
1827  #define PT_SC         4    /* Script (e.g. Han) */  #define PT_SC         4    /* Script (e.g. Han) */
1828    #define PT_ALNUM      5    /* Alphanumeric - the union of L and N */
1829    #define PT_SPACE      6    /* Perl space - Z plus 9,10,12,13 */
1830    #define PT_PXSPACE    7    /* POSIX space - Z plus 9,10,11,12,13 */
1831    #define PT_WORD       8    /* Word - L plus N plus underscore */
1832    #define PT_CLIST      9    /* Pseudo-property: match character list */
1833    
1834  /* Flag bits and data types for the extended class (OP_XCLASS) for classes that  /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
1835  contain UTF-8 characters with values greater than 255. */  contain characters with values greater than 255. */
1836    
1837  #define XCL_NOT    0x01    /* Flag: this is a negative class */  #define XCL_NOT    0x01    /* Flag: this is a negative class */
1838  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */
# Line 594  contain UTF-8 characters with values gre Line 1845  contain UTF-8 characters with values gre
1845    
1846  /* These are escaped items that aren't just an encoding of a particular data  /* These are escaped items that aren't just an encoding of a particular data
1847  value such as \n. They must have non-zero values, as check_escape() returns  value such as \n. They must have non-zero values, as check_escape() returns
1848  their negation. Also, they must appear in the same order as in the opcode  0 for a data character.  Also, they must appear in the same order as in the opcode
1849  definitions below, up to ESC_z. There's a dummy for OP_ANY because it  definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
1850  corresponds to "." rather than an escape sequence. The final one must be  corresponds to "." in DOTALL mode rather than an escape sequence. It is also
1851  ESC_REF as subsequent values are used for backreferences (\1, \2, \3, etc).  used for [^] in JavaScript compatibility mode, and for \C in non-utf mode. In
1852  There are two tests in the code for an escape greater than ESC_b and less than  non-DOTALL mode, "." behaves like \N.
1853  ESC_Z to detect the types that may be repeated. These are the types that  
1854  consume characters. If any new escapes are put in between that don't consume a  The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
1855  character, that code will have to change. */  when PCRE_UCP is set and replacement of \d etc by \p sequences is required.
1856    They must be contiguous, and remain in order so that the replacements can be
1857    looked up from a table.
1858    
1859    Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in
1860    check_escape(). There are two tests in the code for an escape
1861    greater than ESC_b and less than ESC_Z to detect the types that may be
1862    repeated. These are the types that consume characters. If any new escapes are
1863    put in between that don't consume a character, that code will have to change.
1864    */
1865    
1866  enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,  enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1867         ESC_W, ESC_w, ESC_dum1, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H, ESC_h,         ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1868         ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, ESC_E, ESC_Q, ESC_k, ESC_REF };         ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
1869           ESC_E, ESC_Q, ESC_g, ESC_k,
1870           ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu };
1871    
1872  /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to  /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
1873  OP_EOD must correspond in order to the list of escapes immediately above.  OP_EOD must correspond in order to the list of escapes immediately above.
1874    
1875  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1876  that follow must also be updated to match. There is also a table called  that follow must also be updated to match. There are also tables called
1877  "coptable" in pcre_dfa_exec.c that must be updated. */  "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1878    
1879  enum {  enum {
1880    OP_END,            /* 0 End of pattern */    OP_END,            /* 0 End of pattern */
# Line 631  enum { Line 1892  enum {
1892    OP_WHITESPACE,         /*  9 \s */    OP_WHITESPACE,         /*  9 \s */
1893    OP_NOT_WORDCHAR,       /* 10 \W */    OP_NOT_WORDCHAR,       /* 10 \W */
1894    OP_WORDCHAR,           /* 11 \w */    OP_WORDCHAR,           /* 11 \w */
   OP_ANY,            /* 12 Match any character */  
   OP_ANYBYTE,        /* 13 Match any byte (\C); different to OP_ANY for UTF-8 */  
   OP_NOTPROP,        /* 14 \P (not Unicode property) */  
   OP_PROP,           /* 15 \p (Unicode property) */  
   OP_ANYNL,          /* 16 \R (any newline sequence) */  
   OP_NOT_HSPACE,     /* 17 \H (not horizontal whitespace) */  
   OP_HSPACE,         /* 18 \h (horizontal whitespace) */  
   OP_NOT_VSPACE,     /* 19 \V (not vertical whitespace) */  
   OP_VSPACE,         /* 20 \v (vertical whitespace) */  
   OP_EXTUNI,         /* 21 \X (extended Unicode sequence */  
   OP_EODN,           /* 22 End of data or \n at end of data: \Z. */  
   OP_EOD,            /* 23 End of data: \z */  
   
   OP_OPT,            /* 24 Set runtime options */  
   OP_CIRC,           /* 25 Start of line - varies with multiline switch */  
   OP_DOLL,           /* 26 End of line - varies with multiline switch */  
   OP_CHAR,           /* 27 Match one character, casefully */  
   OP_CHARNC,         /* 28 Match one character, caselessly */  
   OP_NOT,            /* 29 Match one character, not the following one */  
   
   OP_STAR,           /* 30 The maximizing and minimizing versions of */  
   OP_MINSTAR,        /* 31 these six opcodes must come in pairs, with */  
   OP_PLUS,           /* 32 the minimizing one second. */  
   OP_MINPLUS,        /* 33 This first set applies to single characters.*/  
   OP_QUERY,          /* 34 */  
   OP_MINQUERY,       /* 35 */  
   
   OP_UPTO,           /* 36 From 0 to n matches */  
   OP_MINUPTO,        /* 37 */  
   OP_EXACT,          /* 38 Exactly n matches */  
   
   OP_POSSTAR,        /* 39 Possessified star */  
   OP_POSPLUS,        /* 40 Possessified plus */  
   OP_POSQUERY,       /* 41 Posesssified query */  
   OP_POSUPTO,        /* 42 Possessified upto */  
   
   OP_NOTSTAR,        /* 43 The maximizing and minimizing versions of */  
   OP_NOTMINSTAR,     /* 44 these six opcodes must come in pairs, with */  
   OP_NOTPLUS,        /* 45 the minimizing one second. They must be in */  
   OP_NOTMINPLUS,     /* 46 exactly the same order as those above. */  
   OP_NOTQUERY,       /* 47 This set applies to "not" single characters. */  
   OP_NOTMINQUERY,    /* 48 */  
   
   OP_NOTUPTO,        /* 49 From 0 to n matches */  
   OP_NOTMINUPTO,     /* 50 */  
   OP_NOTEXACT,       /* 51 Exactly n matches */  
   
   OP_NOTPOSSTAR,     /* 52 Possessified versions */  
   OP_NOTPOSPLUS,     /* 53 */  
   OP_NOTPOSQUERY,    /* 54 */  
   OP_NOTPOSUPTO,     /* 55 */  
   
   OP_TYPESTAR,       /* 56 The maximizing and minimizing versions of */  
   OP_TYPEMINSTAR,    /* 57 these six opcodes must come in pairs, with */  
   OP_TYPEPLUS,       /* 58 the minimizing one second. These codes must */  
   OP_TYPEMINPLUS,    /* 59 be in exactly the same order as those above. */  
   OP_TYPEQUERY,      /* 60 This set applies to character types such as \d */  
   OP_TYPEMINQUERY,   /* 61 */  
   
   OP_TYPEUPTO,       /* 62 From 0 to n matches */  
   OP_TYPEMINUPTO,    /* 63 */  
   OP_TYPEEXACT,      /* 64 Exactly n matches */  
   
   OP_TYPEPOSSTAR,    /* 65 Possessified versions */  
   OP_TYPEPOSPLUS,    /* 66 */  
   OP_TYPEPOSQUERY,   /* 67 */  
   OP_TYPEPOSUPTO,    /* 68 */  
   
