/[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 192 by ph10, Mon Jul 30 11:23:21 2007 UTC revision 1197 by chpe, Thu Nov 1 19:23:27 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    
57    #if !defined COMPILE_PCRE16 && !defined COMPILE_PCRE32
58    #define COMPILE_PCRE8
59    #endif
60    
61    /* If SUPPORT_UCP is defined, SUPPORT_UTF must also be defined. The
62    "configure" script ensures this, but not everybody uses "configure". */
63    
64    #if defined SUPPORT_UCP && !(defined SUPPORT_UTF)
65    #define SUPPORT_UTF 1
66    #endif
67    
68    /* We define SUPPORT_UTF if SUPPORT_UTF8 is enabled for compatibility
69    reasons with existing code. */
70    
71    #if defined SUPPORT_UTF8 && !(defined SUPPORT_UTF)
72    #define SUPPORT_UTF 1
73    #endif
74    
75    /* Fixme: SUPPORT_UTF8 should be eventually disappear from the code.
76    Until then we define it if SUPPORT_UTF is defined. */
77    
78    #if defined SUPPORT_UTF && !(defined SUPPORT_UTF8)
79    #define SUPPORT_UTF8 1
80    #endif
81    
82    /* We do not support both EBCDIC and UTF-8/16/32 at the same time. The "configure"
83    script prevents both being selected, but not everybody uses "configure". */
84    
85    #if defined EBCDIC && defined SUPPORT_UTF
86    #error The use of both EBCDIC and SUPPORT_UTF is not supported.
87  #endif  #endif
88    
89  /* 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 95  It turns out that the Mac Debugging.h he
95  be absolutely sure we get our version. */  be absolutely sure we get our version. */
96    
97  #undef DPRINTF  #undef DPRINTF
98  #ifdef DEBUG  #ifdef PCRE_DEBUG
99  #define DPRINTF(p) printf p  #define DPRINTF(p) printf p
100  #else  #else
101  #define DPRINTF(p) /* Nothing */  #define DPRINTF(p) /* Nothing */
102  #endif  #endif
103    
104    
 /* Get the definitions provided by running "configure" */  
   
 #include "config.h"  
   
105  /* Standard C headers plus the external interface definition. The only time  /* Standard C headers plus the external interface definition. The only time
106  setjmp and stdarg are used is when NO_RECURSE is set. */  setjmp and stdarg are used is when NO_RECURSE is set. */
107    
108  #include <ctype.h>  #include <ctype.h>
109  #include <limits.h>  #include <limits.h>
 #include <setjmp.h>  
 #include <stdarg.h>  
110  #include <stddef.h>  #include <stddef.h>
111  #include <stdio.h>  #include <stdio.h>
112  #include <stdlib.h>  #include <stdlib.h>
113  #include <string.h>  #include <string.h>
114    
115    /* Valgrind (memcheck) support */
116    
117    #ifdef SUPPORT_VALGRIND
118    #include <valgrind/memcheck.h>
119    #endif
120    
121  /* When compiling a DLL for Windows, the exported symbols have to be declared  /* When compiling a DLL for Windows, the exported symbols have to be declared
122  using some MS magic. I found some useful information on this web page:  using some MS magic. I found some useful information on this web page:
123  http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the  http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
# Line 112  PCRE_EXP_DATA_DEFN only if they are not Line 147  PCRE_EXP_DATA_DEFN only if they are not
147    
148  #ifndef PCRE_EXP_DECL  #ifndef PCRE_EXP_DECL
149  #  ifdef _WIN32  #  ifdef _WIN32
150  #    ifdef DLL_EXPORT  #    ifndef PCRE_STATIC
151  #      define PCRE_EXP_DECL       extern __declspec(dllexport)  #      define PCRE_EXP_DECL       extern __declspec(dllexport)
152  #      define PCRE_EXP_DEFN       __declspec(dllexport)  #      define PCRE_EXP_DEFN       __declspec(dllexport)
153  #      define PCRE_EXP_DATA_DEFN  __declspec(dllexport)  #      define PCRE_EXP_DATA_DEFN  __declspec(dllexport)
# Line 121  PCRE_EXP_DATA_DEFN only if they are not Line 156  PCRE_EXP_DATA_DEFN only if they are not
156  #      define PCRE_EXP_DEFN  #      define PCRE_EXP_DEFN
157  #      define PCRE_EXP_DATA_DEFN  #      define PCRE_EXP_DATA_DEFN
158  #    endif  #    endif
 #  
159  #  else  #  else
160  #    ifdef __cplusplus  #    ifdef __cplusplus
161  #      define PCRE_EXP_DECL       extern "C"  #      define PCRE_EXP_DECL       extern "C"
# Line 137  PCRE_EXP_DATA_DEFN only if they are not Line 171  PCRE_EXP_DATA_DEFN only if they are not
171  #  endif  #  endif
172  #endif  #endif
173    
174  /* 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
175    a "calling convention" before exported function names. (This is secondhand
176    information; I know nothing about MSVC myself). For example, something like
177    
178      void __cdecl function(....)
179    
180    might be needed. In order so make this easy, all the exported functions have
181    PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
182    set, we ensure here that it has no effect. */
183    
184    #ifndef PCRE_CALL_CONVENTION
185    #define PCRE_CALL_CONVENTION
186    #endif
187    
188    /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
189  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
190  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
191  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
192  preprocessor time in standard C environments. */  preprocessor time in standard C environments. */
193    
194    typedef unsigned char pcre_uint8;
195    
196  #if USHRT_MAX == 65535  #if USHRT_MAX == 65535
197    typedef unsigned short pcre_uint16;    typedef unsigned short pcre_uint16;
198      typedef short pcre_int16;
199  #elif UINT_MAX == 65535  #elif UINT_MAX == 65535
200    typedef unsigned int pcre_uint16;    typedef unsigned int pcre_uint16;
201      typedef int pcre_int16;
202  #else  #else
203    #error Cannot determine a type for 16-bit unsigned integers    #error Cannot determine a type for 16-bit unsigned integers
204  #endif  #endif
205    
206  #if UINT_MAX == 4294967295  #if UINT_MAX == 4294967295
207    typedef unsigned int pcre_uint32;    typedef unsigned int pcre_uint32;
208      typedef int pcre_int32;
209  #elif ULONG_MAX == 4294967295  #elif ULONG_MAX == 4294967295
210    typedef unsigned long int pcre_uint32;    typedef unsigned long int pcre_uint32;
211      typedef long int pcre_int32;
212  #else  #else
213    #error Cannot determine a type for 32-bit unsigned integers    #error Cannot determine a type for 32-bit unsigned integers
214  #endif  #endif
215    
216    /* When checking for integer overflow in pcre_compile(), we need to handle
217    large integers. If a 64-bit integer type is available, we can use that.
218    Otherwise we have to cast to double, which of course requires floating point
219    arithmetic. Handle this by defining a macro for the appropriate type. If
220    stdint.h is available, include it; it may define INT64_MAX. Systems that do not
221    have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
222    by "configure". */
223    
224    #if defined HAVE_STDINT_H
225    #include <stdint.h>
226    #elif defined HAVE_INTTYPES_H
227    #include <inttypes.h>
228    #endif
229    
230    #if defined INT64_MAX || defined int64_t
231    #define INT64_OR_DOUBLE int64_t
232    #else
233    #define INT64_OR_DOUBLE double
234    #endif
235    
236  /* All character handling must be done as unsigned characters. Otherwise there  /* All character handling must be done as unsigned characters. Otherwise there
237  are problems with top-bit-set characters and functions such as isspace().  are problems with top-bit-set characters and functions such as isspace().
238  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 *,
239  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
240  to save lots of typing. I tried "uchar", but it causes problems on Digital  called pcre_uchar.
241  Unix, where it is defined in sys/types, so use "uschar" instead. */  
242    The IN_UCHARS macro multiply its argument with the byte size of the current
243    pcre_uchar type. Useful for memcpy and such operations, whose require the
244    byte size of their input/output buffers.
245    
246    The MAX_255 macro checks whether its pcre_uchar input is less than 256.
247    
248    The TABLE_GET macro is designed for accessing elements of tables whose contain
249    exactly 256 items. When the character is able to contain more than 256
250    items, some check is needed before accessing these tables.
251    */
252    
253    #if defined COMPILE_PCRE8
254    
255  typedef unsigned char uschar;  typedef unsigned char pcre_uchar;
256    #define IN_UCHARS(x) (x)
257    #define MAX_255(c) 1
258    #define TABLE_GET(c, table, default) ((table)[c])
259    
260    #elif defined COMPILE_PCRE16
261    
262    #if USHRT_MAX != 65535
263    /* This is a warning message. Change PCRE_UCHAR16 to a 16 bit data type in
264    pcre.h(.in) and disable (comment out) this message. */
265    #error Warning: PCRE_UCHAR16 is not a 16 bit data type.
266    #endif
267    
268    typedef pcre_uint16 pcre_uchar;
269    #define UCHAR_SHIFT (1)
270    #define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
271    #define MAX_255(c) ((c) <= 255u)
272    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
273    
274    #elif defined COMPILE_PCRE32
275    
276    typedef pcre_uint32 pcre_uchar;
277    #define UCHAR_SHIFT (2)
278    #define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
279    #define MAX_255(c) ((c) <= 255u)
280    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
281    
282    /* Assert that pcre_uchar32 is a 32-bit type */
283    typedef int __assert_pcre_uchar32_size[sizeof(pcre_uchar) == 4 ? 1 : -1];
284    
285    #else
286    #error Unsupported compiling mode
287    #endif /* COMPILE_PCRE[8|16|32] */
288    
289  /* 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
290  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 189  start/end of string field names are. */ Line 307  start/end of string field names are. */
307  #define IS_NEWLINE(p) \  #define IS_NEWLINE(p) \
308    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
309      ((p) < NLBLOCK->PSEND && \      ((p) < NLBLOCK->PSEND && \
310       _pcre_is_newline((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\       PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
311         utf8)) \         &(NLBLOCK->nllen), utf)) \
312      : \      : \
313      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
314       (p)[0] == NLBLOCK->nl[0] && \       RAWUCHARTEST(p) == NLBLOCK->nl[0] && \
315       (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \       (NLBLOCK->nllen == 1 || RAWUCHARTEST(p+1) == NLBLOCK->nl[1])       \
316      ) \      ) \
317    )    )
318    
# Line 203  start/end of string field names are. */ Line 321  start/end of string field names are. */
321  #define WAS_NEWLINE(p) \  #define WAS_NEWLINE(p) \
322    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
323      ((p) > NLBLOCK->PSSTART && \      ((p) > NLBLOCK->PSSTART && \
324       _pcre_was_newline((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \       PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
325         &(NLBLOCK->nllen), utf8)) \         &(NLBLOCK->nllen), utf)) \
326      : \      : \
327      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
328       (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \       RAWUCHARTEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] &&              \
329       (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \       (NLBLOCK->nllen == 1 || RAWUCHARTEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \
330      ) \      ) \
331    )    )
332    
# Line 222  used for the external interface and appe Line 340  used for the external interface and appe
340  must begin with PCRE_. */  must begin with PCRE_. */
341    
342  #ifdef CUSTOM_SUBJECT_PTR  #ifdef CUSTOM_SUBJECT_PTR
343  #define PCRE_SPTR CUSTOM_SUBJECT_PTR  #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
 #define USPTR CUSTOM_SUBJECT_PTR  
344  #else  #else
345  #define PCRE_SPTR const char *  #define PCRE_PUCHAR const pcre_uchar *
 #define USPTR const unsigned char *  
346  #endif  #endif
347    
   
   
348  /* Include the public PCRE header and the definitions of UCP character property  /* Include the public PCRE header and the definitions of UCP character property
349  values. */  values. */
350    
# Line 298  The macros are controlled by the value o Line 412  The macros are controlled by the value o
412  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
413  is automated on Unix systems via the "configure" command. */  is automated on Unix systems via the "configure" command. */
414    
415    #if defined COMPILE_PCRE8
416    
417  #if LINK_SIZE == 2  #if LINK_SIZE == 2
418    
419  #define PUT(a,n,d)   \  #define PUT(a,n,d)   \
# Line 334  is automated on Unix systems via the "co Line 450  is automated on Unix systems via the "co
450  #define GET(a,n) \  #define GET(a,n) \
451    (((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])
452    
453  #define MAX_PATTERN_SIZE (1 << 30)   /* Keep it positive */  /* Keep it positive */
454    #define MAX_PATTERN_SIZE (1 << 30)
455    
456    #else
457    #error LINK_SIZE must be either 2, 3, or 4
458    #endif
459    
460    #elif defined COMPILE_PCRE16
461    
462    #if LINK_SIZE == 2
463    
464    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
465    #undef LINK_SIZE
466    #define LINK_SIZE 1
467    
468    #define PUT(a,n,d)   \
469      (a[n] = (d))
470    
471    #define GET(a,n) \
472      (a[n])
473    
474    #define MAX_PATTERN_SIZE (1 << 16)
475    
476    #elif LINK_SIZE == 3 || LINK_SIZE == 4
477    
478    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
479    #undef LINK_SIZE
480    #define LINK_SIZE 2
481    
482    #define PUT(a,n,d)   \
483      (a[n] = (d) >> 16), \
484      (a[(n)+1] = (d) & 65535)
485    
486    #define GET(a,n) \
487      (((a)[n] << 16) | (a)[(n)+1])
488    
489    /* Keep it positive */
490    #define MAX_PATTERN_SIZE (1 << 30)
491    
492  #else  #else
493  #error LINK_SIZE must be either 2, 3, or 4  #error LINK_SIZE must be either 2, 3, or 4
494  #endif  #endif
495    
496    #elif defined COMPILE_PCRE32
497    
498    /* Only supported LINK_SIZE is 4 */
499    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
500    #undef LINK_SIZE
501    #define LINK_SIZE 1
502    
503    #define PUT(a,n,d)   \
504      (a[n] = (d))
505    
506    #define GET(a,n) \
507      (a[n])
508    
509    /* Keep it positive */
510    #define MAX_PATTERN_SIZE (1 << 30)
511    
512    #else
513    #error Unsupported compiling mode
514    #endif /* COMPILE_PCRE[8|16|32] */
515    
516  /* Convenience macro defined in terms of the others */  /* Convenience macro defined in terms of the others */
517    
# Line 351  is automated on Unix systems via the "co Line 522  is automated on Unix systems via the "co
522  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
523  capturing parenthesis numbers in back references. */  capturing parenthesis numbers in back references. */
524    
525    #if defined COMPILE_PCRE8
526    
527    #define IMM2_SIZE 2
528    
529  #define PUT2(a,n,d)   \  #define PUT2(a,n,d)   \
530    a[n] = (d) >> 8; \    a[n] = (d) >> 8; \
531    a[(n)+1] = (d) & 255    a[(n)+1] = (d) & 255
532    
533    /* For reasons that I do not understand, the expression in this GET2 macro is
534    treated by gcc as a signed expression, even when a is declared as unsigned. It
535    seems that any kind of arithmetic results in a signed value. */
536    
537  #define GET2(a,n) \  #define GET2(a,n) \
538    (((a)[n] << 8) | (a)[(n)+1])    (unsigned int)(((a)[n] << 8) | (a)[(n)+1])
539    
540  #define PUT2INC(a,n,d)  PUT2(a,n,d), a += 2  #elif defined COMPILE_PCRE16
541    
542    #define IMM2_SIZE 1
543    
544  /* When UTF-8 encoding is being used, a character is no longer just a single  #define PUT2(a,n,d)   \
545  byte. The macros for character handling generate simple sequences when used in     a[n] = d
546  byte-mode, and more complicated ones for UTF-8 characters. */  
547    #define GET2(a,n) \
548       a[n]
549    
550    #elif defined COMPILE_PCRE32
551    
552    #define IMM2_SIZE 1
553    
554  #ifndef SUPPORT_UTF8  #define PUT2(a,n,d)   \
555       a[n] = d
556    
557    #define GET2(a,n) \
558       a[n]
559    
560    #else
561    #error Unsupported compiling mode
562    #endif /* COMPILE_PCRE[8|16|32] */
563    
564    #define PUT2INC(a,n,d)  PUT2(a,n,d), a += IMM2_SIZE
565    
566    /* The maximum length of a MARK name is currently one data unit; it may be
567    changed in future to be a fixed number of bytes or to depend on LINK_SIZE. */
568    
569    #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
570    #define MAX_MARK ((1u << 16) - 1)
571    #else
572    #define MAX_MARK ((1u << 8) - 1)
573    #endif
574    
575    /* When UTF encoding is being used, a character is no longer just a single
576    byte. The macros for character handling generate simple sequences when used in