   OP_CRSTAR,         /* 69 The maximizing and minimizing versions of */  
   OP_CRMINSTAR,      /* 70 all these opcodes must come in pairs, with */  
   OP_CRPLUS,         /* 71 the minimizing one second. These codes must */  
   OP_CRMINPLUS,      /* 72 be in exactly the same order as those above. */  
   OP_CRQUERY,        /* 73 These are for character classes and back refs */  
   OP_CRMINQUERY,     /* 74 */  
   OP_CRRANGE,        /* 75 These are different to the three sets above. */  
   OP_CRMINRANGE,     /* 76 */  
   
   OP_CLASS,          /* 77 Match a character class, chars < 256 only */  
   OP_NCLASS,         /* 78 Same, but the bitmap was created from a negative  
                            class - the difference is relevant only when a UTF-8  
                            character > 255 is encountered. */  
   
   OP_XCLASS,         /* 79 Extended class for handling UTF-8 chars within the  
                            class. This does both positive and negative. */  
   
   OP_REF,            /* 80 Match a back reference */  
   OP_RECURSE,        /* 81 Match a numbered subpattern (possibly recursive) */  
   OP_CALLOUT,        /* 82 Call out to external function if provided */  
   
   OP_ALT,            /* 83 Start of alternation */  
   OP_KET,            /* 84 End of group that doesn't have an unbounded repeat */  
   OP_KETRMAX,        /* 85 These two must remain together and in this */  
   OP_KETRMIN,        /* 86 order. They are for groups the repeat for ever. */  
   
   /* The assertions must come before BRA, CBRA, ONCE, and COND.*/  
   
   OP_ASSERT,         /* 87 Positive lookahead */  
   OP_ASSERT_NOT,     /* 88 Negative lookahead */  
   OP_ASSERTBACK,     /* 89 Positive lookbehind */  
   OP_ASSERTBACK_NOT, /* 90 Negative lookbehind */  
   OP_REVERSE,        /* 91 Move pointer back - used in lookbehind assertions */  
   
   /* ONCE, BRA, CBRA, and COND must come after the assertions, with ONCE first,  
   as there's a test for >= ONCE for a subpattern that isn't an assertion. */  
   
   OP_ONCE,           /* 92 Atomic group */  
   OP_BRA,            /* 93 Start of non-capturing bracket */  
   OP_CBRA,           /* 94 Start of capturing bracket */  
   OP_COND,           /* 95 Conditional group */  
1895    
1896    /* These three must follow the previous three, in the same order. There's a    OP_ANY,            /* 12 Match any character except newline (\N) */
1897    check for >= SBRA to distinguish the two sets. */    OP_ALLANY,         /* 13 Match any character */
1898      OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1899      OP_NOTPROP,        /* 15 \P (not Unicode property) */
1900      OP_PROP,           /* 16 \p (Unicode property) */
1901      OP_ANYNL,          /* 17 \R (any newline sequence) */
1902      OP_NOT_HSPACE,     /* 18 \H (not horizontal whitespace) */
1903      OP_HSPACE,         /* 19 \h (horizontal whitespace) */
1904      OP_NOT_VSPACE,     /* 20 \V (not vertical whitespace) */
1905      OP_VSPACE,         /* 21 \v (vertical whitespace) */
1906      OP_EXTUNI,         /* 22 \X (extended Unicode sequence */
1907      OP_EODN,           /* 23 End of data or \n at end of data (\Z) */
1908      OP_EOD,            /* 24 End of data (\z) */
1909    
1910      OP_CIRC,           /* 25 Start of line - not multiline */
1911      OP_CIRCM,          /* 26 Start of line - multiline */
1912      OP_DOLL,           /* 27 End of line - not multiline */
1913      OP_DOLLM,          /* 28 End of line - multiline */
1914      OP_CHAR,           /* 29 Match one character, casefully */
1915      OP_CHARI,          /* 30 Match one character, caselessly */
1916      OP_NOT,            /* 31 Match one character, not the given one, casefully */
1917      OP_NOTI,           /* 32 Match one character, not the given one, caselessly */
1918    
1919      /* The following sets of 13 opcodes must always be kept in step because
1920      the offset from the first one is used to generate the others. */
1921    
1922      /**** Single characters, caseful, must precede the caseless ones ****/
1923    
1924      OP_STAR,           /* 33 The maximizing and minimizing versions of */
1925      OP_MINSTAR,        /* 34 these six opcodes must come in pairs, with */
1926      OP_PLUS,           /* 35 the minimizing one second. */
1927      OP_MINPLUS,        /* 36 */
1928      OP_QUERY,          /* 37 */
1929      OP_MINQUERY,       /* 38 */
1930    
1931      OP_UPTO,           /* 39 From 0 to n matches of one character, caseful*/
1932      OP_MINUPTO,        /* 40 */
1933      OP_EXACT,          /* 41 Exactly n matches */
1934    
1935      OP_POSSTAR,        /* 42 Possessified star, caseful */
1936      OP_POSPLUS,        /* 43 Possessified plus, caseful */
1937      OP_POSQUERY,       /* 44 Posesssified query, caseful */
1938      OP_POSUPTO,        /* 45 Possessified upto, caseful */
1939    
1940      /**** Single characters, caseless, must follow the caseful ones */
1941    
1942      OP_STARI,          /* 46 */
1943      OP_MINSTARI,       /* 47 */
1944      OP_PLUSI,          /* 48 */
1945      OP_MINPLUSI,       /* 49 */
1946      OP_QUERYI,         /* 50 */
1947      OP_MINQUERYI,      /* 51 */
1948    
1949      OP_UPTOI,          /* 52 From 0 to n matches of one character, caseless */
1950      OP_MINUPTOI,       /* 53 */
1951      OP_EXACTI,         /* 54 */
1952    
1953      OP_POSSTARI,       /* 55 Possessified star, caseless */
1954      OP_POSPLUSI,       /* 56 Possessified plus, caseless */
1955      OP_POSQUERYI,      /* 57 Posesssified query, caseless */
1956      OP_POSUPTOI,       /* 58 Possessified upto, caseless */
1957    
1958      /**** The negated ones must follow the non-negated ones, and match them ****/
1959      /**** Negated single character, caseful; must precede the caseless ones ****/