577    character-mode, and more complicated ones for UTF characters. GETCHARLENTEST
578    and other macros are not used when UTF is not supported, so they are not
579    defined. To make sure they can never even appear when UTF support is omitted,
580    we don't even define them. */
581    
582    #ifndef SUPPORT_UTF
583    
584    /* #define MAX_VALUE_FOR_SINGLE_CHAR */
585    /* #define HAS_EXTRALEN(c) */
586    /* #define GET_EXTRALEN(c) */
587    /* #define NOT_FIRSTCHAR(c) */
588  #define GETCHAR(c, eptr) c = *eptr;  #define GETCHAR(c, eptr) c = *eptr;
589  #define GETCHARTEST(c, eptr) c = *eptr;  #define GETCHARTEST(c, eptr) c = *eptr;
590  #define GETCHARINC(c, eptr) c = *eptr++;  #define GETCHARINC(c, eptr) c = *eptr++;
591  #define GETCHARINCTEST(c, eptr) c = *eptr++;  #define GETCHARINCTEST(c, eptr) c = *eptr++;
592  #define GETCHARLEN(c, eptr, len) c = *eptr;  #define GETCHARLEN(c, eptr, len) c = *eptr;
593  #define BACKCHAR(eptr)  #define RAWUCHAR(eptr) (*(eptr))
594    #define RAWUCHARINC(eptr) (*(eptr)++)
595    #define RAWUCHARTEST(eptr) (*(eptr))
596    #define RAWUCHARINCTEST(eptr) (*(eptr)++)
597    /* #define GETCHARLENTEST(c, eptr, len) */
598    /* #define BACKCHAR(eptr) */
599    /* #define FORWARDCHAR(eptr) */
600    /* #define ACROSSCHAR(condition, eptr, action) */
601    
602  #else   /* SUPPORT_UTF8 */  #else   /* SUPPORT_UTF */
603    
604  /* 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. */  
605    
606  #define GETCHAR(c, eptr) \  #define HASUTF8EXTRALEN(c) ((c) >= 0xc0)
607    c = *eptr; \  
608    if (c >= 0xc0) \  /* Base macro to pick up the remaining bytes of a UTF-8 character, not
609    advancing the pointer. */
610    
611    #define GETUTF8(c, eptr) \
612        { \
613        if ((c & 0x20) == 0) \
614          c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
615        else if ((c & 0x10) == 0) \
616          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
617        else if ((c & 0x08) == 0) \
618          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
619          ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
620        else if ((c & 0x04) == 0) \
621          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
622              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
623              (eptr[4] & 0x3f); \
624        else \
625          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
626              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
627              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
628        }
629    
630    /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
631    the pointer. */
632    
633    #define GETUTF8INC(c, eptr) \
634      { \      { \
635      int gcii; \      if ((c & 0x20) == 0) \
636      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \        c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
637      int gcss = 6*gcaa; \      else if ((c & 0x10) == 0) \
638      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        { \
639      for (gcii = 1; gcii <= gcaa; gcii++) \        c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
640          eptr += 2; \
641          } \
642        else if ((c & 0x08) == 0) \
643          { \
644          c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
645              ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
646          eptr += 3; \
647          } \
648        else if ((c & 0x04) == 0) \
649          { \
650          c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
651              ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
652              (eptr[3] & 0x3f); \
653          eptr += 4; \
654          } \
655        else \
656        { \        { \
657        gcss -= 6; \        c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
658        c |= (eptr[gcii] & 0x3f) << gcss; \            ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
659              ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
660          eptr += 5; \
661        } \        } \
662      }      }
663    
664    #if defined COMPILE_PCRE8
665    
666    /* These macros were originally written in the form of loops that used data
667    from the tables whose names start with PRIV(utf8_table). They were rewritten by
668    a user so as not to use loops, because in some environments this gives a
669    significant performance advantage, and it seems never to do any harm. */
670    
671    /* Tells the biggest code point which can be encoded as a single character. */
672    
673    #define MAX_VALUE_FOR_SINGLE_CHAR 127
674    
675    /* Tests whether the code point needs extra characters to decode. */
676    
677    #define HAS_EXTRALEN(c) ((c) >= 0xc0)
678    
679    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
680    Otherwise it has an undefined behaviour. */
681    
682    #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
683    
684    /* Returns TRUE, if the given character is not the first character
685    of a UTF sequence. */
686    
687    #define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80)
688    
689    /* Get the next UTF-8 character, not advancing the pointer. This is called when
690    we know we are in UTF-8 mode. */
691    
692    #define GETCHAR(c, eptr) \
693      c = *eptr; \
694      if (c >= 0xc0) GETUTF8(c, eptr);
695    
696  /* 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
697  pointer. */  pointer. */
698    
699  #define GETCHARTEST(c, eptr) \  #define GETCHARTEST(c, eptr) \
700    c = *eptr; \    c = *eptr; \
701    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; \  
       } \  
     }  
702    
703  /* 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
704  know we are in UTF-8 mode. */  know we are in UTF-8 mode. */
705    
706  #define GETCHARINC(c, eptr) \  #define GETCHARINC(c, eptr) \
707    c = *eptr++; \    c = *eptr++; \
708    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; \  
       } \  
     }  
709    
710  /* 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.
711    This is called when we don't know if we are in UTF-8 mode. */
712    
713  #define GETCHARINCTEST(c, eptr) \  #define GETCHARINCTEST(c, eptr) \
714    c = *eptr++; \    c = *eptr++; \
715    if (utf8 && c >= 0xc0) \    if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
716    
717    /* Base macro to pick up the remaining bytes of a UTF-8 character, not
718    advancing the pointer, incrementing the length. */
719    
720    #define GETUTF8LEN(c, eptr, len) \
721      { \      { \
722      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \      if ((c & 0x20) == 0) \
723      int gcss = 6*gcaa; \        { \
724      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
725      while (gcaa-- > 0) \        len++; \
726          } \
727        else if ((c & 0x10)  == 0) \
728        { \        { \
729        gcss -= 6; \        c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
730        c |= (*eptr++ & 0x3f) << gcss; \        len += 2; \
731          } \
732        else if ((c & 0x08)  == 0) \
733          {\
734          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
735              ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
736          len += 3; \
737          } \
738        else if ((c & 0x04)  == 0) \
739          { \
740          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
741              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
742              (eptr[4] & 0x3f); \
743          len += 4; \
744          } \
745        else \
746          {\
747          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
748              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
749              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
750          len += 5; \
751        } \        } \
752      }      }
753    
# Line 449  if there are extra bytes. This is called Line 756  if there are extra bytes. This is called
756    
757  #define GETCHARLEN(c, eptr, len) \  #define GETCHARLEN(c, eptr, len) \
758    c = *eptr; \    c = *eptr; \
759    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
760      { \  
761      int gcii; \  /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
762      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  pointer, incrementing length if there are extra bytes. This is called when we
763      int gcss = 6*gcaa; \  do not know if we are in UTF-8 mode. */
764      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
765      for (gcii = 1; gcii <= gcaa; gcii++) \  #define GETCHARLENTEST(c, eptr, len) \
766        { \    c = *eptr; \
767        gcss -= 6; \    if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
768        c |= (eptr[gcii] & 0x3f) << gcss; \  
769        } \  /* Returns the next uchar, not advancing the pointer. This is called when
770      len += gcaa; \  we know we are in UTF mode. */
771      }  
772    #define RAWUCHAR(eptr) \
773      (*(eptr))
774    
775    /* Returns the next uchar, advancing the pointer. This is called when
776    we know we are in UTF mode. */
777    
778    #define RAWUCHARINC(eptr) \
779      (*((eptr)++))
780    
781    /* Returns the next uchar, testing for UTF mode, and not advancing the
782    pointer. */
783    
784    #define RAWUCHARTEST(eptr) \
785      (*(eptr))
786    
787    /* Returns the next uchar, testing for UTF mode, advancing the
788    pointer. */
789    
790    #define RAWUCHARINCTEST(eptr) \
791      (*((eptr)++))
792    
793  /* 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
794  it is. Called only in UTF-8 mode. */  it is. This is called only in UTF-8 mode - we don't put a test within the macro
795    because almost all calls are already within a block of UTF-8 only code. */
796    
797  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--;  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
798    
799  #endif  /* Same as above, just in the other direction. */
800    #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
801    
802    /* Same as above, but it allows a fully customizable form. */
803    #define ACROSSCHAR(condition, eptr, action) \
804      while((condition) && ((eptr) & 0xc0) == 0x80) action
805    
806  /* In case there is no definition of offsetof() provided - though any proper  #elif defined COMPILE_PCRE16
 Standard C system should have one. */  
807    
808  #ifndef offsetof  /* Tells the biggest code point which can be encoded as a single character. */
809  #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))  
810  #endif  #define MAX_VALUE_FOR_SINGLE_CHAR 65535
811    
812    /* Tests whether the code point needs extra characters to decode. */
813    
814  /* These are the public options that can change during matching. */  #define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
815    
816  #define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)  /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
817    Otherwise it has an undefined behaviour. */
818    
819  /* Private options flags start at the most significant end of the four bytes.  #define GET_EXTRALEN(c) 1
 The public options defined in pcre.h start at the least significant end. Make  
 sure they don't overlap! The bits are getting a bit scarce now -- when we run  
 out, there is a dummy word in the structure that could be used for the private  
 bits. */  
820    
821  #define PCRE_NOPARTIAL     0x80000000  /* can't use partial with this regex */  /* Returns TRUE, if the given character is not the first character
822  #define PCRE_FIRSTSET      0x40000000  /* first_byte is set */  of a UTF sequence. */
 #define PCRE_REQCHSET      0x20000000  /* req_byte is set */  
 #define PCRE_STARTLINE     0x10000000  /* start after \n for multiline */  
 #define PCRE_JCHANGED      0x08000000  /* j option changes within regex */  
823    
824  /* Options for the "extra" block produced by pcre_study(). */  #define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00)
825    
826  #define PCRE_STUDY_MAPPED   0x01     /* a map of starting chars exists */  /* Base macro to pick up the low surrogate of a UTF-16 character, not
827    advancing the pointer. */
828    
829    #define GETUTF16(c, eptr) \
830       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
831    
832    /* Get the next UTF-16 character, not advancing the pointer. This is called when
833    we know we are in UTF-16 mode. */
834    
835    #define GETCHAR(c, eptr) \
836      c = *eptr; \
837      if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
838    
839    /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
840    pointer. */
841    
842    #define GETCHARTEST(c, eptr) \
843      c = *eptr; \
844      if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
845    
846    /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
847    the pointer. */
848    
849    #define GETUTF16INC(c, eptr) \
850       { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
851    
852    /* Get the next UTF-16 character, advancing the pointer. This is called when we
853    know we are in UTF-16 mode. */
854    
855    #define GETCHARINC(c, eptr) \
856      c = *eptr++; \
857      if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
858    
859    /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
860    This is called when we don't know if we are in UTF-16 mode. */
861    
862    #define GETCHARINCTEST(c, eptr) \
863      c = *eptr++; \
864      if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
865    
866    /* Base macro to pick up the low surrogate of a UTF-16 character, not
867    advancing the pointer, incrementing the length. */
868    
869    #define GETUTF16LEN(c, eptr, len) \
870       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
871    
872    /* Get the next UTF-16 character, not advancing the pointer, incrementing
873    length if there is a low surrogate. This is called when we know we are in
874    UTF-16 mode. */
875    
876    #define GETCHARLEN(c, eptr, len) \
877      c = *eptr; \
878      if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
879    
880    /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
881    pointer, incrementing length if there is a low surrogate. This is called when
882    we do not know if we are in UTF-16 mode. */
883    
884    #define GETCHARLENTEST(c, eptr, len) \
885      c = *eptr; \
886      if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
887    
888    /* Returns the next uchar, not advancing the pointer. This is called when
889    we know we are in UTF mode. */
890    
891    #define RAWUCHAR(eptr) \
892      (*(eptr))
893    
894    /* Returns the next uchar, advancing the pointer. This is called when
895    we know we are in UTF mode. */
896    
897    #define RAWUCHARINC(eptr) \
898      (*((eptr)++))
899    
900    /* Returns the next uchar, testing for UTF mode, and not advancing the
901    pointer. */
902    
903    #define RAWUCHARTEST(eptr) \
904      (*(eptr))
905    
906    /* Returns the next uchar, testing for UTF mode, advancing the
907    pointer. */
908    
909    #define RAWUCHARINCTEST(eptr) \
910      (*((eptr)++))
911    
912    /* If the pointer is not at the start of a character, move it back until
913    it is. This is called only in UTF-16 mode - we don't put a test within the
914    macro because almost all calls are already within a block of UTF-16 only
915    code. */
916    
917    #define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
918    
919    /* Same as above, just in the other direction. */
920    #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
921    
922    /* Same as above, but it allows a fully customizable form. */
923    #define ACROSSCHAR(condition, eptr, action) \
924      if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
925    
926    #elif defined COMPILE_PCRE32
927    
928    /* These are trivial for the 32-bit library, since all UTF-32 characters fit
929    into one pcre_uchar unit. */
930    #define MAX_VALUE_FOR_SINGLE_CHAR (0x10ffffu)
931    #define HAS_EXTRALEN(c) (0)
932    #define GET_EXTRALEN(c) (0)
933    #define NOT_FIRSTCHAR(c) (0)
934    
935    #define UTF32_MASK (0x1fffffu)
936    
937    /* Get the next UTF-32 character, not advancing the pointer. This is called when
938    we know we are in UTF-32 mode. */
939    
940    #define GETCHAR(c, eptr) \
941      c = (*eptr) & UTF32_MASK;
942    
943    /* Get the next UTF-32 character, testing for UTF-32 mode, and not advancing the
944    pointer. */
945    
946    #define GETCHARTEST(c, eptr) \
947      c = *eptr; \
948      if (utf) c &= UTF32_MASK;
949    
950    /* Get the next UTF-32 character, advancing the pointer. This is called when we
951    know we are in UTF-32 mode. */
952    
953    #define GETCHARINC(c, eptr) \
954      c = (*eptr++) & UTF32_MASK;
955    
956    /* Get the next character, testing for UTF-32 mode, and advancing the pointer.
957    This is called when we don't know if we are in UTF-32 mode. */
958    
959    #define GETCHARINCTEST(c, eptr) \
960      c = *eptr++; \
961      if (utf) c &= UTF32_MASK;
962    
963    /* Get the next UTF-32 character, not advancing the pointer, not incrementing
964    length (since all UTF-32 is of length 1). This is called when we know we are in
965    UTF-32 mode. */
966    
967    #define GETCHARLEN(c, eptr, len) \
968      GETCHAR(c, eptr)
969    
970    /* Get the next UTF-32character, testing for UTF-32 mode, not advancing the
971    pointer, not incrementing the length (since all UTF-32 is of length 1).
972    This is called when we do not know if we are in UTF-32 mode. */
973    
974    #define GETCHARLENTEST(c, eptr, len) \
975      GETCHARTEST(c, eptr)
976    
977    /* Returns the next uchar, not advancing the pointer. This is called when
978    we know we are in UTF mode. */
979    
980    #define RAWUCHAR(eptr) \
981      (*(eptr) & UTF32_MASK)
982    
983    /* Returns the next uchar, advancing the pointer. This is called when
984    we know we are in UTF mode. */
985    
986    #define RAWUCHARINC(eptr) \
987      (*((eptr)++) & UTF32_MASK)
988    
989    /* Returns the next uchar, testing for UTF mode, and not advancing the
990    pointer. */
991    
992    #define RAWUCHARTEST(eptr) \
993      (utf ? (*(eptr) & UTF32_MASK) : *(eptr))
994    
995    /* Returns the next uchar, testing for UTF mode, advancing the
996    pointer. */
997    
998    #define RAWUCHARINCTEST(eptr) \
999      (utf ? (*((eptr)++) & UTF32_MASK) : *((eptr)++))
1000    
1001    /* If the pointer is not at the start of a character, move it back until
1002    it is. This is called only in UTF-32 mode - we don't put a test within the
1003    macro because almost all calls are already within a block of UTF-32 only
1004    code.
1005    These are all no-ops since all UTF-32 characters fit into one pcre_uchar. */
1006    
1007    #define BACKCHAR(eptr) do { } while (0)
1008    
1009    /* Same as above, just in the other direction. */
1010    #define FORWARDCHAR(eptr) do { } while (0)
1011    
1012    /* Same as above, but it allows a fully customizable form. */
1013    #define ACROSSCHAR(condition, eptr, action) do { } while (0)
1014    
1015    #else
1016    #error Unsupported compiling mode
1017    #endif /* COMPILE_PCRE[8|16|32] */
1018    
1019    #endif  /* SUPPORT_UTF */
1020    
1021    /* Tests for Unicode horizontal and vertical whitespace characters must check a
1022    number of different values. Using a switch statement for this generates the
1023    fastest code (no loop, no memory access), and there are several places in the
1024    interpreter code where this happens. In order to ensure that all the case lists
1025    remain in step, we use macros so that there is only one place where the lists
1026    are defined.
1027    
1028    These values are also required as lists in pcre_compile.c when processing \h,
1029    \H, \v and \V in a character class. The lists are defined in pcre_tables.c, but
1030    macros that define the values are here so that all the definitions are
1031    together. The lists must be in ascending character order, terminated by
1032    NOTACHAR (which is 0xffffffff).
1033    
1034    Any changes should ensure that the various macros are kept in step with each
1035    other. NOTE: The values also appear in pcre_jit_compile.c. */
1036    
1037    /* ------ ASCII/Unicode environments ------ */
1038    
1039    #ifndef EBCDIC
1040    
1041    #define HSPACE_LIST \
1042      CHAR_HT, CHAR_SPACE, 0xa0, \
1043      0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \
1044      0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \
1045      NOTACHAR
1046    
1047    #define HSPACE_MULTIBYTE_CASES \
1048      case 0x1680:  /* OGHAM SPACE MARK */ \
1049      case 0x180e:  /* MONGOLIAN VOWEL SEPARATOR */ \
1050      case 0x2000:  /* EN QUAD */ \
1051      case 0x2001:  /* EM QUAD */ \
1052      case 0x2002:  /* EN SPACE */ \
1053      case 0x2003:  /* EM SPACE */ \
1054      case 0x2004:  /* THREE-PER-EM SPACE */ \
1055      case 0x2005:  /* FOUR-PER-EM SPACE */ \
1056      case 0x2006:  /* SIX-PER-EM SPACE */ \
1057      case 0x2007:  /* FIGURE SPACE */ \
1058      case 0x2008:  /* PUNCTUATION SPACE */ \
1059      case 0x2009:  /* THIN SPACE */ \
1060      case 0x200A:  /* HAIR SPACE */ \
1061      case 0x202f:  /* NARROW NO-BREAK SPACE */ \
1062      case 0x205f:  /* MEDIUM MATHEMATICAL SPACE */ \
1063      case 0x3000   /* IDEOGRAPHIC SPACE */
1064    
1065    #define HSPACE_BYTE_CASES \
1066      case CHAR_HT: \
1067      case CHAR_SPACE: \
1068      case 0xa0     /* NBSP */
1069    
1070    #define HSPACE_CASES \
1071      HSPACE_BYTE_CASES: \
1072      HSPACE_MULTIBYTE_CASES
1073    
1074    #define VSPACE_LIST \
1075      CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR
1076    
1077    #define VSPACE_MULTIBYTE_CASES \
1078      case 0x2028:    /* LINE SEPARATOR */ \
1079      case 0x2029     /* PARAGRAPH SEPARATOR */
1080    
1081    #define VSPACE_BYTE_CASES \
1082      case CHAR_LF: \
1083      case CHAR_VT: \
1084      case CHAR_FF: \
1085      case CHAR_CR: \
1086      case CHAR_NEL
1087    
1088    #define VSPACE_CASES \
1089      VSPACE_BYTE_CASES: \
1090      VSPACE_MULTIBYTE_CASES
1091    
1092    /* ------ EBCDIC environments ------ */
1093    
1094    #else
1095    #define HSPACE_LIST CHAR_HT, CHAR_SPACE
1096    
1097    #define HSPACE_BYTE_CASES \
1098      case CHAR_HT: \
1099      case CHAR_SPACE
1100    
1101    #define HSPACE_CASES HSPACE_BYTE_CASES
1102    
1103    #ifdef EBCDIC_NL25
1104    #define VSPACE_LIST \
1105      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR
1106    #else
1107    #define VSPACE_LIST \
1108      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR
1109    #endif
1110    
1111    #define VSPACE_BYTE_CASES \
1112      case CHAR_LF: \
1113      case CHAR_VT: \
1114      case CHAR_FF: \
1115      case CHAR_CR: \
1116      case CHAR_NEL
1117    
1118    #define VSPACE_CASES VSPACE_BYTE_CASES
1119    #endif  /* EBCDIC */
1120    
1121    /* ------ End of whitespace macros ------ */
1122    
1123    
1124    
1125    /* Private flags containing information about the compiled regex. They used to
1126    live at the top end of the options word, but that got almost full, so now they
1127    are in a 16-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as
1128    the restrictions on partial matching have been lifted. It remains for backwards
1129    compatibility. */
1130    
1131    #define PCRE_MODE8         0x0001  /* compiled in 8 bit mode */
1132    #define PCRE_MODE16        0x0002  /* compiled in 16 bit mode */
1133    #define PCRE_MODE32        0x0004  /* compiled in 32 bit mode */
1134    #define PCRE_FIRSTSET      0x0010  /* first_char is set */
1135    #define PCRE_FCH_CASELESS  0x0020  /* caseless first char */
1136    #define PCRE_REQCHSET      0x0040  /* req_byte is set */
1137    #define PCRE_RCH_CASELESS  0x0080  /* caseless requested char */
1138    #define PCRE_STARTLINE     0x0100  /* start after \n for multiline */
1139    #define PCRE_NOPARTIAL     0x0200  /* can't use partial with this regex */
1140    #define PCRE_JCHANGED      0x0400  /* j option used in regex */
1141    #define PCRE_HASCRORLF     0x0800  /* explicit \r or \n in pattern */
1142    #define PCRE_HASTHEN       0x1000  /* pattern contains (*THEN) */
1143    
1144    #if defined COMPILE_PCRE8
1145    #define PCRE_MODE          PCRE_MODE8
1146    #elif defined COMPILE_PCRE16
1147    #define PCRE_MODE          PCRE_MODE16
1148    #elif defined COMPILE_PCRE32
1149    #define PCRE_MODE          PCRE_MODE32
1150    #endif
1151    #define PCRE_MODE_MASK     (PCRE_MODE8 | PCRE_MODE16 | PCRE_MODE32)
1152    
1153    /* Flags for the "extra" block produced by pcre_study(). */
1154    
1155    #define PCRE_STUDY_MAPPED  0x0001  /* a map of starting chars exists */
1156    #define PCRE_STUDY_MINLEN  0x0002  /* a minimum length field exists */
1157    
1158  /* Masks for identifying the public options that are permitted at compile  /* Masks for identifying the public options that are permitted at compile
1159  time, run time, or study time, respectively. */  time, run time, or study time, respectively. */
# Line 505  time, run time, or study time, respectiv Line 1161  time, run time, or study time, respectiv
1161  #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| \
1162                             PCRE_NEWLINE_ANYCRLF)                             PCRE_NEWLINE_ANYCRLF)
1163    
1164  #define PUBLIC_OPTIONS \  #define PUBLIC_COMPILE_OPTIONS \
1165    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
1166     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
1167     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| \
1168     PCRE_DUPNAMES|PCRE_NEWLINE_BITS)     PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1169       PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE)
1170    
1171  #define PUBLIC_EXEC_OPTIONS \  #define PUBLIC_EXEC_OPTIONS \
1172    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1173     PCRE_PARTIAL|PCRE_NEWLINE_BITS)     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
1174       PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
1175    
1176  #define PUBLIC_DFA_EXEC_OPTIONS \  #define PUBLIC_DFA_EXEC_OPTIONS \
1177    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1178     PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_BITS)     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
1179       PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1180  #define PUBLIC_STUDY_OPTIONS 0   /* None defined */     PCRE_NO_START_OPTIMIZE)
1181    
1182    #define PUBLIC_STUDY_OPTIONS \
1183       (PCRE_STUDY_JIT_COMPILE|PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE| \
1184        PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE|PCRE_STUDY_EXTRA_NEEDED)
1185    
1186    #define PUBLIC_JIT_EXEC_OPTIONS \
1187       (PCRE_NO_UTF8_CHECK|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|\
1188        PCRE_NOTEMPTY_ATSTART|PCRE_PARTIAL_SOFT|PCRE_PARTIAL_HARD)
1189    
1190  /* Magic number to provide a small check against being handed junk. Also used  /* Magic number to provide a small check against being handed junk. */
 to detect whether a pattern was compiled on a host of different endianness. */  
1191    
1192  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */
1193    
1194  /* Negative values for the firstchar and reqchar variables */  /* This variable is used to detect a loaded regular expression
1195    in different endianness. */
1196    
1197  #define REQ_UNSET (-2)  #define REVERSED_MAGIC_NUMBER  0x45524350UL   /* 'ERCP' */
 #define REQ_NONE  (-1)  
1198    
1199  /* 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
1200  req_byte match. */  req_byte match. */
1201    
1202  #define REQ_BYTE_MAX 1000  #define REQ_BYTE_MAX 1000
1203    
1204  /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a  /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
1205  variable-length repeat, or a anything other than literal characters. */  environments where these macros are defined elsewhere. Unfortunately, there
1206    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 */  
1207    
1208  typedef int BOOL;  typedef int BOOL;
1209    
1210    #ifndef FALSE
1211  #define FALSE   0  #define FALSE   0
1212  #define TRUE    1  #define TRUE    1
1213    #endif
1214    
1215    /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
1216    character constants like '*' because the compiler would emit their EBCDIC code,
1217    which is different from their ASCII/UTF-8 code. Instead we define macros for
1218    the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
1219    is enabled. When UTF-8 support is not enabled, the definitions use character
1220    literals. Both character and string versions of each character are needed, and
1221    there are some longer strings as well.
1222    
1223    This means that, on EBCDIC platforms, the PCRE library can handle either
1224    EBCDIC, or UTF-8, but not both. To support both in the same compiled library
1225    would need different lookups depending on whether PCRE_UTF8 was set or not.
1226    This would make it impossible to use characters in switch/case statements,
1227    which would reduce performance. For a theoretical use (which nobody has asked
1228    for) in a minority area (EBCDIC platforms), this is not sensible. Any
1229    application that did need both could compile two versions of the library, using
1230    macros to give the functions distinct names. */
1231    
1232    #ifndef SUPPORT_UTF
1233    
1234    /* UTF-8 support is not enabled; use the platform-dependent character literals
1235    so that PCRE works in both ASCII and EBCDIC environments, but only in non-UTF
1236    mode. Newline characters are problematic in EBCDIC. Though it has CR and LF
1237    characters, a common practice has been to use its NL (0x15) character as the
1238    line terminator in C-like processing environments. However, sometimes the LF
1239    (0x25) character is used instead, according to this Unicode document:
1240    
1241    http://unicode.org/standard/reports/tr13/tr13-5.html
1242    
1243    PCRE defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25
1244    instead. Whichever is *not* chosen is defined as NEL.
1245    
1246    In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the
1247    same code point. */
1248    
1249    #ifdef EBCDIC
1250    
1251    #ifndef EBCDIC_NL25
1252    #define CHAR_NL                     '\x15'
1253    #define CHAR_NEL                    '\x25'
1254    #define STR_NL                      "\x15"
1255    #define STR_NEL                     "\x25"
1256    #else
1257    #define CHAR_NL                     '\x25'
1258    #define CHAR_NEL                    '\x15'
1259    #define STR_NL                      "\x25"
1260    #define STR_NEL                     "\x15"
1261    #endif
1262    
1263    #define CHAR_LF                     CHAR_NL
1264    #define STR_LF                      STR_NL
1265    
1266    #define CHAR_ESC                    '\047'
1267    #define CHAR_DEL                    '\007'
1268    #define STR_ESC                     "\047"
1269    #define STR_DEL                     "\007"
1270    
1271    #else  /* Not EBCDIC */
1272    
1273    /* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for
1274    compatibility. NEL is the Unicode newline character; make sure it is
1275    a positive value. */
1276    
1277    #define CHAR_LF                     '\n'
1278    #define CHAR_NL                     CHAR_LF
1279    #define CHAR_NEL                    ((unsigned char)'\x85')
1280    #define CHAR_ESC                    '\033'
1281    #define CHAR_DEL                    '\177'
1282    
1283    #define STR_LF                      "\n"
1284    #define STR_NL                      STR_LF
1285    #define STR_NEL                     "\x85"
1286    #define STR_ESC                     "\033"
1287    #define STR_DEL                     "\177"
1288    
1289    #endif  /* EBCDIC */
1290    
1291    /* The remaining definitions work in both environments. */
1292    
1293    #define CHAR_NULL                   '\0'
1294    #define CHAR_HT                     '\t'
1295    #define CHAR_VT                     '\v'
1296    #define CHAR_FF                     '\f'
1297    #define CHAR_CR                     '\r'
1298    #define CHAR_BS                     '\b'
1299    #define CHAR_BEL                    '\a'
1300    
1301    #define CHAR_SPACE                  ' '
1302    #define CHAR_EXCLAMATION_MARK       '!'