1960    
1961      OP_NOTSTAR,        /* 59 The maximizing and minimizing versions of */
1962      OP_NOTMINSTAR,     /* 60 these six opcodes must come in pairs, with */
1963      OP_NOTPLUS,        /* 61 the minimizing one second. They must be in */
1964      OP_NOTMINPLUS,     /* 62 exactly the same order as those above. */
1965      OP_NOTQUERY,       /* 63 */
1966      OP_NOTMINQUERY,    /* 64 */
1967    
1968      OP_NOTUPTO,        /* 65 From 0 to n matches, caseful */
1969      OP_NOTMINUPTO,     /* 66 */
1970      OP_NOTEXACT,       /* 67 Exactly n matches */
1971    
1972      OP_NOTPOSSTAR,     /* 68 Possessified versions, caseful */
1973      OP_NOTPOSPLUS,     /* 69 */
1974      OP_NOTPOSQUERY,    /* 70 */
1975      OP_NOTPOSUPTO,     /* 71 */
1976    
1977      /**** Negated single character, caseless; must follow the caseful ones ****/
1978    
1979      OP_NOTSTARI,       /* 72 */
1980      OP_NOTMINSTARI,    /* 73 */
1981      OP_NOTPLUSI,       /* 74 */
1982      OP_NOTMINPLUSI,    /* 75 */
1983      OP_NOTQUERYI,      /* 76 */
1984      OP_NOTMINQUERYI,   /* 77 */
1985    
1986      OP_NOTUPTOI,       /* 78 From 0 to n matches, caseless */
1987      OP_NOTMINUPTOI,    /* 79 */
1988      OP_NOTEXACTI,      /* 80 Exactly n matches */
1989    
1990      OP_NOTPOSSTARI,    /* 81 Possessified versions, caseless */
1991      OP_NOTPOSPLUSI,    /* 82 */
1992      OP_NOTPOSQUERYI,   /* 83 */
1993      OP_NOTPOSUPTOI,    /* 84 */
1994    
1995      /**** Character types ****/
1996    
1997      OP_TYPESTAR,       /* 85 The maximizing and minimizing versions of */
1998      OP_TYPEMINSTAR,    /* 86 these six opcodes must come in pairs, with */
1999      OP_TYPEPLUS,       /* 87 the minimizing one second. These codes must */
2000      OP_TYPEMINPLUS,    /* 88 be in exactly the same order as those above. */
2001      OP_TYPEQUERY,      /* 89 */
2002      OP_TYPEMINQUERY,   /* 90 */
2003    
2004      OP_TYPEUPTO,       /* 91 From 0 to n matches */
2005      OP_TYPEMINUPTO,    /* 92 */
2006      OP_TYPEEXACT,      /* 93 Exactly n matches */
2007    
2008      OP_TYPEPOSSTAR,    /* 94 Possessified versions */
2009      OP_TYPEPOSPLUS,    /* 95 */
2010      OP_TYPEPOSQUERY,   /* 96 */
2011      OP_TYPEPOSUPTO,    /* 97 */
2012    
2013      /* These are used for character classes and back references; only the
2014      first six are the same as the sets above. */
2015    
2016      OP_CRSTAR,         /* 98 The maximizing and minimizing versions of */
2017      OP_CRMINSTAR,      /* 99 all these opcodes must come in pairs, with */
2018      OP_CRPLUS,         /* 100 the minimizing one second. These codes must */
2019      OP_CRMINPLUS,      /* 101 be in exactly the same order as those above. */
2020      OP_CRQUERY,        /* 102 */
2021      OP_CRMINQUERY,     /* 103 */
2022    
2023      OP_CRRANGE,        /* 104 These are different to the three sets above. */
2024      OP_CRMINRANGE,     /* 105 */
2025    
2026      /* End of quantifier opcodes */
2027    
2028      OP_CLASS,          /* 106 Match a character class, chars < 256 only */
2029      OP_NCLASS,         /* 107 Same, but the bitmap was created from a negative
2030                                  class - the difference is relevant only when a
2031                                  character > 255 is encountered. */
2032      OP_XCLASS,         /* 108 Extended class for handling > 255 chars within the
2033                                  class. This does both positive and negative. */
2034      OP_REF,            /* 109 Match a back reference, casefully */
2035      OP_REFI,           /* 110 Match a back reference, caselessly */
2036      OP_RECURSE,        /* 111 Match a numbered subpattern (possibly recursive) */
2037      OP_CALLOUT,        /* 112 Call out to external function if provided */
2038    
2039      OP_ALT,            /* 113 Start of alternation */
2040      OP_KET,            /* 114 End of group that doesn't have an unbounded repeat */
2041      OP_KETRMAX,        /* 115 These two must remain together and in this */
2042      OP_KETRMIN,        /* 116 order. They are for groups the repeat for ever. */
2043      OP_KETRPOS,        /* 117 Possessive unlimited repeat. */
2044    
2045      /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
2046      asserts must remain in order. */
2047    
2048      OP_REVERSE,        /* 118 Move pointer back - used in lookbehind assertions */
2049      OP_ASSERT,         /* 119 Positive lookahead */
2050      OP_ASSERT_NOT,     /* 120 Negative lookahead */
2051      OP_ASSERTBACK,     /* 121 Positive lookbehind */
2052      OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
2053    
2054      /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
2055      after the assertions, with ONCE first, as there's a test for >= ONCE for a
2056      subpattern that isn't an assertion. The POS versions must immediately follow
2057      the non-POS versions in each case. */
2058    
2059      OP_ONCE,           /* 123 Atomic group, contains captures */
2060      OP_ONCE_NC,        /* 124 Atomic group containing no captures */
2061      OP_BRA,            /* 125 Start of non-capturing bracket */
2062      OP_BRAPOS,         /* 126 Ditto, with unlimited, possessive repeat */
2063      OP_CBRA,           /* 127 Start of capturing bracket */
2064      OP_CBRAPOS,        /* 128 Ditto, with unlimited, possessive repeat */
2065      OP_COND,           /* 129 Conditional group */
2066    
2067    OP_SBRA,           /* 96 Start of non-capturing bracket, check empty  */    /* These five must follow the previous five, in the same order. There's a
2068    OP_SCBRA,          /* 97 Start of capturing bracket, check empty */    check for >= SBRA to distinguish the two sets. */
   OP_SCOND,          /* 98 Conditional group, check empty */  
   