1303    #define CHAR_QUOTATION_MARK         '"'
1304    #define CHAR_NUMBER_SIGN            '#'
1305    #define CHAR_DOLLAR_SIGN            '$'
1306    #define CHAR_PERCENT_SIGN           '%'
1307    #define CHAR_AMPERSAND              '&'
1308    #define CHAR_APOSTROPHE             '\''
1309    #define CHAR_LEFT_PARENTHESIS       '('
1310    #define CHAR_RIGHT_PARENTHESIS      ')'
1311    #define CHAR_ASTERISK               '*'
1312    #define CHAR_PLUS                   '+'
1313    #define CHAR_COMMA                  ','
1314    #define CHAR_MINUS                  '-'
1315    #define CHAR_DOT                    '.'
1316    #define CHAR_SLASH                  '/'
1317    #define CHAR_0                      '0'
1318    #define CHAR_1                      '1'
1319    #define CHAR_2                      '2'
1320    #define CHAR_3                      '3'
1321    #define CHAR_4                      '4'
1322    #define CHAR_5                      '5'
1323    #define CHAR_6                      '6'
1324    #define CHAR_7                      '7'
1325    #define CHAR_8                      '8'
1326    #define CHAR_9                      '9'
1327    #define CHAR_COLON                  ':'
1328    #define CHAR_SEMICOLON              ';'
1329    #define CHAR_LESS_THAN_SIGN         '<'
1330    #define CHAR_EQUALS_SIGN            '='
1331    #define CHAR_GREATER_THAN_SIGN      '>'
1332    #define CHAR_QUESTION_MARK          '?'
1333    #define CHAR_COMMERCIAL_AT          '@'
1334    #define CHAR_A                      'A'
1335    #define CHAR_B                      'B'
1336    #define CHAR_C                      'C'
1337    #define CHAR_D                      'D'
1338    #define CHAR_E                      'E'
1339    #define CHAR_F                      'F'
1340    #define CHAR_G                      'G'
1341    #define CHAR_H                      'H'
1342    #define CHAR_I                      'I'
1343    #define CHAR_J                      'J'
1344    #define CHAR_K                      'K'
1345    #define CHAR_L                      'L'
1346    #define CHAR_M                      'M'
1347    #define CHAR_N                      'N'
1348    #define CHAR_O                      'O'
1349    #define CHAR_P                      'P'
1350    #define CHAR_Q                      'Q'
1351    #define CHAR_R                      'R'
1352    #define CHAR_S                      'S'
1353    #define CHAR_T                      'T'
1354    #define CHAR_U                      'U'
1355    #define CHAR_V                      'V'
1356    #define CHAR_W                      'W'
1357    #define CHAR_X                      'X'
1358    #define CHAR_Y                      'Y'
1359    #define CHAR_Z                      'Z'
1360    #define CHAR_LEFT_SQUARE_BRACKET    '['
1361    #define CHAR_BACKSLASH              '\\'
1362    #define CHAR_RIGHT_SQUARE_BRACKET   ']'
1363    #define CHAR_CIRCUMFLEX_ACCENT      '^'
1364    #define CHAR_UNDERSCORE             '_'
1365    #define CHAR_GRAVE_ACCENT           '`'
1366    #define CHAR_a                      'a'
1367    #define CHAR_b                      'b'
1368    #define CHAR_c                      'c'
1369    #define CHAR_d                      'd'
1370    #define CHAR_e                      'e'
1371    #define CHAR_f                      'f'
1372    #define CHAR_g                      'g'
1373    #define CHAR_h                      'h'
1374    #define CHAR_i                      'i'
1375    #define CHAR_j                      'j'
1376    #define CHAR_k                      'k'
1377    #define CHAR_l                      'l'
1378    #define CHAR_m                      'm'
1379    #define CHAR_n                      'n'
1380    #define CHAR_o                      'o'
1381    #define CHAR_p                      'p'
1382    #define CHAR_q                      'q'
1383    #define CHAR_r                      'r'
1384    #define CHAR_s                      's'
1385    #define CHAR_t                      't'
1386    #define CHAR_u                      'u'
1387    #define CHAR_v                      'v'
1388    #define CHAR_w                      'w'
1389    #define CHAR_x                      'x'
1390    #define CHAR_y                      'y'
1391    #define CHAR_z                      'z'
1392    #define CHAR_LEFT_CURLY_BRACKET     '{'
1393    #define CHAR_VERTICAL_LINE          '|'
1394    #define CHAR_RIGHT_CURLY_BRACKET    '}'
1395    #define CHAR_TILDE                  '~'
1396    
1397    #define STR_HT                      "\t"
1398    #define STR_VT                      "\v"
1399    #define STR_FF                      "\f"
1400    #define STR_CR                      "\r"
1401    #define STR_BS                      "\b"
1402    #define STR_BEL                     "\a"
1403    
1404    #define STR_SPACE                   " "
1405    #define STR_EXCLAMATION_MARK        "!"
1406    #define STR_QUOTATION_MARK          "\""
1407    #define STR_NUMBER_SIGN             "#"
1408    #define STR_DOLLAR_SIGN             "$"
1409    #define STR_PERCENT_SIGN            "%"
1410    #define STR_AMPERSAND               "&"
1411    #define STR_APOSTROPHE              "'"
1412    #define STR_LEFT_PARENTHESIS        "("
1413    #define STR_RIGHT_PARENTHESIS       ")"
1414    #define STR_ASTERISK                "*"
1415    #define STR_PLUS                    "+"
1416    #define STR_COMMA                   ","
1417    #define STR_MINUS                   "-"
1418    #define STR_DOT                     "."
1419    #define STR_SLASH                   "/"
1420    #define STR_0                       "0"
1421    #define STR_1                       "1"
1422    #define STR_2                       "2"
1423    #define STR_3                       "3"
1424    #define STR_4                       "4"
1425    #define STR_5                       "5"
1426    #define STR_6                       "6"
1427    #define STR_7                       "7"
1428    #define STR_8                       "8"
1429    #define STR_9                       "9"
1430    #define STR_COLON                   ":"
1431    #define STR_SEMICOLON               ";"
1432    #define STR_LESS_THAN_SIGN          "<"
1433    #define STR_EQUALS_SIGN             "="
1434    #define STR_GREATER_THAN_SIGN       ">"
1435    #define STR_QUESTION_MARK           "?"
1436    #define STR_COMMERCIAL_AT           "@"
1437    #define STR_A                       "A"
1438    #define STR_B                       "B"
1439    #define STR_C                       "C"
1440    #define STR_D                       "D"
1441    #define STR_E                       "E"
1442    #define STR_F                       "F"
1443    #define STR_G                       "G"
1444    #define STR_H                       "H"
1445    #define STR_I                       "I"
1446    #define STR_J                       "J"
1447    #define STR_K                       "K"
1448    #define STR_L                       "L"
1449    #define STR_M                       "M"
1450    #define STR_N                       "N"
1451    #define STR_O                       "O"
1452    #define STR_P                       "P"
1453    #define STR_Q                       "Q"
1454    #define STR_R                       "R"
1455    #define STR_S                       "S"
1456    #define STR_T                       "T"
1457    #define STR_U                       "U"
1458    #define STR_V                       "V"
1459    #define STR_W                       "W"
1460    #define STR_X                       "X"
1461    #define STR_Y                       "Y"
1462    #define STR_Z                       "Z"
1463    #define STR_LEFT_SQUARE_BRACKET     "["
1464    #define STR_BACKSLASH               "\\"
1465    #define STR_RIGHT_SQUARE_BRACKET    "]"
1466    #define STR_CIRCUMFLEX_ACCENT       "^"
1467    #define STR_UNDERSCORE              "_"
1468    #define STR_GRAVE_ACCENT            "`"
1469    #define STR_a                       "a"
1470    #define STR_b                       "b"
1471    #define STR_c                       "c"
1472    #define STR_d                       "d"
1473    #define STR_e                       "e"
1474    #define STR_f                       "f"
1475    #define STR_g                       "g"
1476    #define STR_h                       "h"
1477    #define STR_i                       "i"
1478    #define STR_j                       "j"
1479    #define STR_k                       "k"
1480    #define STR_l                       "l"
1481    #define STR_m                       "m"
1482    #define STR_n                       "n"
1483    #define STR_o                       "o"
1484    #define STR_p                       "p"
1485    #define STR_q                       "q"
1486    #define STR_r                       "r"
1487    #define STR_s                       "s"
1488    #define STR_t                       "t"
1489    #define STR_u                       "u"
1490    #define STR_v                       "v"
1491    #define STR_w                       "w"
1492    #define STR_x                       "x"
1493    #define STR_y                       "y"
1494    #define STR_z                       "z"
1495    #define STR_LEFT_CURLY_BRACKET      "{"
1496    #define STR_VERTICAL_LINE           "|"
1497    #define STR_RIGHT_CURLY_BRACKET     "}"
1498    #define STR_TILDE                   "~"
1499    
1500    #define STRING_ACCEPT0              "ACCEPT\0"
1501    #define STRING_COMMIT0              "COMMIT\0"
1502    #define STRING_F0                   "F\0"
1503    #define STRING_FAIL0                "FAIL\0"
1504    #define STRING_MARK0                "MARK\0"
1505    #define STRING_PRUNE0               "PRUNE\0"
1506    #define STRING_SKIP0                "SKIP\0"
1507    #define STRING_THEN                 "THEN"
1508    
1509    #define STRING_alpha0               "alpha\0"
1510    #define STRING_lower0               "lower\0"
1511    #define STRING_upper0               "upper\0"
1512    #define STRING_alnum0               "alnum\0"
1513    #define STRING_ascii0               "ascii\0"
1514    #define STRING_blank0               "blank\0"
1515    #define STRING_cntrl0               "cntrl\0"
1516    #define STRING_digit0               "digit\0"
1517    #define STRING_graph0               "graph\0"
1518    #define STRING_print0               "print\0"
1519    #define STRING_punct0               "punct\0"
1520    #define STRING_space0               "space\0"
1521    #define STRING_word0                "word\0"
1522    #define STRING_xdigit               "xdigit"
1523    
1524    #define STRING_DEFINE               "DEFINE"
1525    
1526    #define STRING_CR_RIGHTPAR             "CR)"
1527    #define STRING_LF_RIGHTPAR             "LF)"
1528    #define STRING_CRLF_RIGHTPAR           "CRLF)"
1529    #define STRING_ANY_RIGHTPAR            "ANY)"
1530    #define STRING_ANYCRLF_RIGHTPAR        "ANYCRLF)"
1531    #define STRING_BSR_ANYCRLF_RIGHTPAR    "BSR_ANYCRLF)"
1532    #define STRING_BSR_UNICODE_RIGHTPAR    "BSR_UNICODE)"
1533    #ifdef COMPILE_PCRE8
1534    #define STRING_UTF_RIGHTPAR            "UTF8)"
1535    #endif
1536    #ifdef COMPILE_PCRE16
1537    #define STRING_UTF_RIGHTPAR            "UTF16)"
1538    #endif
1539    #ifdef COMPILE_PCRE32
1540    #define STRING_UTF_RIGHTPAR            "UTF32)"
1541    #endif
1542    #define STRING_UCP_RIGHTPAR            "UCP)"
1543    #define STRING_NO_START_OPT_RIGHTPAR   "NO_START_OPT)"
1544    
1545    #else  /* SUPPORT_UTF */
1546    
1547    /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1548    works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1549    only. */
1550    
1551    #define CHAR_HT                     '\011'
1552    #define CHAR_VT                     '\013'
1553    #define CHAR_FF                     '\014'
1554    #define CHAR_CR                     '\015'
1555    #define CHAR_LF                     '\012'
1556    #define CHAR_NL                     CHAR_LF
1557    #define CHAR_NEL                    ((unsigned char)'\x85')
1558    #define CHAR_BS                     '\010'
1559    #define CHAR_BEL                    '\007'
1560    #define CHAR_ESC                    '\033'
1561    #define CHAR_DEL                    '\177'
1562    
1563    #define CHAR_NULL                   '\0'
1564    #define CHAR_SPACE                  '\040'
1565    #define CHAR_EXCLAMATION_MARK       '\041'
1566    #define CHAR_QUOTATION_MARK         '\042'
1567    #define CHAR_NUMBER_SIGN            '\043'
1568    #define CHAR_DOLLAR_SIGN            '\044'
1569    #define CHAR_PERCENT_SIGN           '\045'
1570    #define CHAR_AMPERSAND              '\046'
1571    #define CHAR_APOSTROPHE             '\047'
1572    #define CHAR_LEFT_PARENTHESIS       '\050'
1573    #define CHAR_RIGHT_PARENTHESIS      '\051'
1574    #define CHAR_ASTERISK               '\052'
1575    #define CHAR_PLUS                   '\053'
1576    #define CHAR_COMMA                  '\054'
1577    #define CHAR_MINUS                  '\055'
1578    #define CHAR_DOT                    '\056'
1579    #define CHAR_SLASH                  '\057'
1580    #define CHAR_0                      '\060'
1581    #define CHAR_1                      '\061'
1582    #define CHAR_2                      '\062'
1583    #define CHAR_3                      '\063'
1584    #define CHAR_4                      '\064'
1585    #define CHAR_5                      '\065'
1586    #define CHAR_6                      '\066'
1587    #define CHAR_7                      '\067'
1588    #define CHAR_8                      '\070'
1589    #define CHAR_9                      '\071'
1590    #define CHAR_COLON                  '\072'
1591    #define CHAR_SEMICOLON              '\073'
1592    #define CHAR_LESS_THAN_SIGN         '\074'
1593    #define CHAR_EQUALS_SIGN            '\075'
1594    #define CHAR_GREATER_THAN_SIGN      '\076'
1595    #define CHAR_QUESTION_MARK          '\077'
1596    #define CHAR_COMMERCIAL_AT          '\100'
1597    #define CHAR_A                      '\101'
1598    #define CHAR_B                      '\102'
1599    #define CHAR_C                      '\103'
1600    #define CHAR_D                      '\104'
1601    #define CHAR_E                      '\105'
1602    #define CHAR_F                      '\106'
1603    #define CHAR_G                      '\107'
1604    #define CHAR_H                      '\110'
1605    #define CHAR_I                      '\111'
1606    #define CHAR_J                      '\112'
1607    #define CHAR_K                      '\113'
1608    #define CHAR_L                      '\114'
1609    #define CHAR_M                      '\115'
1610    #define CHAR_N                      '\116'
1611    #define CHAR_O                      '\117'
1612    #define CHAR_P                      '\120'
1613    #define CHAR_Q                      '\121'
1614    #define CHAR_R                      '\122'
1615    #define CHAR_S                      '\123'
1616    #define CHAR_T                      '\124'
1617    #define CHAR_U                      '\125'
1618    #define CHAR_V                      '\126'
1619    #define CHAR_W                      '\127'
1620    #define CHAR_X                      '\130'
1621    #define CHAR_Y                      '\131'
1622    #define CHAR_Z                      '\132'
1623    #define CHAR_LEFT_SQUARE_BRACKET    '\133'
1624    #define CHAR_BACKSLASH              '\134'
1625    #define CHAR_RIGHT_SQUARE_BRACKET   '\135'
1626    #define CHAR_CIRCUMFLEX_ACCENT      '\136'
1627    #define CHAR_UNDERSCORE             '\137'
1628    #define CHAR_GRAVE_ACCENT           '\140'
1629    #define CHAR_a                      '\141'
1630    #define CHAR_b                      '\142'
1631    #define CHAR_c                      '\143'
1632    #define CHAR_d                      '\144'
1633    #define CHAR_e                      '\145'
1634    #define CHAR_f                      '\146'
1635    #define CHAR_g                      '\147'
1636    #define CHAR_h                      '\150'
1637    #define CHAR_i                      '\151'
1638    #define CHAR_j                      '\152'
1639    #define CHAR_k                      '\153'
1640    #define CHAR_l                      '\154'
1641    #define CHAR_m                      '\155'
1642    #define CHAR_n                      '\156'
1643    #define CHAR_o                      '\157'
1644    #define CHAR_p                      '\160'
1645    #define CHAR_q                      '\161'
1646    #define CHAR_r                      '\162'
1647    #define CHAR_s                      '\163'
1648    #define CHAR_t                      '\164'
1649    #define CHAR_u                      '\165'
1650    #define CHAR_v                      '\166'
1651    #define CHAR_w                      '\167'
1652    #define CHAR_x                      '\170'
1653    #define CHAR_y                      '\171'
1654    #define CHAR_z                      '\172'
1655    #define CHAR_LEFT_CURLY_BRACKET     '\173'
1656    #define CHAR_VERTICAL_LINE          '\174'
1657    #define CHAR_RIGHT_CURLY_BRACKET    '\175'
1658    #define CHAR_TILDE                  '\176'
1659    
1660    #define STR_HT                      "\011"
1661    #define STR_VT                      "\013"
1662    #define STR_FF                      "\014"
1663    #define STR_CR                      "\015"
1664    #define STR_NL                      "\012"
1665    #define STR_BS                      "\010"
1666    #define STR_BEL                     "\007"
1667    #define STR_ESC                     "\033"
1668    #define STR_DEL                     "\177"
1669    
1670    #define STR_SPACE                   "\040"
1671    #define STR_EXCLAMATION_MARK        "\041"
1672    #define STR_QUOTATION_MARK          "\042"
1673    #define STR_NUMBER_SIGN             "\043"
1674    #define STR_DOLLAR_SIGN             "\044"
1675    #define STR_PERCENT_SIGN            "\045"
1676    #define STR_AMPERSAND               "\046"
1677    #define STR_APOSTROPHE              "\047"
1678    #define STR_LEFT_PARENTHESIS        "\050"
1679    #define STR_RIGHT_PARENTHESIS       "\051"
1680    #define STR_ASTERISK                "\052"
1681    #define STR_PLUS                    "\053"
1682    #define STR_COMMA                   "\054"
1683    #define STR_MINUS                   "\055"
1684    #define STR_DOT                     "\056"
1685    #define STR_SLASH                   "\057"
1686    #define STR_0                       "\060"
1687    #define STR_1                       "\061"
1688    #define STR_2                       "\062"
1689    #define STR_3                       "\063"
1690    #define STR_4                       "\064"
1691    #define STR_5                       "\065"
1692    #define STR_6                       "\066"
1693    #define STR_7                       "\067"
1694    #define STR_8                       "\070"
1695    #define STR_9                       "\071"
1696    #define STR_COLON                   "\072"
1697    #define STR_SEMICOLON               "\073"
1698    #define STR_LESS_THAN_SIGN          "\074"
1699    #define STR_EQUALS_SIGN             "\075"
1700    #define STR_GREATER_THAN_SIGN       "\076"
1701    #define STR_QUESTION_MARK           "\077"
1702    #define STR_COMMERCIAL_AT           "\100"
1703    #define STR_A                       "\101"
1704    #define STR_B                       "\102"
1705    #define STR_C                       "\103"
1706    #define STR_D                       "\104"
1707    #define STR_E                       "\105"
1708    #define STR_F                       "\106"
1709    #define STR_G                       "\107"
1710    #define STR_H                       "\110"
1711    #define STR_I                       "\111"
1712    #define STR_J                       "\112"
1713    #define STR_K                       "\113"
1714    #define STR_L                       "\114"
1715    #define STR_M                       "\115"
1716    #define STR_N                       "\116"
1717    #define STR_O                       "\117"
1718    #define STR_P                       "\120"
1719    #define STR_Q                       "\121"
1720    #define STR_R                       "\122"
1721    #define STR_S                       "\123"
1722    #define STR_T                       "\124"
1723    #define STR_U                       "\125"
1724    #define STR_V                       "\126"
1725    #define STR_W                       "\127"
1726    #define STR_X                       "\130"
1727    #define STR_Y                       "\131"
1728    #define STR_Z                       "\132"
1729    #define STR_LEFT_SQUARE_BRACKET     "\133"
1730    #define STR_BACKSLASH               "\134"
1731    #define STR_RIGHT_SQUARE_BRACKET    "\135"
1732    #define STR_CIRCUMFLEX_ACCENT       "\136"
1733    #define STR_UNDERSCORE              "\137"
1734    #define STR_GRAVE_ACCENT            "\140"
1735    #define STR_a                       "\141"
1736    #define STR_b                       "\142"
1737    #define STR_c                       "\143"
1738    #define STR_d                       "\144"
1739    #define STR_e                       "\145"
1740    #define STR_f                       "\146"
1741    #define STR_g                       "\147"
1742    #define STR_h                       "\150"
1743    #define STR_i                       "\151"
1744    #define STR_j                       "\152"
1745    #define STR_k                       "\153"
1746    #define STR_l                       "\154"
1747    #define STR_m                       "\155"
1748    #define STR_n                       "\156"
1749    #define STR_o                       "\157"
1750    #define STR_p                       "\160"
1751    #define STR_q                       "\161"
1752    #define STR_r                       "\162"
1753    #define STR_s                       "\163"
1754    #define STR_t                       "\164"
1755    #define STR_u                       "\165"
1756    #define STR_v                       "\166"
1757    #define STR_w                       "\167"
1758    #define STR_x                       "\170"
1759    #define STR_y                       "\171"
1760    #define STR_z                       "\172"
1761    #define STR_LEFT_CURLY_BRACKET      "\173"
1762    #define STR_VERTICAL_LINE           "\174"
1763    #define STR_RIGHT_CURLY_BRACKET     "\175"
1764    #define STR_TILDE                   "\176"
1765    
1766    #define STRING_ACCEPT0              STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1767    #define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1768    #define STRING_F0                   STR_F "\0"
1769    #define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"
1770    #define STRING_MARK0                STR_M STR_A STR_R STR_K "\0"
1771    #define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"
1772    #define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"
1773    #define STRING_THEN                 STR_T STR_H STR_E STR_N
1774    
1775    #define STRING_alpha0               STR_a STR_l STR_p STR_h STR_a "\0"
1776    #define STRING_lower0               STR_l STR_o STR_w STR_e STR_r "\0"
1777    #define STRING_upper0               STR_u STR_p STR_p STR_e STR_r "\0"
1778    #define STRING_alnum0               STR_a STR_l STR_n STR_u STR_m "\0"
1779    #define STRING_ascii0               STR_a STR_s STR_c STR_i STR_i "\0"
1780    #define STRING_blank0               STR_b STR_l STR_a STR_n STR_k "\0"
1781    #define STRING_cntrl0               STR_c STR_n STR_t STR_r STR_l "\0"
1782    #define STRING_digit0               STR_d STR_i STR_g STR_i STR_t "\0"
1783    #define STRING_graph0               STR_g STR_r STR_a STR_p STR_h "\0"
1784    #define STRING_print0               STR_p STR_r STR_i STR_n STR_t "\0"
1785    #define STRING_punct0               STR_p STR_u STR_n STR_c STR_t "\0"
1786    #define STRING_space0               STR_s STR_p STR_a STR_c STR_e "\0"
1787    #define STRING_word0                STR_w STR_o STR_r STR_d       "\0"
1788    #define STRING_xdigit               STR_x STR_d STR_i STR_g STR_i STR_t
1789    
1790    #define STRING_DEFINE               STR_D STR_E STR_F STR_I STR_N STR_E
1791    
1792    #define STRING_CR_RIGHTPAR             STR_C STR_R STR_RIGHT_PARENTHESIS
1793    #define STRING_LF_RIGHTPAR             STR_L STR_F STR_RIGHT_PARENTHESIS
1794    #define STRING_CRLF_RIGHTPAR           STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1795    #define STRING_ANY_RIGHTPAR            STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1796    #define STRING_ANYCRLF_RIGHTPAR        STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1797    #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
1798    #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
1799    #ifdef COMPILE_PCRE8
1800    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1801    #endif
1802    #ifdef COMPILE_PCRE16
1803    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
1804    #endif
1805    #ifdef COMPILE_PCRE32
1806    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS
1807    #endif
1808    #define STRING_UCP_RIGHTPAR            STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1809    #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
1810    
1811    #endif  /* SUPPORT_UTF */
1812    
1813  /* Escape items that are just an encoding of a particular data value. */  /* Escape items that are just an encoding of a particular data value. */
1814    
1815  #ifndef ESC_e  #ifndef ESC_e
1816  #define ESC_e 27  #define ESC_e CHAR_ESC
1817  #endif  #endif
1818    
1819  #ifndef ESC_f  #ifndef ESC_f
1820  #define ESC_f '\f'  #define ESC_f CHAR_FF
1821  #endif  #endif
1822    
1823  #ifndef ESC_n  #ifndef ESC_n
1824  #define ESC_n '\n'  #define ESC_n CHAR_LF
1825  #endif  #endif
1826    
1827  #ifndef ESC_r  #ifndef ESC_r
1828  #define ESC_r '\r'  #define ESC_r CHAR_CR
1829  #endif  #endif
1830    
1831  /* 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
1832  (presumably because of all the others like size_t). */  (presumably because of all the others like size_t). */
1833    
1834  #ifndef ESC_tee  #ifndef ESC_tee
1835  #define ESC_tee '\t'  #define ESC_tee CHAR_HT
1836  #endif  #endif
1837    
1838  /* Codes for different types of Unicode property */  /* Codes for different types of Unicode property */
1839    
1840  #define PT_ANY        0    /* Any property - matches all chars */  #define PT_ANY        0    /* Any property - matches all chars */
1841  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */
1842  #define PT_GC         2    /* General characteristic (e.g. L) */  #define PT_GC         2    /* Specified general characteristic (e.g. L) */
1843  #define PT_PC         3    /* Particular characteristic (e.g. Lu) */  #define PT_PC         3    /* Specified particular characteristic (e.g. Lu) */
1844  #define PT_SC         4    /* Script (e.g. Han) */  #define PT_SC         4    /* Script (e.g. Han) */
1845    #define PT_ALNUM      5    /* Alphanumeric - the union of L and N */
1846    #define PT_SPACE      6    /* Perl space - Z plus 9,10,12,13 */
1847    #define PT_PXSPACE    7    /* POSIX space - Z plus 9,10,11,12,13 */
1848    #define PT_WORD       8    /* Word - L plus N plus underscore */
1849    #define PT_CLIST      9    /* Pseudo-property: match character list */
1850    
1851  /* 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
1852  contain UTF-8 characters with values greater than 255. */  contain characters with values greater than 255. */
1853    
1854  #define XCL_NOT    0x01    /* Flag: this is a negative class */  #define XCL_NOT    0x01    /* Flag: this is a negative class */
1855  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */
# Line 596  contain UTF-8 characters with values gre Line 1862  contain UTF-8 characters with values gre
1862    
1863  /* 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
1864  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
1865  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
1866  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
1867  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
1868  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
1869  There are two tests in the code for an escape greater than ESC_b and less than  non-DOTALL mode, "." behaves like \N.
1870  ESC_Z to detect the types that may be repeated. These are the types that  
1871  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.
1872  character, that code will have to change. */  when PCRE_UCP is set and replacement of \d etc by \p sequences is required.
1873    They must be contiguous, and remain in order so that the replacements can be
1874    looked up from a table.
1875    
1876    Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in
1877    check_escape(). There are two tests in the code for an escape
1878    greater than ESC_b and less than ESC_Z to detect the types that may be
1879    repeated. These are the types that consume characters. If any new escapes are
1880    put in between that don't consume a character, that code will have to change.
1881    */
1882    
1883  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,
1884         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,
1885         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,
1886           ESC_E, ESC_Q, ESC_g, ESC_k,
1887           ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu };
1888    
1889    /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
 /* Opcode table: OP_BRA must be last, as all values >= it are used for brackets  
 that extract substrings. Starting from 1 (i.e. after OP_END), the values up to  
1890  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.
1891    
 To keep stored, compiled patterns compatible, new opcodes should be added  
 immediately before OP_BRA, where (since release 7.0) a gap is left for this  
 purpose.  