   OP_CREF,           /* 99 Used to hold a capture number as condition */  
   OP_RREF,           /* 100 Used to hold a recursion number as condition */  
   OP_DEF,            /* 101 The DEFINE condition */  
2069    
2070    OP_BRAZERO,        /* 102 These two must remain together and in this */    OP_SBRA,           /* 130 Start of non-capturing bracket, check empty  */
2071    OP_BRAMINZERO,     /* 103 order. */    OP_SBRAPOS,        /* 131 Ditto, with unlimited, possessive repeat */
2072      OP_SCBRA,          /* 132 Start of capturing bracket, check empty */
2073      OP_SCBRAPOS,       /* 133 Ditto, with unlimited, possessive repeat */
2074      OP_SCOND,          /* 134 Conditional group, check empty */
2075    
2076      /* The next two pairs must (respectively) be kept together. */
2077    
2078      OP_CREF,           /* 135 Used to hold a capture number as condition */
2079      OP_NCREF,          /* 136 Same, but generated by a name reference*/
2080      OP_RREF,           /* 137 Used to hold a recursion number as condition */
2081      OP_NRREF,          /* 138 Same, but generated by a name reference*/
2082      OP_DEF,            /* 139 The DEFINE condition */
2083    
2084      OP_BRAZERO,        /* 140 These two must remain together and in this */
2085      OP_BRAMINZERO,     /* 141 order. */
2086      OP_BRAPOSZERO,     /* 142 */
2087    
2088    /* These are backtracking control verbs */    /* These are backtracking control verbs */
2089    
2090    OP_PRUNE,          /* 104 */    OP_MARK,           /* 143 always has an argument */
2091    OP_SKIP,           /* 105 */    OP_PRUNE,          /* 144 */
2092    OP_THEN,           /* 106 */    OP_PRUNE_ARG,      /* 145 same, but with argument */
2093    OP_COMMIT,         /* 107 */    OP_SKIP,           /* 146 */
2094      OP_SKIP_ARG,       /* 147 same, but with argument */
2095      OP_THEN,           /* 148 */
2096      OP_THEN_ARG,       /* 149 same, but with argument */
2097      OP_COMMIT,         /* 150 */
2098    
2099    /* These are forced failure and success verbs */    /* These are forced failure and success verbs */
2100    
2101    OP_FAIL,           /* 108 */    OP_FAIL,           /* 151 */
2102    OP_ACCEPT          /* 109 */    OP_ACCEPT,         /* 152 */
2103      OP_ASSERT_ACCEPT,  /* 153 Used inside assertions */
2104      OP_CLOSE,          /* 154 Used before OP_ACCEPT to close open captures */
2105    
2106      /* This is used to skip a subpattern with a {0} quantifier */
2107    
2108      OP_SKIPZERO,       /* 155 */
2109    
2110      /* This is not an opcode, but is used to check that tables indexed by opcode
2111      are the correct length, in order to catch updating errors - there have been
2112      some in the past. */
2113    
2114      OP_TABLE_LENGTH
2115  };  };
2116    
2117    /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
2118    definitions that follow must also be updated to match. There are also tables
2119    called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
2120    
2121    
2122  /* This macro defines textual names for all the opcodes. These are used only  /* This macro defines textual names for all the opcodes. These are used only
2123  for debugging. The macro is referenced only in pcre_printint.c. */  for debugging, and some of them are only partial names. The macro is referenced
2124    only in pcre_printint.c, which fills out the full names in many cases (and in
2125    some cases doesn't actually use these names at all). */
2126    
2127  #define OP_NAME_LIST \  #define OP_NAME_LIST \
2128    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \
2129    "\\S", "\\s", "\\W", "\\w", "Any", "Anybyte",                   \    "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte",         \
2130    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \
2131    "extuni",  "\\Z", "\\z",                                        \    "extuni",  "\\Z", "\\z",                                        \
2132    "Opt", "^", "$", "char", "charnc", "not",                       \    "^", "^", "$", "$", "char", "chari", "not", "noti",             \
2133    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2134      "{", "{", "{",                                                  \
2135    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2136    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2137      "{", "{", "{",                                                  \
2138      "*+","++", "?+", "{",                                           \
2139      "*", "*?", "+", "+?", "?", "??",                                \
2140      "{", "{", "{",                                                  \
2141      "*+","++", "?+", "{",                                           \
2142      "*", "*?", "+", "+?", "?", "??",                                \
2143      "{", "{", "{",                                                  \
2144    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2145    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \
2146    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2147    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \
2148    "class", "nclass", "xclass", "Ref", "Recurse", "Callout",       \    "class", "nclass", "xclass", "Ref", "Refi",                     \
2149    "Alt", "Ket", "KetRmax", "KetRmin", "Assert", "Assert not",     \    "Recurse", "Callout",                                           \
2150    "AssertB", "AssertB not", "Reverse",                            \    "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos",                  \
2151    "Once", "Bra", "CBra", "Cond", "SBra", "SCBra", "SCond",        \    "Reverse", "Assert", "Assert not", "AssertB", "AssertB not",    \
2152    "Cond ref", "Cond rec", "Cond def", "Brazero", "Braminzero",    \    "Once", "Once_NC",                                              \
2153    "*PRUNE", "*SKIP", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT"    "Bra", "BraPos", "CBra", "CBraPos",                             \
2154      "Cond",                                                         \
2155      "SBra", "SBraPos", "SCBra", "SCBraPos",                         \
2156      "SCond",                                                        \
2157      "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def",   \
2158      "Brazero", "Braminzero", "Braposzero",                          \
2159      "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP",                  \
2160      "*THEN", "*THEN", "*COMMIT", "*FAIL",                           \
2161      "*ACCEPT", "*ASSERT_ACCEPT",                                    \
2162      "Close", "Skip zero"
2163    
2164    
2165  /* This macro defines the length of fixed length operations in the compiled  /* This macro defines the length of fixed length operations in the compiled
# Line 806  in UTF-8 mode. The code that uses this t Line 2175  in UTF-8 mode. The code that uses this t
2175    1,                             /* End                                    */ \    1,                             /* End                                    */ \
2176    1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \    1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \
2177    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \
2178    1, 1,                          /* Any, Anybyte                           */ \    1, 1, 1,                       /* Any, AllAny, Anybyte                   */ \
2179    3, 3, 1,                       /* NOTPROP, PROP, EXTUNI                  */ \    3, 3,                          /* \P, \p                                 */ \
2180    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \
2181    1, 1, 2, 1, 1,                 /* \Z, \z, Opt, ^, $                      */ \    1,                             /* \X                                     */ \
2182      1, 1, 1, 1, 1, 1,              /* \Z, \z, ^, ^M, $, $M                   */ \
2183    2,                             /* Char  - the minimum length             */ \    2,                             /* Char  - the minimum length             */ \
2184    2,                             /* Charnc  - the minimum length           */ \    2,                             /* Chari  - the minimum length            */ \
2185    2,                             /* not                                    */ \    2,                             /* not                                    */ \
2186    /* Positive single-char repeats                            ** These are  */ \    2,                             /* noti                                   */ \
2187    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??      ** minima in  */ \    /* Positive single-char repeats                             ** These are */ \
2188    4, 4, 4,                       /* upto, minupto, exact     ** UTF-8 mode */ \    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??       ** minima in */ \
2189    2, 2, 2, 4,                    /* *+, ++, ?+, upto+                      */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto, minupto             ** mode      */ \
2190      2+IMM2_SIZE,                   /* exact                                  */ \
2191      2, 2, 2, 2+IMM2_SIZE,          /* *+, ++, ?+, upto+                      */ \
2192      2, 2, 2, 2, 2, 2,              /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8     */ \
2193      2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto I, minupto I                      */ \
2194      2+IMM2_SIZE,                   /* exact I                                */ \
2195      2, 2, 2, 2+IMM2_SIZE,          /* *+I, ++I, ?+I, upto+I                  */ \
2196    /* Negative single-char repeats - only for chars < 256                   */ \    /* Negative single-char repeats - only for chars < 256                   */ \