   
1892  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1893  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
1894  "coptable" in pcre_dfa_exec.c that must be updated. */  "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1895    
1896  enum {  enum {
1897    OP_END,            /* 0 End of pattern */    OP_END,            /* 0 End of pattern */
# Line 638  enum { Line 1909  enum {
1909    OP_WHITESPACE,         /*  9 \s */    OP_WHITESPACE,         /*  9 \s */
1910    OP_NOT_WORDCHAR,       /* 10 \W */    OP_NOT_WORDCHAR,       /* 10 \W */
1911    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 */  
1912    
1913    /* These three must follow the previous three, in the same order. There's a    OP_ANY,            /* 12 Match any character except newline (\N) */
1914      OP_ALLANY,         /* 13 Match any character */
1915      OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1916      OP_NOTPROP,        /* 15 \P (not Unicode property) */
1917      OP_PROP,           /* 16 \p (Unicode property) */
1918      OP_ANYNL,          /* 17 \R (any newline sequence) */
1919      OP_NOT_HSPACE,     /* 18 \H (not horizontal whitespace) */
1920      OP_HSPACE,         /* 19 \h (horizontal whitespace) */
1921      OP_NOT_VSPACE,     /* 20 \V (not vertical whitespace) */
1922      OP_VSPACE,         /* 21 \v (vertical whitespace) */
1923      OP_EXTUNI,         /* 22 \X (extended Unicode sequence */
1924      OP_EODN,           /* 23 End of data or \n at end of data (\Z) */
1925      OP_EOD,            /* 24 End of data (\z) */
1926    
1927      OP_CIRC,           /* 25 Start of line - not multiline */
1928      OP_CIRCM,          /* 26 Start of line - multiline */
1929      OP_DOLL,           /* 27 End of line - not multiline */
1930      OP_DOLLM,          /* 28 End of line - multiline */
1931      OP_CHAR,           /* 29 Match one character, casefully */
1932      OP_CHARI,          /* 30 Match one character, caselessly */
1933      OP_NOT,            /* 31 Match one character, not the given one, casefully */
1934      OP_NOTI,           /* 32 Match one character, not the given one, caselessly */
1935    
1936      /* The following sets of 13 opcodes must always be kept in step because
1937      the offset from the first one is used to generate the others. */
1938    
1939      /**** Single characters, caseful, must precede the caseless ones ****/
1940    
1941      OP_STAR,           /* 33 The maximizing and minimizing versions of */
1942      OP_MINSTAR,        /* 34 these six opcodes must come in pairs, with */
1943      OP_PLUS,           /* 35 the minimizing one second. */
1944      OP_MINPLUS,        /* 36 */
1945      OP_QUERY,          /* 37 */
1946      OP_MINQUERY,       /* 38 */
1947    
1948      OP_UPTO,           /* 39 From 0 to n matches of one character, caseful*/
1949      OP_MINUPTO,        /* 40 */
1950      OP_EXACT,          /* 41 Exactly n matches */
1951    
1952      OP_POSSTAR,        /* 42 Possessified star, caseful */
1953      OP_POSPLUS,        /* 43 Possessified plus, caseful */
1954      OP_POSQUERY,       /* 44 Posesssified query, caseful */
1955      OP_POSUPTO,        /* 45 Possessified upto, caseful */
1956    
1957      /**** Single characters, caseless, must follow the caseful ones */
1958    
1959      OP_STARI,          /* 46 */
1960      OP_MINSTARI,       /* 47 */
1961      OP_PLUSI,          /* 48 */
1962      OP_MINPLUSI,       /* 49 */
1963      OP_QUERYI,         /* 50 */
1964      OP_MINQUERYI,      /* 51 */
1965    
1966      OP_UPTOI,          /* 52 From 0 to n matches of one character, caseless */
1967      OP_MINUPTOI,       /* 53 */
1968      OP_EXACTI,         /* 54 */
1969    
1970      OP_POSSTARI,       /* 55 Possessified star, caseless */
1971      OP_POSPLUSI,       /* 56 Possessified plus, caseless */
1972      OP_POSQUERYI,      /* 57 Posesssified query, caseless */
1973      OP_POSUPTOI,       /* 58 Possessified upto, caseless */
1974    
1975      /**** The negated ones must follow the non-negated ones, and match them ****/
1976      /**** Negated single character, caseful; must precede the caseless ones ****/
1977    
1978      OP_NOTSTAR,        /* 59 The maximizing and minimizing versions of */
1979      OP_NOTMINSTAR,     /* 60 these six opcodes must come in pairs, with */
1980      OP_NOTPLUS,        /* 61 the minimizing one second. They must be in */
1981      OP_NOTMINPLUS,     /* 62 exactly the same order as those above. */
1982      OP_NOTQUERY,       /* 63 */
1983      OP_NOTMINQUERY,    /* 64 */
1984    
1985      OP_NOTUPTO,        /* 65 From 0 to n matches, caseful */
1986      OP_NOTMINUPTO,     /* 66 */
1987      OP_NOTEXACT,       /* 67 Exactly n matches */
1988    
1989      OP_NOTPOSSTAR,     /* 68 Possessified versions, caseful */
1990      OP_NOTPOSPLUS,     /* 69 */
1991      OP_NOTPOSQUERY,    /* 70 */
1992      OP_NOTPOSUPTO,     /* 71 */
1993    
1994      /**** Negated single character, caseless; must follow the caseful ones ****/
1995    
1996      OP_NOTSTARI,       /* 72 */
1997      OP_NOTMINSTARI,    /* 73 */
1998      OP_NOTPLUSI,       /* 74 */
1999      OP_NOTMINPLUSI,    /* 75 */
2000      OP_NOTQUERYI,      /* 76 */
2001      OP_NOTMINQUERYI,   /* 77 */
2002    
2003      OP_NOTUPTOI,       /* 78 From 0 to n matches, caseless */
2004      OP_NOTMINUPTOI,    /* 79 */
2005      OP_NOTEXACTI,      /* 80 Exactly n matches */
2006    
2007      OP_NOTPOSSTARI,    /* 81 Possessified versions, caseless */
2008      OP_NOTPOSPLUSI,    /* 82 */
2009      OP_NOTPOSQUERYI,   /* 83 */
2010      OP_NOTPOSUPTOI,    /* 84 */
2011    
2012      /**** Character types ****/
2013    
2014      OP_TYPESTAR,       /* 85 The maximizing and minimizing versions of */
2015      OP_TYPEMINSTAR,    /* 86 these six opcodes must come in pairs, with */
2016      OP_TYPEPLUS,       /* 87 the minimizing one second. These codes must */
2017      OP_TYPEMINPLUS,    /* 88 be in exactly the same order as those above. */
2018      OP_TYPEQUERY,      /* 89 */
2019      OP_TYPEMINQUERY,   /* 90 */
2020    
2021      OP_TYPEUPTO,       /* 91 From 0 to n matches */
2022      OP_TYPEMINUPTO,    /* 92 */
2023      OP_TYPEEXACT,      /* 93 Exactly n matches */
2024    
2025      OP_TYPEPOSSTAR,    /* 94 Possessified versions */
2026      OP_TYPEPOSPLUS,    /* 95 */
2027      OP_TYPEPOSQUERY,   /* 96 */
2028      OP_TYPEPOSUPTO,    /* 97 */
2029    
2030      /* These are used for character classes and back references; only the
2031      first six are the same as the sets above. */
2032    
2033      OP_CRSTAR,         /* 98 The maximizing and minimizing versions of */
2034      OP_CRMINSTAR,      /* 99 all these opcodes must come in pairs, with */
2035      OP_CRPLUS,         /* 100 the minimizing one second. These codes must */
2036      OP_CRMINPLUS,      /* 101 be in exactly the same order as those above. */
2037      OP_CRQUERY,        /* 102 */
2038      OP_CRMINQUERY,     /* 103 */
2039    
2040      OP_CRRANGE,        /* 104 These are different to the three sets above. */
2041      OP_CRMINRANGE,     /* 105 */
2042    
2043      /* End of quantifier opcodes */
2044    
2045      OP_CLASS,          /* 106 Match a character class, chars < 256 only */
2046      OP_NCLASS,         /* 107 Same, but the bitmap was created from a negative
2047                                  class - the difference is relevant only when a
2048                                  character > 255 is encountered. */
2049      OP_XCLASS,         /* 108 Extended class for handling > 255 chars within the
2050                                  class. This does both positive and negative. */
2051      OP_REF,            /* 109 Match a back reference, casefully */
2052      OP_REFI,           /* 110 Match a back reference, caselessly */
2053      OP_RECURSE,        /* 111 Match a numbered subpattern (possibly recursive) */
2054      OP_CALLOUT,        /* 112 Call out to external function if provided */
2055    
2056      OP_ALT,            /* 113 Start of alternation */
2057      OP_KET,            /* 114 End of group that doesn't have an unbounded repeat */
2058      OP_KETRMAX,        /* 115 These two must remain together and in this */
2059      OP_KETRMIN,        /* 116 order. They are for groups the repeat for ever. */
2060      OP_KETRPOS,        /* 117 Possessive unlimited repeat. */
2061    
2062      /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
2063      asserts must remain in order. */
2064    
2065      OP_REVERSE,        /* 118 Move pointer back - used in lookbehind assertions */
2066      OP_ASSERT,         /* 119 Positive lookahead */
2067      OP_ASSERT_NOT,     /* 120 Negative lookahead */
2068      OP_ASSERTBACK,     /* 121 Positive lookbehind */
2069      OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
2070    
2071      /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
2072      after the assertions, with ONCE first, as there's a test for >= ONCE for a
2073      subpattern that isn't an assertion. The POS versions must immediately follow
2074      the non-POS versions in each case. */
2075    
2076      OP_ONCE,           /* 123 Atomic group, contains captures */
2077      OP_ONCE_NC,        /* 124 Atomic group containing no captures */
2078      OP_BRA,            /* 125 Start of non-capturing bracket */
2079      OP_BRAPOS,         /* 126 Ditto, with unlimited, possessive repeat */
2080      OP_CBRA,           /* 127 Start of capturing bracket */
2081      OP_CBRAPOS,        /* 128 Ditto, with unlimited, possessive repeat */
2082      OP_COND,           /* 129 Conditional group */
2083    
2084      /* These five must follow the previous five, in the same order. There's a
2085    check for >= SBRA to distinguish the two sets. */    check for >= SBRA to distinguish the two sets. */
2086    
2087    OP_SBRA,           /* 96 Start of non-capturing bracket, check empty  */    OP_SBRA,           /* 130 Start of non-capturing bracket, check empty  */
2088    OP_SCBRA,          /* 97 Start of capturing bracket, check empty */    OP_SBRAPOS,        /* 131 Ditto, with unlimited, possessive repeat */
2089    OP_SCOND,          /* 98 Conditional group, check empty */    OP_SCBRA,          /* 132 Start of capturing bracket, check empty */
2090      OP_SCBRAPOS,       /* 133 Ditto, with unlimited, possessive repeat */
2091    OP_CREF,           /* 99 Used to hold a capture number as condition */    OP_SCOND,          /* 134 Conditional group, check empty */
2092    OP_RREF,           /* 100 Used to hold a recursion number as condition */  
2093    OP_DEF,            /* 101 The DEFINE condition */    /* The next two pairs must (respectively) be kept together. */
2094    
2095      OP_CREF,           /* 135 Used to hold a capture number as condition */
2096      OP_NCREF,          /* 136 Same, but generated by a name reference*/
2097      OP_RREF,           /* 137 Used to hold a recursion number as condition */
2098      OP_NRREF,          /* 138 Same, but generated by a name reference*/
2099      OP_DEF,            /* 139 The DEFINE condition */
2100    
2101      OP_BRAZERO,        /* 140 These two must remain together and in this */
2102      OP_BRAMINZERO,     /* 141 order. */
2103      OP_BRAPOSZERO,     /* 142 */
2104    
2105      /* These are backtracking control verbs */
2106    
2107      OP_MARK,           /* 143 always has an argument */
2108      OP_PRUNE,          /* 144 */
2109      OP_PRUNE_ARG,      /* 145 same, but with argument */
2110      OP_SKIP,           /* 146 */
2111      OP_SKIP_ARG,       /* 147 same, but with argument */
2112      OP_THEN,           /* 148 */
2113      OP_THEN_ARG,       /* 149 same, but with argument */
2114      OP_COMMIT,         /* 150 */
2115    
2116      /* These are forced failure and success verbs */
2117    
2118      OP_FAIL,           /* 151 */
2119      OP_ACCEPT,         /* 152 */
2120      OP_ASSERT_ACCEPT,  /* 153 Used inside assertions */
2121      OP_CLOSE,          /* 154 Used before OP_ACCEPT to close open captures */
2122    
2123      /* This is used to skip a subpattern with a {0} quantifier */
2124    
2125      OP_SKIPZERO,       /* 155 */
2126    
2127      /* This is not an opcode, but is used to check that tables indexed by opcode
2128      are the correct length, in order to catch updating errors - there have been
2129      some in the past. */
2130    
2131    OP_BRAZERO,        /* 102 These two must remain together and in this */    OP_TABLE_LENGTH
   OP_BRAMINZERO      /* 103 order. */  
2132  };  };
2133    
2134    /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
2135    definitions that follow must also be updated to match. There are also tables
2136    called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
2137    
2138    
2139  /* 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
2140  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
2141    only in pcre_printint.c, which fills out the full names in many cases (and in
2142    some cases doesn't actually use these names at all). */
2143    
2144  #define OP_NAME_LIST \  #define OP_NAME_LIST \
2145    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \
2146    "\\S", "\\s", "\\W", "\\w", "Any", "Anybyte",                   \    "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte",         \
2147    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \
2148    "extuni",  "\\Z", "\\z",                                        \    "extuni",  "\\Z", "\\z",                                        \
2149    "Opt", "^", "$", "char", "charnc", "not",                       \    "^", "^", "$", "$", "char", "chari", "not", "noti",             \
2150    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2151      "{", "{", "{",                                                  \
2152    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2153    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2154      "{", "{", "{",                                                  \
2155      "*+","++", "?+", "{",                                           \
2156      "*", "*?", "+", "+?", "?", "??",                                \
2157      "{", "{", "{",                                                  \
2158      "*+","++", "?+", "{",                                           \
2159      "*", "*?", "+", "+?", "?", "??",                                \
2160      "{", "{", "{",                                                  \
2161    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2162    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \
2163    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2164    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \
2165    "class", "nclass", "xclass", "Ref", "Recurse", "Callout",       \    "class", "nclass", "xclass", "Ref", "Refi",                     \
2166    "Alt", "Ket", "KetRmax", "KetRmin", "Assert", "Assert not",     \    "Recurse", "Callout",                                           \
2167    "AssertB", "AssertB not", "Reverse",                            \    "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos",                  \
2168    "Once", "Bra", "CBra", "Cond", "SBra", "SCBra", "SCond",        \    "Reverse", "Assert", "Assert not", "AssertB", "AssertB not",    \
2169    "Cond ref", "Cond rec", "Cond def", "Brazero", "Braminzero"    "Once", "Once_NC",                                              \
2170      "Bra", "BraPos", "CBra", "CBraPos",                             \
2171      "Cond",                                                         \
2172      "SBra", "SBraPos", "SCBra", "SCBraPos",                         \
2173      "SCond",                                                        \
2174      "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def",   \
2175      "Brazero", "Braminzero", "Braposzero",                          \
2176      "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP",                  \
2177      "*THEN", "*THEN", "*COMMIT", "*FAIL",                           \
2178      "*ACCEPT", "*ASSERT_ACCEPT",                                    \
2179      "Close", "Skip zero"
2180    
2181    
2182  /* 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 800  in UTF-8 mode. The code that uses this t Line 2192  in UTF-8 mode. The code that uses this t
2192    1,                             /* End                                    */ \    1,                             /* End                                    */ \
2193    1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \    1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \
2194    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \
2195    1, 1,                          /* Any, Anybyte                           */ \    1, 1, 1,                       /* Any, AllAny, Anybyte                   */ \
2196    3, 3, 1,                       /* NOTPROP, PROP, EXTUNI                  */ \    3, 3,                          /* \P, \p                                 */ \
2197    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \
2198    1, 1, 2, 1, 1,                 /* \Z, \z, Opt, ^, $                      */ \    1,                             /* \X                                     */ \
2199      1, 1, 1, 1, 1, 1,              /* \Z, \z, ^, ^M, $, $M                   */ \
2200    2,                             /* Char  - the minimum length             */ \    2,                             /* Char  - the minimum length             */ \
2201    2,                             /* Charnc  - the minimum length           */ \    2,                             /* Chari  - the minimum length            */ \
2202    2,                             /* not                                    */ \    2,                             /* not                                    */ \
2203    /* Positive single-char repeats                            ** These are  */ \    2,                             /* noti                                   */ \
2204    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??      ** minima in  */ \    /* Positive single-char repeats                             ** These are */ \
2205    4, 4, 4,                       /* upto, minupto, exact     ** UTF-8 mode */ \    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??       ** minima in */ \
2206    2, 2, 2, 4,                    /* *+, ++, ?+, upto+                      */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto, minupto             ** mode      */ \
2207      2+IMM2_SIZE,                   /* exact                                  */ \
2208      2, 2, 2, 2+IMM2_SIZE,          /* *+, ++, ?+, upto+                      */ \
2209      2, 2, 2, 2, 2, 2,              /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8     */ \
2210      2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto I, minupto I                      */ \
2211      2+IMM2_SIZE,                   /* exact I                                */ \
2212      2, 2, 2, 2+IMM2_SIZE,          /* *+I, ++I, ?+I, upto+I                  */ \
2213    /* Negative single-char repeats - only for chars < 256                   */ \    /* Negative single-char repeats - only for chars < 256                   */ \
2214    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \
2215    4, 4, 4,                       /* NOT upto, minupto, exact               */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto, minupto                      */ \
2216    2, 2, 2, 4,                    /* Possessive *, +, ?, upto               */ \    2+IMM2_SIZE,                   /* NOT exact                              */ \
2217      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *, +, ?, upto           */ \
2218      2, 2, 2, 2, 2, 2,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */ \
2219      2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto I, minupto I                  */ \
2220      2+IMM2_SIZE,                   /* NOT exact I                            */ \
2221      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *I, +I, ?I, upto I      */ \
2222    /* Positive type repeats                                                 */ \    /* Positive type repeats                                                 */ \
2223    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \
2224    4, 4, 4,                       /* Type upto, minupto, exact              */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* Type upto, minupto                     */ \
2225    2, 2, 2, 4,                    /* Possessive *+, ++, ?+, upto+           */ \    2+IMM2_SIZE,                   /* Type exact                             */ \
2226      2, 2, 2, 2+IMM2_SIZE,          /* Possessive *+, ++, ?+, upto+           */ \
2227    /* Character class & ref repeats                                         */ \    /* Character class & ref repeats                                         */ \
2228    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \
2229    5, 5,                          /* CRRANGE, CRMINRANGE                    */ \    1+2*IMM2_SIZE, 1+2*IMM2_SIZE,  /* CRRANGE, CRMINRANGE                    */ \
2230   33,                             /* CLASS                                  */ \    1+(32/sizeof(pcre_uchar)),     /* CLASS                                  */ \
2231   33,                             /* NCLASS                                 */ \    1+(32/sizeof(pcre_uchar)),     /* NCLASS                                 */ \
2232    0,                             /* XCLASS - variable length               */ \    0,                             /* XCLASS - variable length               */ \
2233    3,                             /* REF                                    */ \    1+IMM2_SIZE,                   /* REF                                    */ \
2234      1+IMM2_SIZE,                   /* REFI                                   */ \
2235    1+LINK_SIZE,                   /* RECURSE                                */ \    1+LINK_SIZE,                   /* RECURSE                                */ \
2236    2+2*LINK_SIZE,                 /* CALLOUT                                */ \    2+2*LINK_SIZE,                 /* CALLOUT                                */ \
2237    1+LINK_SIZE,                   /* Alt                                    */ \    1+LINK_SIZE,                   /* Alt                                    */ \
2238    1+LINK_SIZE,                   /* Ket                                    */ \    1+LINK_SIZE,                   /* Ket                                    */ \
2239    1+LINK_SIZE,                   /* KetRmax                                */ \    1+LINK_SIZE,                   /* KetRmax                                */ \
2240    1+LINK_SIZE,                   /* KetRmin                                */ \    1+LINK_SIZE,                   /* KetRmin                                */ \
2241      1+LINK_SIZE,                   /* KetRpos                                */ \
2242      1+LINK_SIZE,                   /* Reverse                                */ \
2243    1+LINK_SIZE,                   /* Assert                                 */ \    1+LINK_SIZE,                   /* Assert                                 */ \
2244    1+LINK_SIZE,                   /* Assert not                             */ \    1+LINK_SIZE,                   /* Assert not                             */ \
2245    1+LINK_SIZE,                   /* Assert behind                          */ \    1+LINK_SIZE,                   /* Assert behind                          */ \
2246    1+LINK_SIZE,                   /* Assert behind not                      */ \    1+LINK_SIZE,                   /* Assert behind not                      */ \
   1+LINK_SIZE,                   /* Reverse                                */ \  
2247    1+LINK_SIZE,                   /* ONCE                                   */ \    1+LINK_SIZE,                   /* ONCE                                   */ \
2248      1+LINK_SIZE,                   /* ONCE_NC                                */ \
2249    1+LINK_SIZE,                   /* BRA                                    */ \    1+LINK_SIZE,                   /* BRA                                    */ \
2250    3+LINK_SIZE,                   /* CBRA                                   */ \    1+LINK_SIZE,                   /* BRAPOS                                 */ \
2251      1+LINK_SIZE+IMM2_SIZE,         /* CBRA                                   */ \
2252      1+LINK_SIZE+IMM2_SIZE,         /* CBRAPOS                                */ \
2253    1+LINK_SIZE,                   /* COND                                   */ \    1+LINK_SIZE,                   /* COND                                   */ \
2254    1+LINK_SIZE,                   /* SBRA                                   */ \    1+LINK_SIZE,                   /* SBRA                                   */ \
2255    3+LINK_SIZE,                   /* SCBRA                                  */ \    1+LINK_SIZE,                   /* SBRAPOS                                */ \
2256      1+LINK_SIZE+IMM2_SIZE,         /* SCBRA                                  */ \
2257      1+LINK_SIZE+IMM2_SIZE,         /* SCBRAPOS                               */ \
2258    1+LINK_SIZE,                   /* SCOND                                  */ \    1+LINK_SIZE,                   /* SCOND                                  */ \
2259    3,                             /* CREF                                   */ \    1+IMM2_SIZE, 1+IMM2_SIZE,      /* CREF, NCREF                            */ \
2260    3,                             /* RREF                                   */ \    1+IMM2_SIZE, 1+IMM2_SIZE,      /* RREF, NRREF                            */ \
2261    1,                             /* DEF                                    */ \    1,                             /* DEF                                    */ \
2262    1, 1,                          /* BRAZERO, BRAMINZERO                    */ \    1, 1, 1,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */ \
2263      3, 1, 3,                       /* MARK, PRUNE, PRUNE_ARG                 */ \
2264      1, 3,                          /* SKIP, SKIP_ARG                         */ \
2265      1, 3,                          /* THEN, THEN_ARG                         */ \
2266      1, 1, 1, 1,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */ \
2267      1+IMM2_SIZE, 1                 /* CLOSE, SKIPZERO                        */
2268    
2269    /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
2270  /* A magic value for OP_RREF to indicate the "any recursion" condition. */  condition. */
2271    
2272  #define RREF_ANY  0xffff  #define RREF_ANY  0xffff
2273    
2274  /* 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
2275  tracked. */  easily be tracked. When a new number is added, the table called eint in
2276    pcreposix.c must be updated. */
2277    
2278  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
2279         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
2280         ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,         ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
2281         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
2282         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
2283         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58 };         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
2284           ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
2285           ERR70, ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERR77, ERRCOUNT };
2286    
2287    /* JIT compiling modes. The function list is indexed by them. */
2288    enum { JIT_COMPILE, JIT_PARTIAL_SOFT_COMPILE, JIT_PARTIAL_HARD_COMPILE,
2289           JIT_NUMBER_OF_COMPILE_MODES };
2290    
2291  /* 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
2292  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 878  Because people can now save and re-use c Line 2302  Because people can now save and re-use c
2302  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
2303  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
2304  fields are present. Currently PCRE always sets the dummy fields to zero.  fields are present. Currently PCRE always sets the dummy fields to zero.
2305  NOTE NOTE NOTE:  NOTE NOTE NOTE
2306  */  */
2307    
2308  typedef struct real_pcre {  #if defined COMPILE_PCRE8
2309    #define REAL_PCRE real_pcre
2310    #elif defined COMPILE_PCRE16
2311    #define REAL_PCRE real_pcre16
2312    #elif defined COMPILE_PCRE32
2313    #define REAL_PCRE real_pcre32
2314    #endif
2315    
2316    /* It is necessary to fork the struct for 32 bit, since it needs to use
2317     * pcre_uchar for first_char and req_char. Can't put an ifdef inside the
2318     * typedef since pcretest needs access to  the struct of the 8-, 16-
2319     * and 32-bit variants. */
2320    
2321    typedef struct real_pcre8_or_16 {
2322    pcre_uint32 magic_number;    pcre_uint32 magic_number;
2323    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2324    pcre_uint32 options;    pcre_uint32 options;            /* Public options */
2325    pcre_uint32 dummy1;             /* For future use, maybe */    pcre_uint16 flags;              /* Private flags */
2326      pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2327    pcre_uint16 top_bracket;    pcre_uint16 top_bracket;        /* Highest numbered group */
2328    pcre_uint16 top_backref;    pcre_uint16 top_backref;        /* Highest numbered back reference */
2329    pcre_uint16 first_byte;    pcre_uint16 first_char;         /* Starting character */
2330    pcre_uint16 req_byte;    pcre_uint16 req_char;           /* This character must be seen */
2331    pcre_uint16 name_table_offset;  /* Offset to name table that follows */    pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2332    pcre_uint16 name_entry_size;    /* Size of any name items */    pcre_uint16 name_entry_size;    /* Size of any name items */
2333    pcre_uint16 name_count;         /* Number of name items */    pcre_uint16 name_count;         /* Number of name items */
2334    pcre_uint16 ref_count;          /* Reference count */    pcre_uint16 ref_count;          /* Reference count */
2335      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2336      const pcre_uint8 *nullpad;      /* NULL padding */
2337    } real_pcre8_or_16;
2338    
2339    typedef struct real_pcre8_or_16 real_pcre;
2340    typedef struct real_pcre8_or_16 real_pcre16;
2341    
2342    const unsigned char *tables;    /* Pointer to tables or NULL for std */  typedef struct real_pcre32 {
2343    const unsigned char *nullpad;   /* NULL padding */    pcre_uint32 magic_number;
2344  } real_pcre;    pcre_uint32 size;               /* Total that was malloced */
2345      pcre_uint32 options;            /* Public options */
2346      pcre_uint16 flags;              /* Private flags */
2347      pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2348      pcre_uint16 top_bracket;        /* Highest numbered group */
2349      pcre_uint16 top_backref;        /* Highest numbered back reference */
2350      pcre_uint32 first_char;         /* Starting character */
2351      pcre_uint32 req_char;           /* This character must be seen */
2352      pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2353      pcre_uint16 name_entry_size;    /* Size of any name items */
2354      pcre_uint16 name_count;         /* Number of name items */
2355      pcre_uint16 ref_count;          /* Reference count */
2356      pcre_uint16 dummy1;             /* for later expansion */
2357      pcre_uint16 dummy2;             /* for later expansion */
2358      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2359      void *nullpad;                  /* for later expansion */
2360    } real_pcre32;
2361    
2362    /* Assert that the size of REAL_PCRE is divisible by 8 */
2363    typedef int __assert_real_pcre_size_divisible_8[(sizeof(REAL_PCRE) % 8) == 0 ? 1 : -1];
2364    
2365    /* Needed in pcretest to access some fields in the real_pcre* structures
2366     * directly. They're unified for 8/16/32 bits since the structs only differ
2367     * after these fields; if that ever changes, need to fork those defines into
2368     * 8/16 and 32 bit versions. */
2369    #define REAL_PCRE_MAGIC(re)     (((REAL_PCRE*)re)->magic_number)
2370    #define REAL_PCRE_SIZE(re)      (((REAL_PCRE*)re)->size)
2371    #define REAL_PCRE_OPTIONS(re)   (((REAL_PCRE*)re)->options)
2372    #define REAL_PCRE_FLAGS(re)     (((REAL_PCRE*)re)->flags)
2373    
2374  /* 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
2375  remark (see NOTE above) about extending this structure applies. */  remark (see NOTE above) about extending this structure applies. */
2376    
2377  typedef struct pcre_study_data {  typedef struct pcre_study_data {
2378    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2379    pcre_uint32 options;    pcre_uint32 flags;              /* Private flags */
2380    uschar start_bits[32];    pcre_uint8 start_bits[32];      /* Starting char bits */
2381      pcre_uint32 minlength;          /* Minimum subject length */
2382  } pcre_study_data;  } pcre_study_data;
2383    
2384    /* Structure for building a chain of open capturing subpatterns during
2385    compiling, so that instructions to close them can be compiled when (*ACCEPT) is
2386    encountered. This is also used to identify subpatterns that contain recursive
2387    back references to themselves, so that they can be made atomic. */
2388    
2389    typedef struct open_capitem {
2390      struct open_capitem *next;    /* Chain link */
2391      pcre_uint16 number;           /* Capture number */
2392      pcre_uint16 flag;             /* Set TRUE if recursive back ref */
2393    } open_capitem;
2394    
2395  /* Structure for passing "static" information around between the functions  /* Structure for passing "static" information around between the functions
2396  doing the compiling, so that they are thread-safe. */  doing the compiling, so that they are thread-safe. */
2397    
2398  typedef struct compile_data {  typedef struct compile_data {
2399    const uschar *lcc;            /* Points to lower casing table */    const pcre_uint8 *lcc;            /* Points to lower casing table */
2400    const uschar *fcc;            /* Points to case-flipping table */    const pcre_uint8 *fcc;            /* Points to case-flipping table */
2401    const uschar *cbits;          /* Points to character type table */    const pcre_uint8 *cbits;          /* Points to character type table */
2402    const uschar *ctypes;         /* Points to table of type maps */    const pcre_uint8 *ctypes;         /* Points to table of type maps */
2403    const uschar *start_workspace;/* The start of working space */    const pcre_uchar *start_workspace;/* The start of working space */
2404    const uschar *start_code;     /* The start of the compiled code */    const pcre_uchar *start_code;     /* The start of the compiled code */
2405    const uschar *start_pattern;  /* The start of the pattern */    const pcre_uchar *start_pattern;  /* The start of the pattern */
2406    const uschar *end_pattern;    /* The end of the pattern */    const pcre_uchar *end_pattern;    /* The end of the pattern */
2407    uschar *hwm;                  /* High watermark of workspace */    open_capitem *open_caps;          /* Chain of open capture items */
2408    uschar *name_table;           /* The name/number table */    pcre_uchar *hwm;                  /* High watermark of workspace */
2409    int  names_found;             /* Number of entries so far */    pcre_uchar *name_table;           /* The name/number table */
2410    int  name_entry_size;         /* Size of each entry */    int  names_found;                 /* Number of entries so far */
2411    int  bracount;                /* Count of capturing parens */    int  name_entry_size;             /* Size of each entry */
2412    int  top_backref;             /* Maximum back reference */    int  workspace_size;              /* Size of workspace */
2413    unsigned int backref_map;     /* Bitmap of low back refs */    unsigned int  bracount;           /* Count of capturing parens as we compile */
2414    int  external_options;        /* External (initial) options */    int  final_bracount;              /* Saved value after first pass */
2415    int  req_varyopt;             /* "After variable item" flag for reqbyte */    int  max_lookbehind;              /* Maximum lookbehind (characters) */
2416    BOOL nopartial;               /* Set TRUE if partial won't work */    int  top_backref;                 /* Maximum back reference */