2197    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \
2198    4, 4, 4,                       /* NOT upto, minupto, exact               */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto, minupto                      */ \
2199    2, 2, 2, 4,                    /* Possessive *, +, ?, upto               */ \    2+IMM2_SIZE,                   /* NOT exact                              */ \
2200      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *, +, ?, upto           */ \
2201      2, 2, 2, 2, 2, 2,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */ \
2202      2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto I, minupto I                  */ \
2203      2+IMM2_SIZE,                   /* NOT exact I                            */ \
2204      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *I, +I, ?I, upto I      */ \
2205    /* Positive type repeats                                                 */ \    /* Positive type repeats                                                 */ \
2206    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \
2207    4, 4, 4,                       /* Type upto, minupto, exact              */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* Type upto, minupto                     */ \
2208    2, 2, 2, 4,                    /* Possessive *+, ++, ?+, upto+           */ \    2+IMM2_SIZE,                   /* Type exact                             */ \
2209      2, 2, 2, 2+IMM2_SIZE,          /* Possessive *+, ++, ?+, upto+           */ \
2210    /* Character class & ref repeats                                         */ \    /* Character class & ref repeats                                         */ \
2211    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \
2212    5, 5,                          /* CRRANGE, CRMINRANGE                    */ \    1+2*IMM2_SIZE, 1+2*IMM2_SIZE,  /* CRRANGE, CRMINRANGE                    */ \
2213   33,                             /* CLASS                                  */ \    1+(32/sizeof(pcre_uchar)),     /* CLASS                                  */ \
2214   33,                             /* NCLASS                                 */ \    1+(32/sizeof(pcre_uchar)),     /* NCLASS                                 */ \
2215    0,                             /* XCLASS - variable length               */ \    0,                             /* XCLASS - variable length               */ \
2216    3,                             /* REF                                    */ \    1+IMM2_SIZE,                   /* REF                                    */ \
2217      1+IMM2_SIZE,                   /* REFI                                   */ \
2218    1+LINK_SIZE,                   /* RECURSE                                */ \    1+LINK_SIZE,                   /* RECURSE                                */ \
2219    2+2*LINK_SIZE,                 /* CALLOUT                                */ \    2+2*LINK_SIZE,                 /* CALLOUT                                */ \
2220    1+LINK_SIZE,                   /* Alt                                    */ \    1+LINK_SIZE,                   /* Alt                                    */ \
2221    1+LINK_SIZE,                   /* Ket                                    */ \    1+LINK_SIZE,                   /* Ket                                    */ \
2222    1+LINK_SIZE,                   /* KetRmax                                */ \    1+LINK_SIZE,                   /* KetRmax                                */ \
2223    1+LINK_SIZE,                   /* KetRmin                                */ \    1+LINK_SIZE,                   /* KetRmin                                */ \
2224      1+LINK_SIZE,                   /* KetRpos                                */ \
2225      1+LINK_SIZE,                   /* Reverse                                */ \
2226    1+LINK_SIZE,                   /* Assert                                 */ \    1+LINK_SIZE,                   /* Assert                                 */ \
2227    1+LINK_SIZE,                   /* Assert not                             */ \    1+LINK_SIZE,                   /* Assert not                             */ \
2228    1+LINK_SIZE,                   /* Assert behind                          */ \    1+LINK_SIZE,                   /* Assert behind                          */ \
2229    1+LINK_SIZE,                   /* Assert behind not                      */ \    1+LINK_SIZE,                   /* Assert behind not                      */ \
   1+LINK_SIZE,                   /* Reverse                                */ \  
2230    1+LINK_SIZE,                   /* ONCE                                   */ \    1+LINK_SIZE,                   /* ONCE                                   */ \
2231      1+LINK_SIZE,                   /* ONCE_NC                                */ \
2232    1+LINK_SIZE,                   /* BRA                                    */ \    1+LINK_SIZE,                   /* BRA                                    */ \
2233    3+LINK_SIZE,                   /* CBRA                                   */ \    1+LINK_SIZE,                   /* BRAPOS                                 */ \
2234      1+LINK_SIZE+IMM2_SIZE,         /* CBRA                                   */ \
2235      1+LINK_SIZE+IMM2_SIZE,         /* CBRAPOS                                */ \
2236    1+LINK_SIZE,                   /* COND                                   */ \    1+LINK_SIZE,                   /* COND                                   */ \
2237    1+LINK_SIZE,                   /* SBRA                                   */ \    1+LINK_SIZE,                   /* SBRA                                   */ \
2238    3+LINK_SIZE,                   /* SCBRA                                  */ \    1+LINK_SIZE,                   /* SBRAPOS                                */ \
2239      1+LINK_SIZE+IMM2_SIZE,         /* SCBRA                                  */ \
2240      1+LINK_SIZE+IMM2_SIZE,         /* SCBRAPOS                               */ \
2241    1+LINK_SIZE,                   /* SCOND                                  */ \    1+LINK_SIZE,                   /* SCOND                                  */ \
2242    3,                             /* CREF                                   */ \    1+IMM2_SIZE, 1+IMM2_SIZE,      /* CREF, NCREF                            */ \
2243    3,                             /* RREF                                   */ \    1+IMM2_SIZE, 1+IMM2_SIZE,      /* RREF, NRREF                            */ \
2244    1,                             /* DEF                                    */ \    1,                             /* DEF                                    */ \
2245    1, 1,                          /* BRAZERO, BRAMINZERO                    */ \    1, 1, 1,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */ \
2246    1, 1, 1, 1,                    /* PRUNE, SKIP, THEN, COMMIT,             */ \    3, 1, 3,                       /* MARK, PRUNE, PRUNE_ARG                 */ \
2247    1, 1                           /* FAIL, ACCEPT                           */    1, 3,                          /* SKIP, SKIP_ARG                         */ \
2248      1, 3,                          /* THEN, THEN_ARG                         */ \
2249      1, 1, 1, 1,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */ \
2250      1+IMM2_SIZE, 1                 /* CLOSE, SKIPZERO                        */
2251    
2252  /* A magic value for OP_RREF to indicate the "any recursion" condition. */  /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
2253    condition. */
2254    
2255  #define RREF_ANY  0xffff  #define RREF_ANY  0xffff
2256    
2257  /* Error code numbers. They are given names so that they can more easily be  /* Compile time error code numbers. They are given names so that they can more
2258  tracked. */  easily be tracked. When a new number is added, the table called eint in
2259    pcreposix.c must be updated. */
2260    
2261  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
2262         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
# Line 871  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4, Line 2264  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,
2264         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
2265         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
2266         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
2267         ERR60, ERR61, ERR62, ERR63 };         ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
2268           ERR70, ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERR77, ERRCOUNT };
2269    
2270    /* JIT compiling modes. The function list is indexed by them. */
2271    enum { JIT_COMPILE, JIT_PARTIAL_SOFT_COMPILE, JIT_PARTIAL_HARD_COMPILE,
2272           JIT_NUMBER_OF_COMPILE_MODES };
2273    
2274  /* The real format of the start of the pcre block; the index of names and the  /* The real format of the start of the pcre block; the index of names and the
2275  code vector run on as long as necessary after the end. We store an explicit  code vector run on as long as necessary after the end. We store an explicit
# Line 887  Because people can now save and re-use c Line 2285  Because people can now save and re-use c
2285  structure should be made at the end, and something earlier (e.g. a new  structure should be made at the end, and something earlier (e.g. a new
2286  flag in the options or one of the dummy fields) should indicate that the new  flag in the options or one of the dummy fields) should indicate that the new
2287  fields are present. Currently PCRE always sets the dummy fields to zero.  fields are present. Currently PCRE always sets the dummy fields to zero.
2288  NOTE NOTE NOTE:  NOTE NOTE NOTE
2289  */  */
2290    
2291  typedef struct real_pcre {  #if defined COMPILE_PCRE8
2292    #define REAL_PCRE real_pcre
2293    #elif defined COMPILE_PCRE16
2294    #define REAL_PCRE real_pcre16
2295    #elif defined COMPILE_PCRE32
2296    #define REAL_PCRE real_pcre32
2297    #endif
2298    
2299    /* It is necessary to fork the struct for 32 bit, since it needs to use
2300     * pcre_uchar for first_char and req_char. Can't put an ifdef inside the
2301     * typedef since pcretest needs access to  the struct of the 8-, 16-
2302     * and 32-bit variants. */
2303    
2304    typedef struct real_pcre8_or_16 {
2305    pcre_uint32 magic_number;    pcre_uint32 magic_number;
2306    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2307    pcre_uint32 options;            /* Public options */    pcre_uint32 options;            /* Public options */
2308    pcre_uint16 flags;              /* Private flags */    pcre_uint16 flags;              /* Private flags */
2309    pcre_uint16 dummy1;             /* For future use */    pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2310    pcre_uint16 top_bracket;    pcre_uint16 top_bracket;        /* Highest numbered group */
2311    pcre_uint16 top_backref;    pcre_uint16 top_backref;        /* Highest numbered back reference */
2312    pcre_uint16 first_byte;    pcre_uint16 first_char;         /* Starting character */
2313    pcre_uint16 req_byte;    pcre_uint16 req_char;           /* This character must be seen */
2314    pcre_uint16 name_table_offset;  /* Offset to name table that follows */    pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2315    pcre_uint16 name_entry_size;    /* Size of any name items */    pcre_uint16 name_entry_size;    /* Size of any name items */
2316    pcre_uint16 name_count;         /* Number of name items */    pcre_uint16 name_count;         /* Number of name items */
2317    pcre_uint16 ref_count;          /* Reference count */    pcre_uint16 ref_count;          /* Reference count */
2318      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2319      const pcre_uint8 *nullpad;      /* NULL padding */
2320    } real_pcre8_or_16;
2321    
2322    typedef struct real_pcre8_or_16 real_pcre;
2323    typedef struct real_pcre8_or_16 real_pcre16;
2324    
2325    const unsigned char *tables;    /* Pointer to tables or NULL for std */  typedef struct real_pcre32 {
2326    const unsigned char *nullpad;   /* NULL padding */    pcre_uint32 magic_number;
2327  } real_pcre;    pcre_uint32 size;               /* Total that was malloced */
2328      pcre_uint32 options;            /* Public options */
2329      pcre_uint16 flags;              /* Private flags */
2330      pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2331      pcre_uint16 top_bracket;        /* Highest numbered group */
2332      pcre_uint16 top_backref;        /* Highest numbered back reference */
2333      pcre_uint32 first_char;         /* Starting character */
2334      pcre_uint32 req_char;           /* This character must be seen */
2335      pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2336      pcre_uint16 name_entry_size;    /* Size of any name items */
2337      pcre_uint16 name_count;         /* Number of name items */
2338      pcre_uint16 ref_count;          /* Reference count */
2339      pcre_uint16 dummy1;             /* for later expansion */
2340      pcre_uint16 dummy2;             /* for later expansion */
2341      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2342      void *nullpad;                  /* for later expansion */
2343    } real_pcre32;
2344    
2345    /* Assert that the size of REAL_PCRE is divisible by 8 */
2346    typedef int __assert_real_pcre_size_divisible_8[(sizeof(REAL_PCRE) % 8) == 0 ? 1 : -1];
2347    
2348    /* Needed in pcretest to access some fields in the real_pcre* structures
2349     * directly. They're unified for 8/16/32 bits since the structs only differ
2350     * after these fields; if that ever changes, need to fork those defines into
2351     * 8/16 and 32 bit versions. */
2352    #define REAL_PCRE_MAGIC(re)     (((REAL_PCRE*)re)->magic_number)
2353    #define REAL_PCRE_SIZE(re)      (((REAL_PCRE*)re)->size)
2354    #define REAL_PCRE_OPTIONS(re)   (((REAL_PCRE*)re)->options)
2355    #define REAL_PCRE_FLAGS(re)     (((REAL_PCRE*)re)->flags)
2356    
2357  /* The format of the block used to store data from pcre_study(). The same  /* The format of the block used to store data from pcre_study(). The same
2358  remark (see NOTE above) about extending this structure applies. */  remark (see NOTE above) about extending this structure applies. */
2359    
2360  typedef struct pcre_study_data {  typedef struct pcre_study_data {
2361    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2362    pcre_uint32 options;    pcre_uint32 flags;              /* Private flags */
2363    uschar start_bits[32];    pcre_uint8 start_bits[32];      /* Starting char bits */
2364      pcre_uint32 minlength;          /* Minimum subject length */
2365  } pcre_study_data;  } pcre_study_data;
2366    
2367    /* Structure for building a chain of open capturing subpatterns during
2368    compiling, so that instructions to close them can be compiled when (*ACCEPT) is
2369    encountered. This is also used to identify subpatterns that contain recursive
2370    back references to themselves, so that they can be made atomic. */
2371    
2372    typedef struct open_capitem {
2373      struct open_capitem *next;    /* Chain link */
2374      pcre_uint16 number;           /* Capture number */
2375      pcre_uint16 flag;             /* Set TRUE if recursive back ref */
2376    } open_capitem;
2377    
2378  /* Structure for passing "static" information around between the functions  /* Structure for passing "static" information around between the functions
2379  doing the compiling, so that they are thread-safe. */  doing the compiling, so that they are thread-safe. */
2380    
2381  typedef struct compile_data {  typedef struct compile_data {
2382    const uschar *lcc;            /* Points to lower casing table */    const pcre_uint8 *lcc;            /* Points to lower casing table */
2383    const uschar *fcc;            /* Points to case-flipping table */    const pcre_uint8 *fcc;            /* Points to case-flipping table */
2384    const uschar *cbits;          /* Points to character type table */    const pcre_uint8 *cbits;          /* Points to character type table */
2385    const uschar *ctypes;         /* Points to table of type maps */    const pcre_uint8 *ctypes;         /* Points to table of type maps */
2386    const uschar *start_workspace;/* The start of working space */    const pcre_uchar *start_workspace;/* The start of working space */
2387    const uschar *start_code;     /* The start of the compiled code */    const pcre_uchar *start_code;     /* The start of the compiled code */
2388    const uschar *start_pattern;  /* The start of the pattern */    const pcre_uchar *start_pattern;  /* The start of the pattern */
2389    const uschar *end_pattern;    /* The end of the pattern */    const pcre_uchar *end_pattern;    /* The end of the pattern */
2390    uschar *hwm;                  /* High watermark of workspace */    open_capitem *open_caps;          /* Chain of open capture items */
2391    uschar *name_table;           /* The name/number table */    pcre_uchar *hwm;                  /* High watermark of workspace */
2392    int  names_found;             /* Number of entries so far */    pcre_uchar *name_table;           /* The name/number table */
2393    int  name_entry_size;         /* Size of each entry */    int  names_found;                 /* Number of entries so far */
2394    int  bracount;                /* Count of capturing parens as we compile */    int  name_entry_size;             /* Size of each entry */
2395    int  final_bracount;          /* Saved value after first pass */    int  workspace_size;              /* Size of workspace */
2396    int  top_backref;             /* Maximum back reference */    int  bracount;                    /* Count of capturing parens as we compile */
2397    unsigned int backref_map;     /* Bitmap of low back refs */    int  final_bracount;              /* Saved value after first pass */
2398    int  external_options;        /* External (initial) options */    int  max_lookbehind;              /* Maximum lookbehind (characters) */
2399    int  external_flags;          /* External flag bits to be set */    int  top_backref;                 /* Maximum back reference */
2400    int  req_varyopt;             /* "After variable item" flag for reqbyte */    unsigned int backref_map;         /* Bitmap of low back refs */
2401    BOOL had_accept;              /* (*ACCEPT) encountered */    int  assert_depth;                /* Depth of nested assertions */
2402    int  nltype;                  /* Newline type */    int  external_options;            /* External (initial) options */
2403    int  nllen;                   /* Newline string length */    int  external_flags;              /* External flag bits to be set */
2404    uschar nl[4];                 /* Newline string when fixed length */    int  req_varyopt;                 /* "After variable item" flag for reqbyte */
2405      BOOL had_accept;                  /* (*ACCEPT) encountered */
2406      BOOL had_pruneorskip;             /* (*PRUNE) or (*SKIP) encountered */
2407      BOOL check_lookbehind;            /* Lookbehinds need later checking */
2408      int  nltype;                      /* Newline type */
2409      int  nllen;                       /* Newline string length */
2410      pcre_uchar nl[4];                 /* Newline string when fixed length */
2411  } compile_data;  } compile_data;
2412    
2413  /* Structure for maintaining a chain of pointers to the currently incomplete  /* Structure for maintaining a chain of pointers to the currently incomplete
2414  branches, for testing for left recursion. */  branches, for testing for left recursion while compiling. */
2415    
2416  typedef struct branch_chain {  typedef struct branch_chain {
2417    struct branch_chain *outer;    struct branch_chain *outer;
2418    uschar *current;    pcre_uchar *current_branch;
2419  } branch_chain;  } branch_chain;
2420    
2421  /* Structure for items in a linked list that represents an explicit recursive  /* Structure for items in a linked list that represents an explicit recursive
2422  call within the pattern. */  call within the pattern; used by pcre_exec(). */
2423    
2424  typedef struct recursion_info {  typedef struct recursion_info {
2425    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
2426    int group_num;                /* Number of group that was called */    int group_num;                  /* Number of group that was called */
2427    const uschar *after_call;     /* "Return value": points after the call in the expr */    int *offset_save;               /* Pointer to start of saved offsets */
2428    USPTR save_start;             /* Old value of mstart */    int saved_max;                  /* Number of saved offsets */
2429    int *offset_save;             /* Pointer to start of saved offsets */    PCRE_PUCHAR subject_position;   /* Position at start of recursion */
   int saved_max;                /* Number of saved offsets */  
2430  } recursion_info;  } recursion_info;
2431    
2432    /* A similar structure for pcre_dfa_exec(). */
2433    
2434    typedef struct dfa_recursion_info {
2435      struct dfa_recursion_info *prevrec;
2436      int group_num;
2437      PCRE_PUCHAR subject_position;
2438    } dfa_recursion_info;
2439    
2440  /* Structure for building a chain of data for holding the values of the subject  /* Structure for building a chain of data for holding the values of the subject
2441  pointer at the start of each subpattern, so as to detect when an empty string  pointer at the start of each subpattern, so as to detect when an empty string
2442  has been matched by a subpattern - to break infinite loops. */  has been matched by a subpattern - to break infinite loops; used by
2443    pcre_exec(). */
2444    
2445  typedef struct eptrblock {  typedef struct eptrblock {
2446    struct eptrblock *epb_prev;    struct eptrblock *epb_prev;
2447    USPTR epb_saved_eptr;    PCRE_PUCHAR epb_saved_eptr;
2448  } eptrblock;  } eptrblock;
2449    
2450    
# Line 984  typedef struct match_data { Line 2455  typedef struct match_data {
2455    unsigned long int match_call_count;      /* As it says */    unsigned long int match_call_count;      /* As it says */
2456    unsigned long int match_limit;           /* As it says */    unsigned long int match_limit;           /* As it says */
2457    unsigned long int match_limit_recursion; /* As it says */    unsigned long int match_limit_recursion; /* As it says */
2458    int   *offset_vector;         /* Offset vector */    int   *offset_vector;           /* Offset vector */
2459    int    offset_end;            /* One past the end */    int    offset_end;              /* One past the end */
2460    int    offset_max;            /* The maximum usable for return data */    int    offset_max;              /* The maximum usable for return data */
2461    int    nltype;                /* Newline type */    int    nltype;                  /* Newline type */
2462    int    nllen;                 /* Newline string length */    int    nllen;                   /* Newline string length */
2463    uschar nl[4];                 /* Newline string when fixed */    int    name_count;              /* Number of names in name table */
2464    const uschar *lcc;            /* Points to lower casing table */    int    name_entry_size;         /* Size of entry in names table */
2465    const uschar *ctypes;         /* Points to table of type maps */    pcre_uchar *name_table;         /* Table of names */
2466    BOOL   offset_overflow;       /* Set if too many extractions */    pcre_uchar nl[4];               /* Newline string when fixed */
2467    BOOL   notbol;                /* NOTBOL flag */    const  pcre_uint8 *lcc;         /* Points to lower casing table */
2468    BOOL   noteol;                /* NOTEOL flag */    const  pcre_uint8 *fcc;         /* Points to case-flipping table */
2469    BOOL   utf8;                  /* UTF8 flag */    const  pcre_uint8 *ctypes;      /* Points to table of type maps */
2470    BOOL   endonly;               /* Dollar not before final \n */    BOOL   offset_overflow;         /* Set if too many extractions */
2471    BOOL   notempty;              /* Empty string match not wanted */    BOOL   notbol;                  /* NOTBOL flag */
2472    BOOL   partial;               /* PARTIAL flag */    BOOL   noteol;                  /* NOTEOL flag */
2473    BOOL   hitend;                /* Hit the end of the subject at some point */    BOOL   utf;                     /* UTF-8 / UTF-16 flag */
2474    BOOL   bsr_anycrlf;           /* \R is just any CRLF, not full Unicode */    BOOL   jscript_compat;          /* JAVASCRIPT_COMPAT flag */
2475    const uschar *start_code;     /* For use when recursing */    BOOL   use_ucp;                 /* PCRE_UCP flag */
2476    USPTR  start_subject;         /* Start of the subject string */    BOOL   endonly;                 /* Dollar not before final \n */
2477    USPTR  end_subject;           /* End of the subject string */    BOOL   notempty;                /* Empty string match not wanted */
2478    USPTR  start_match_ptr;       /* Start of matched string */    BOOL   notempty_atstart;        /* Empty string match at start not wanted */
2479    USPTR  end_match_ptr;         /* Subject position at end match */    BOOL   hitend;                  /* Hit the end of the subject at some point */
2480    int    end_offset_top;        /* Highwater mark at end of match */    BOOL   bsr_anycrlf;             /* \R is just any CRLF, not full Unicode */
2481    int    capture_last;          /* Most recent capture number */    BOOL   hasthen;                 /* Pattern contains (*THEN) */
2482    int    start_offset;          /* The start offset value */    BOOL   ignore_skip_arg;         /* For re-run when SKIP name not found */
2483    eptrblock *eptrchain;         /* Chain of eptrblocks for tail recursions */    const  pcre_uchar *start_code;  /* For use when recursing */
2484    int    eptrn;                 /* Next free eptrblock */    PCRE_PUCHAR start_subject;      /* Start of the subject string */
2485    recursion_info *recursive;    /* Linked list of recursion data */    PCRE_PUCHAR end_subject;        /* End of the subject string */
2486    void  *callout_data;          /* To pass back to callouts */    PCRE_PUCHAR start_match_ptr;    /* Start of matched string */
2487      PCRE_PUCHAR end_match_ptr;      /* Subject position at end match */
2488      PCRE_PUCHAR start_used_ptr;     /* Earliest consulted character */
2489      int    partial;                 /* PARTIAL options */
2490      int    end_offset_top;          /* Highwater mark at end of match */
2491      int    capture_last;            /* Most recent capture number */
2492      int    start_offset;            /* The start offset value */
2493      int    match_function_type;     /* Set for certain special calls of MATCH() */
2494      eptrblock *eptrchain;           /* Chain of eptrblocks for tail recursions */
2495      int    eptrn;                   /* Next free eptrblock */
2496      recursion_info *recursive;      /* Linked list of recursion data */
2497      void  *callout_data;            /* To pass back to callouts */
2498      const  pcre_uchar *mark;        /* Mark pointer to pass back on success */
2499      const  pcre_uchar *nomatch_mark;/* Mark pointer to pass back on failure */
2500      const  pcre_uchar *once_target; /* Where to back up to for atomic groups */
2501    #ifdef NO_RECURSE
2502      void  *match_frames_base;       /* For remembering malloc'd frames */
2503    #endif
2504  } match_data;  } match_data;
2505    
2506  /* A similar structure is used for the same purpose by the DFA matching  /* A similar structure is used for the same purpose by the DFA matching
2507  functions. */  functions. */
2508    
2509  typedef struct dfa_match_data {  typedef struct dfa_match_data {
2510    const uschar *start_code;     /* Start of the compiled pattern */    const pcre_uchar *start_code;     /* Start of the compiled pattern */
2511    const uschar *start_subject;  /* Start of the subject string */    const pcre_uchar *start_subject ; /* Start of the subject string */
2512    const uschar *end_subject;    /* End of subject string */    const pcre_uchar *end_subject;    /* End of subject string */
2513    const uschar *tables;         /* Character tables */    const pcre_uchar *start_used_ptr; /* Earliest consulted character */
2514    int   moptions;               /* Match options */    const pcre_uint8 *tables;         /* Character tables */
2515    int   poptions;               /* Pattern options */    int   start_offset;               /* The start offset value */
2516    int    nltype;                /* Newline type */    int   moptions;                   /* Match options */
2517    int    nllen;                 /* Newline string length */    int   poptions;                   /* Pattern options */
2518    uschar nl[4];                 /* Newline string when fixed */    int   nltype;                     /* Newline type */
2519    void  *callout_data;          /* To pass back to callouts */    int   nllen;                      /* Newline string length */
2520      pcre_uchar nl[4];                 /* Newline string when fixed */
2521      void *callout_data;               /* To pass back to callouts */
2522      dfa_recursion_info *recursive;    /* Linked list of recursion data */
2523  } dfa_match_data;  } dfa_match_data;
2524    
2525  /* Bit definitions for entries in the pcre_ctypes table. */  /* Bit definitions for entries in the pcre_ctypes table. */
# Line 1064  total length. */ Line 2555  total length. */
2555  #define ctypes_offset (cbits_offset + cbit_length)  #define ctypes_offset (cbits_offset + cbit_length)
2556  #define tables_length (ctypes_offset + 256)  #define tables_length (ctypes_offset + 256)
2557    
2558    /* Internal function and data prefixes. */
2559    
2560    #if defined COMPILE_PCRE8
2561    #ifndef PUBL
2562    #define PUBL(name) pcre_##name
2563    #endif
2564    #ifndef PRIV
2565    #define PRIV(name) _pcre_##name
2566    #endif
2567    #elif defined COMPILE_PCRE16
2568    #ifndef PUBL
2569    #define PUBL(name) pcre16_##name
2570    #endif
2571    #ifndef PRIV
2572    #define PRIV(name) _pcre16_##name
2573    #endif
2574    #elif defined COMPILE_PCRE32
2575    #ifndef PUBL
2576    #define PUBL(name) pcre32_##name
2577    #endif
2578    #ifndef PRIV
2579    #define PRIV(name) _pcre32_##name
2580    #endif
2581    #else
2582    #error Unsupported compiling mode
2583    #endif /* COMPILE_PCRE[8|16|32] */
2584    
2585  /* Layout of the UCP type table that translates property names into types and  /* Layout of the UCP type table that translates property names into types and
2586  codes. Each entry used to point directly to a name, but to reduce the number of  codes. Each entry used to point directly to a name, but to reduce the number of
2587  relocations in shared libraries, it now has an offset into a single string  relocations in shared libraries, it now has an offset into a single string
# Line 1081  of the exported public functions. They h Line 2599  of the exported public functions. They h
2599  but are not part of the PCRE public API. The data for these tables is in the  but are not part of the PCRE public API. The data for these tables is in the
2600  pcre_tables.c module. */  pcre_tables.c module. */
2601    
2602  extern const int    _pcre_utf8_table1[];  #ifdef COMPILE_PCRE8
2603  extern const int    _pcre_utf8_table2[];  extern const int            PRIV(utf8_table1)[];
2604  extern const int    _pcre_utf8_table3[];  extern const int            PRIV(utf8_table1_size);
2605  extern const uschar _pcre_utf8_table4[];  extern const int            PRIV(utf8_table2)[];
2606    extern const int            PRIV(utf8_table3)[];
2607  extern const int    _pcre_utf8_table1_size;  extern const pcre_uint8     PRIV(utf8_table4)[];
2608    #endif /* COMPILE_PCRE8 */
2609    
2610    extern const char           PRIV(utt_names)[];
2611    extern const ucp_type_table PRIV(utt)[];
2612    extern const int            PRIV(utt_size);
2613    
2614  extern const char   _pcre_utt_names[];  extern const pcre_uint8     PRIV(OP_lengths)[];
2615  extern const ucp_type_table _pcre_utt[];  extern const pcre_uint8     PRIV(default_tables)[];
 extern const int _pcre_utt_size;  
2616    
2617  extern const uschar _pcre_default_tables[];  extern const pcre_uint32    PRIV(hspace_list)[];
2618    extern const pcre_uint32    PRIV(vspace_list)[];
 extern const uschar _pcre_OP_lengths[];  
2619    
2620    
2621  /* Internal shared functions. These are functions that are used by more than  /* Internal shared functions. These are functions that are used by more than
2622  one of the exported public functions. They have to be "external" in the C  one of the exported public functions. They have to be "external" in the C
2623  sense, but are not part of the PCRE public API. */  sense, but are not part of the PCRE public API. */
2624    
2625  extern BOOL         _pcre_is_newline(const uschar *, int, const uschar *,  /* String comparison functions. */
2626                        int *, BOOL);  #if defined COMPILE_PCRE8
2627  extern int          _pcre_ord2utf8(int, uschar *);  
2628  extern real_pcre   *_pcre_try_flipped(const real_pcre *, real_pcre *,  #define STRCMP_UC_UC(str1, str2) \
2629                        const pcre_study_data *, pcre_study_data *);    strcmp((char *)(str1), (char *)(str2))
2630  extern int          _pcre_ucp_findprop(const unsigned int, int *, int *);  #define STRCMP_UC_C8(str1, str2) \
2631  extern unsigned int _pcre_ucp_othercase(const unsigned int);    strcmp((char *)(str1), (str2))
2632  extern int          _pcre_valid_utf8(const uschar *, int);  #define STRNCMP_UC_UC(str1, str2, num) \
2633  extern BOOL         _pcre_was_newline(const uschar *, int, const uschar *,    strncmp((char *)(str1), (char *)(str2), (num))
2634                        int *, BOOL);  #define STRNCMP_UC_C8(str1, str2, num) \
2635  extern BOOL         _pcre_xclass(int, const uschar *);    strncmp((char *)(str1), (str2), (num))
2636    #define STRLEN_UC(str) strlen((const char *)str)
2637    
2638    #elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
2639    
2640    extern int               PRIV(strcmp_uc_uc)(const pcre_uchar *,
2641                               const pcre_uchar *);
2642    extern int               PRIV(strcmp_uc_c8)(const pcre_uchar *,
2643                               const char *);
2644    extern int               PRIV(strncmp_uc_uc)(const pcre_uchar *,
2645                               const pcre_uchar *, unsigned int num);
2646    extern int               PRIV(strncmp_uc_c8)(const pcre_uchar *,
2647                               const char *, unsigned int num);
2648    extern unsigned int      PRIV(strlen_uc)(const pcre_uchar *str);
2649    
2650    #define STRCMP_UC_UC(str1, str2) \
2651      PRIV(strcmp_uc_uc)((str1), (str2))
2652    #define STRCMP_UC_C8(str1, str2) \
2653      PRIV(strcmp_uc_c8)((str1), (str2))
2654    #define STRNCMP_UC_UC(str1, str2, num) \
2655      PRIV(strncmp_uc_uc)((str1), (str2), (num))
2656    #define STRNCMP_UC_C8(str1, str2, num) \
2657      PRIV(strncmp_uc_c8)((str1), (str2), (num))
2658    #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2659    
2660    #endif /* COMPILE_PCRE[8|16|32] */
2661    
2662    #if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
2663    
2664    #define STRCMP_UC_UC_TEST(str1, str2) STRCMP_UC_UC(str1, str2)
2665    #define STRCMP_UC_C8_TEST(str1, str2) STRCMP_UC_C8(str1, str2)
2666    
2667    #elif defined COMPILE_PCRE32
2668    
2669    extern int               PRIV(strcmp_uc_uc_utf)(const pcre_uchar *,
2670                               const pcre_uchar *);
2671    extern int               PRIV(strcmp_uc_c8_utf)(const pcre_uchar *,
2672                               const char *);
2673    
2674    #define STRCMP_UC_UC_TEST(str1, str2) \
2675      (utf ? PRIV(strcmp_uc_uc_utf)((str1), (str2)) : PRIV(strcmp_uc_uc)((str1), (str2)))
2676    #define STRCMP_UC_C8_TEST(str1, str2) \
2677      (utf ? PRIV(strcmp_uc_c8_utf)((str1), (str2)) : PRIV(strcmp_uc_c8)((str1), (str2)))
2678    
2679    #endif /* COMPILE_PCRE[8|16|32] */
2680    
2681    extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2682    extern BOOL              PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2683                               int *, BOOL);
2684    extern int               PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
2685    extern int               PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
2686    extern BOOL              PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2687                               int *, BOOL);
2688    extern BOOL              PRIV(xclass)(pcre_uint32, const pcre_uchar *, BOOL);
2689    
2690    #ifdef SUPPORT_JIT
2691    extern void              PRIV(jit_compile)(const REAL_PCRE *,
2692                               PUBL(extra) *, int);
2693    extern int               PRIV(jit_exec)(const REAL_PCRE *, const PUBL(extra) *,
2694                               const pcre_uchar *, int, int, int, int *, int);
2695    extern void              PRIV(jit_free)(void *);
2696    extern int               PRIV(jit_get_size)(void *);
2697    extern const char*       PRIV(jit_get_target)(void);
2698    #endif
2699    
2700    /* Unicode character database (UCD) */
2701    
2702    typedef struct {
2703      pcre_uint8 script;     /* ucp_Arabic, etc. */
2704      pcre_uint8 chartype;   /* ucp_Cc, etc. (general categories) */
2705      pcre_uint8 gbprop;     /* ucp_gbControl, etc. (grapheme break property) */
2706      pcre_uint8 caseset;    /* offset to multichar other cases or zero */
2707      pcre_int32 other_case; /* offset to other case, or zero if none */
2708    } ucd_record;
2709    
2710    extern const pcre_uint32 PRIV(ucd_caseless_sets)[];
2711    extern const ucd_record  PRIV(ucd_records)[];
2712    extern const pcre_uint8  PRIV(ucd_stage1)[];
2713    extern const pcre_uint16 PRIV(ucd_stage2)[];
2714    extern const pcre_uint32 PRIV(ucp_gentype)[];
2715    extern const pcre_uint32 PRIV(ucp_gbtable)[];
2716    #ifdef SUPPORT_JIT
2717    extern const int         PRIV(ucp_typerange)[];
2718    #endif
2719    
2720    #ifdef SUPPORT_UCP
2721    /* UCD access macros */
2722    
2723    #define UCD_BLOCK_SIZE 128
2724    #define GET_UCD(ch) (PRIV(ucd_records) + \
2725            PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \
2726            UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE])
2727    
2728    #define UCD_CHARTYPE(ch)    GET_UCD(ch)->chartype
2729    #define UCD_SCRIPT(ch)      GET_UCD(ch)->script
2730    #define UCD_CATEGORY(ch)    PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2731    #define UCD_GRAPHBREAK(ch)  GET_UCD(ch)->gbprop
2732    #define UCD_CASESET(ch)     GET_UCD(ch)->caseset
2733    #define UCD_OTHERCASE(ch)   ((pcre_uint32)((int)ch + (int)(GET_UCD(ch)->other_case)))
2734    
2735    #endif /* SUPPORT_UCP */
2736    
2737  #endif  #endif
2738    

Legend:
Removed from v.286  
changed lines
  Added in v.1128

  ViewVC Help
Powered by ViewVC 1.1.5