2417    int  nltype;                  /* Newline type */    unsigned int backref_map;         /* Bitmap of low back refs */
2418    int  nllen;                   /* Newline string length */    int  assert_depth;                /* Depth of nested assertions */
2419    uschar nl[4];                 /* Newline string when fixed length */    int  external_options;            /* External (initial) options */
2420      int  external_flags;              /* External flag bits to be set */
2421      int  req_varyopt;                 /* "After variable item" flag for reqbyte */
2422      BOOL had_accept;                  /* (*ACCEPT) encountered */
2423      BOOL had_pruneorskip;             /* (*PRUNE) or (*SKIP) encountered */
2424      BOOL check_lookbehind;            /* Lookbehinds need later checking */
2425      int  nltype;                      /* Newline type */
2426      int  nllen;                       /* Newline string length */
2427      pcre_uchar nl[4];                 /* Newline string when fixed length */
2428  } compile_data;  } compile_data;
2429    
2430  /* Structure for maintaining a chain of pointers to the currently incomplete  /* Structure for maintaining a chain of pointers to the currently incomplete
2431  branches, for testing for left recursion. */  branches, for testing for left recursion while compiling. */
2432    
2433  typedef struct branch_chain {  typedef struct branch_chain {
2434    struct branch_chain *outer;    struct branch_chain *outer;
2435    uschar *current;    pcre_uchar *current_branch;
2436  } branch_chain;  } branch_chain;
2437    
2438  /* Structure for items in a linked list that represents an explicit recursive  /* Structure for items in a linked list that represents an explicit recursive
2439  call within the pattern. */  call within the pattern; used by pcre_exec(). */
2440    
2441  typedef struct recursion_info {  typedef struct recursion_info {
2442    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
2443    int group_num;                /* Number of group that was called */    unsigned int group_num;         /* Number of group that was called */
2444    const uschar *after_call;     /* "Return value": points after the call in the expr */    int *offset_save;               /* Pointer to start of saved offsets */
2445    USPTR save_start;             /* Old value of mstart */    int saved_max;                  /* Number of saved offsets */
2446    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 */  
2447  } recursion_info;  } recursion_info;
2448    
2449    /* A similar structure for pcre_dfa_exec(). */
2450    
2451    typedef struct dfa_recursion_info {
2452      struct dfa_recursion_info *prevrec;
2453      int group_num;
2454      PCRE_PUCHAR subject_position;
2455    } dfa_recursion_info;
2456    
2457  /* 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
2458  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
2459  has been matched by a subpattern - to break infinite loops. */  has been matched by a subpattern - to break infinite loops; used by
2460    pcre_exec(). */
2461    
2462  typedef struct eptrblock {  typedef struct eptrblock {
2463    struct eptrblock *epb_prev;    struct eptrblock *epb_prev;
2464    USPTR epb_saved_eptr;    PCRE_PUCHAR epb_saved_eptr;
2465  } eptrblock;  } eptrblock;
2466    
2467    
# Line 973  typedef struct match_data { Line 2472  typedef struct match_data {
2472    unsigned long int match_call_count;      /* As it says */    unsigned long int match_call_count;      /* As it says */
2473    unsigned long int match_limit;           /* As it says */    unsigned long int match_limit;           /* As it says */
2474    unsigned long int match_limit_recursion; /* As it says */    unsigned long int match_limit_recursion; /* As it says */
2475    int   *offset_vector;         /* Offset vector */    int   *offset_vector;           /* Offset vector */
2476    int    offset_end;            /* One past the end */    int    offset_end;              /* One past the end */
2477    int    offset_max;            /* The maximum usable for return data */    int    offset_max;              /* The maximum usable for return data */
2478    int    nltype;                /* Newline type */    int    nltype;                  /* Newline type */
2479    int    nllen;                 /* Newline string length */    int    nllen;                   /* Newline string length */
2480    uschar nl[4];                 /* Newline string when fixed */    int    name_count;              /* Number of names in name table */
2481    const uschar *lcc;            /* Points to lower casing table */    int    name_entry_size;         /* Size of entry in names table */
2482    const uschar *ctypes;         /* Points to table of type maps */    pcre_uchar *name_table;         /* Table of names */
2483    BOOL   offset_overflow;       /* Set if too many extractions */    pcre_uchar nl[4];               /* Newline string when fixed */
2484    BOOL   notbol;                /* NOTBOL flag */    const  pcre_uint8 *lcc;         /* Points to lower casing table */
2485    BOOL   noteol;                /* NOTEOL flag */    const  pcre_uint8 *fcc;         /* Points to case-flipping table */
2486    BOOL   utf8;                  /* UTF8 flag */    const  pcre_uint8 *ctypes;      /* Points to table of type maps */
2487    BOOL   endonly;               /* Dollar not before final \n */    BOOL   offset_overflow;         /* Set if too many extractions */
2488    BOOL   notempty;              /* Empty string match not wanted */    BOOL   notbol;                  /* NOTBOL flag */
2489    BOOL   partial;               /* PARTIAL flag */    BOOL   noteol;                  /* NOTEOL flag */
2490    BOOL   hitend;                /* Hit the end of the subject at some point */    BOOL   utf;                     /* UTF-8 / UTF-16 flag */
2491    const uschar *start_code;     /* For use when recursing */    BOOL   jscript_compat;          /* JAVASCRIPT_COMPAT flag */
2492    USPTR  start_subject;         /* Start of the subject string */    BOOL   use_ucp;                 /* PCRE_UCP flag */
2493    USPTR  end_subject;           /* End of the subject string */    BOOL   endonly;                 /* Dollar not before final \n */
2494    USPTR  start_match_ptr;       /* Start of matched string */    BOOL   notempty;                /* Empty string match not wanted */
2495    USPTR  end_match_ptr;         /* Subject position at end match */    BOOL   notempty_atstart;        /* Empty string match at start not wanted */
2496    int    end_offset_top;        /* Highwater mark at end of match */    BOOL   hitend;                  /* Hit the end of the subject at some point */
2497    int    capture_last;          /* Most recent capture number */    BOOL   bsr_anycrlf;             /* \R is just any CRLF, not full Unicode */
2498    int    start_offset;          /* The start offset value */    BOOL   hasthen;                 /* Pattern contains (*THEN) */
2499    eptrblock *eptrchain;         /* Chain of eptrblocks for tail recursions */    BOOL   ignore_skip_arg;         /* For re-run when SKIP name not found */
2500    int    eptrn;                 /* Next free eptrblock */    const  pcre_uchar *start_code;  /* For use when recursing */
2501    recursion_info *recursive;    /* Linked list of recursion data */    PCRE_PUCHAR start_subject;      /* Start of the subject string */
2502    void  *callout_data;          /* To pass back to callouts */    PCRE_PUCHAR end_subject;        /* End of the subject string */
2503      PCRE_PUCHAR start_match_ptr;    /* Start of matched string */
2504      PCRE_PUCHAR end_match_ptr;      /* Subject position at end match */
2505      PCRE_PUCHAR start_used_ptr;     /* Earliest consulted character */
2506      int    partial;                 /* PARTIAL options */
2507      int    end_offset_top;          /* Highwater mark at end of match */
2508      int    capture_last;            /* Most recent capture number */
2509      int    start_offset;            /* The start offset value */
2510      int    match_function_type;     /* Set for certain special calls of MATCH() */
2511      eptrblock *eptrchain;           /* Chain of eptrblocks for tail recursions */
2512      int    eptrn;                   /* Next free eptrblock */
2513      recursion_info *recursive;      /* Linked list of recursion data */
2514      void  *callout_data;            /* To pass back to callouts */
2515      const  pcre_uchar *mark;        /* Mark pointer to pass back on success */
2516      const  pcre_uchar *nomatch_mark;/* Mark pointer to pass back on failure */
2517      const  pcre_uchar *once_target; /* Where to back up to for atomic groups */
2518    #ifdef NO_RECURSE
2519      void  *match_frames_base;       /* For remembering malloc'd frames */
2520    #endif
2521  } match_data;  } match_data;
2522    
2523  /* 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
2524  functions. */  functions. */
2525    
2526  typedef struct dfa_match_data {  typedef struct dfa_match_data {
2527    const uschar *start_code;     /* Start of the compiled pattern */    const pcre_uchar *start_code;     /* Start of the compiled pattern */
2528    const uschar *start_subject;  /* Start of the subject string */    const pcre_uchar *start_subject ; /* Start of the subject string */
2529    const uschar *end_subject;    /* End of subject string */    const pcre_uchar *end_subject;    /* End of subject string */
2530    const uschar *tables;         /* Character tables */    const pcre_uchar *start_used_ptr; /* Earliest consulted character */
2531    int   moptions;               /* Match options */    const pcre_uint8 *tables;         /* Character tables */
2532    int   poptions;               /* Pattern options */    int   start_offset;               /* The start offset value */
2533    int    nltype;                /* Newline type */    int   moptions;                   /* Match options */
2534    int    nllen;                 /* Newline string length */    int   poptions;                   /* Pattern options */
2535    uschar nl[4];                 /* Newline string when fixed */    int   nltype;                     /* Newline type */
2536    void  *callout_data;          /* To pass back to callouts */    int   nllen;                      /* Newline string length */
2537      pcre_uchar nl[4];                 /* Newline string when fixed */
2538      void *callout_data;               /* To pass back to callouts */
2539      dfa_recursion_info *recursive;    /* Linked list of recursion data */
2540  } dfa_match_data;  } dfa_match_data;
2541    
2542  /* Bit definitions for entries in the pcre_ctypes table. */  /* Bit definitions for entries in the pcre_ctypes table. */
# Line 1025  typedef struct dfa_match_data { Line 2545  typedef struct dfa_match_data {
2545  #define ctype_letter  0x02  #define ctype_letter  0x02
2546  #define ctype_digit   0x04  #define ctype_digit   0x04
2547  #define ctype_xdigit  0x08  #define ctype_xdigit  0x08
2548  #define ctype_word    0x10   /* alphameric or '_' */  #define ctype_word    0x10   /* alphanumeric or '_' */
2549  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */
2550    
2551  /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set  /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
# Line 1052  total length. */ Line 2572  total length. */
2572  #define ctypes_offset (cbits_offset + cbit_length)  #define ctypes_offset (cbits_offset + cbit_length)
2573  #define tables_length (ctypes_offset + 256)  #define tables_length (ctypes_offset + 256)
2574    
2575    /* Internal function and data prefixes. */
2576    
2577    #if defined COMPILE_PCRE8
2578    #ifndef PUBL
2579    #define PUBL(name) pcre_##name
2580    #endif
2581    #ifndef PRIV
2582    #define PRIV(name) _pcre_##name
2583    #endif
2584    #elif defined COMPILE_PCRE16
2585    #ifndef PUBL
2586    #define PUBL(name) pcre16_##name
2587    #endif
2588    #ifndef PRIV
2589    #define PRIV(name) _pcre16_##name
2590    #endif
2591    #elif defined COMPILE_PCRE32
2592    #ifndef PUBL
2593    #define PUBL(name) pcre32_##name
2594    #endif
2595    #ifndef PRIV
2596    #define PRIV(name) _pcre32_##name
2597    #endif
2598    #else
2599    #error Unsupported compiling mode
2600    #endif /* COMPILE_PCRE[8|16|32] */
2601    
2602  /* 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
2603  codes. */  codes. Each entry used to point directly to a name, but to reduce the number of
2604    relocations in shared libraries, it now has an offset into a single string
2605    instead. */
2606    
2607  typedef struct {  typedef struct {
2608    const char *name;    pcre_uint16 name_offset;
2609    pcre_uint16 type;    pcre_uint16 type;
2610    pcre_uint16 value;    pcre_uint16 value;
2611  } ucp_type_table;  } ucp_type_table;
# Line 1067  of the exported public functions. They h Line 2616  of the exported public functions. They h
2616  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
2617  pcre_tables.c module. */  pcre_tables.c module. */
2618    
2619  extern const int    _pcre_utf8_table1[];  #ifdef COMPILE_PCRE8
2620  extern const int    _pcre_utf8_table2[];  extern const int            PRIV(utf8_table1)[];
2621  extern const int    _pcre_utf8_table3[];  extern const int            PRIV(utf8_table1_size);
2622  extern const uschar _pcre_utf8_table4[];  extern const int            PRIV(utf8_table2)[];
2623    extern const int            PRIV(utf8_table3)[];
2624  extern const int    _pcre_utf8_table1_size;  extern const pcre_uint8     PRIV(utf8_table4)[];
2625    #endif /* COMPILE_PCRE8 */
2626    
2627    extern const char           PRIV(utt_names)[];
2628    extern const ucp_type_table PRIV(utt)[];
2629    extern const int            PRIV(utt_size);
2630    
2631  extern const ucp_type_table _pcre_utt[];  extern const pcre_uint8     PRIV(OP_lengths)[];
2632  extern const int _pcre_utt_size;  extern const pcre_uint8     PRIV(default_tables)[];
2633    
2634  extern const uschar _pcre_default_tables[];  extern const pcre_uint32    PRIV(hspace_list)[];
2635    extern const pcre_uint32    PRIV(vspace_list)[];
 extern const uschar _pcre_OP_lengths[];  
2636    
2637    
2638  /* Internal shared functions. These are functions that are used by more than  /* Internal shared functions. These are functions that are used by more than
2639  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
2640  sense, but are not part of the PCRE public API. */  sense, but are not part of the PCRE public API. */
2641    
2642  extern BOOL         _pcre_is_newline(const uschar *, int, const uschar *,  /* String comparison functions. */
2643                        int *, BOOL);  #if defined COMPILE_PCRE8
2644  extern int          _pcre_ord2utf8(int, uschar *);  
2645  extern real_pcre   *_pcre_try_flipped(const real_pcre *, real_pcre *,  #define STRCMP_UC_UC(str1, str2) \
2646                        const pcre_study_data *, pcre_study_data *);    strcmp((char *)(str1), (char *)(str2))
2647  extern int          _pcre_ucp_findprop(const unsigned int, int *, int *);  #define STRCMP_UC_C8(str1, str2) \
2648  extern unsigned int _pcre_ucp_othercase(const unsigned int);    strcmp((char *)(str1), (str2))
2649  extern int          _pcre_valid_utf8(const uschar *, int);  #define STRNCMP_UC_UC(str1, str2, num) \
2650  extern BOOL         _pcre_was_newline(const uschar *, int, const uschar *,    strncmp((char *)(str1), (char *)(str2), (num))
2651                        int *, BOOL);  #define STRNCMP_UC_C8(str1, str2, num) \
2652  extern BOOL         _pcre_xclass(int, const uschar *);    strncmp((char *)(str1), (str2), (num))
2653    #define STRLEN_UC(str) strlen((const char *)str)
2654    
2655    #elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
2656    
2657    extern int               PRIV(strcmp_uc_uc)(const pcre_uchar *,
2658                               const pcre_uchar *);
2659    extern int               PRIV(strcmp_uc_c8)(const pcre_uchar *,
2660                               const char *);
2661    extern int               PRIV(strncmp_uc_uc)(const pcre_uchar *,
2662                               const pcre_uchar *, unsigned int num);
2663    extern int               PRIV(strncmp_uc_c8)(const pcre_uchar *,
2664                               const char *, unsigned int num);
2665    extern unsigned int      PRIV(strlen_uc)(const pcre_uchar *str);
2666    
2667    #define STRCMP_UC_UC(str1, str2) \
2668      PRIV(strcmp_uc_uc)((str1), (str2))
2669    #define STRCMP_UC_C8(str1, str2) \
2670      PRIV(strcmp_uc_c8)((str1), (str2))
2671    #define STRNCMP_UC_UC(str1, str2, num) \
2672      PRIV(strncmp_uc_uc)((str1), (str2), (num))
2673    #define STRNCMP_UC_C8(str1, str2, num) \
2674      PRIV(strncmp_uc_c8)((str1), (str2), (num))
2675    #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2676    
2677    #endif /* COMPILE_PCRE[8|16|32] */
2678    
2679    #if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
2680    
2681    #define STRCMP_UC_UC_TEST(str1, str2) STRCMP_UC_UC(str1, str2)
2682    #define STRCMP_UC_C8_TEST(str1, str2) STRCMP_UC_C8(str1, str2)
2683    
2684    #elif defined COMPILE_PCRE32
2685    
2686    extern int               PRIV(strcmp_uc_uc_utf)(const pcre_uchar *,
2687                               const pcre_uchar *);
2688    extern int               PRIV(strcmp_uc_c8_utf)(const pcre_uchar *,
2689                               const char *);
2690    
2691    #define STRCMP_UC_UC_TEST(str1, str2) \
2692      (utf ? PRIV(strcmp_uc_uc_utf)((str1), (str2)) : PRIV(strcmp_uc_uc)((str1), (str2)))
2693    #define STRCMP_UC_C8_TEST(str1, str2) \
2694      (utf ? PRIV(strcmp_uc_c8_utf)((str1), (str2)) : PRIV(strcmp_uc_c8)((str1), (str2)))
2695    
2696    #endif /* COMPILE_PCRE[8|16|32] */
2697    
2698    extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2699    extern BOOL              PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2700                               int *, BOOL);
2701    extern unsigned int      PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
2702    extern int               PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
2703    extern BOOL              PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2704                               int *, BOOL);
2705    extern BOOL              PRIV(xclass)(pcre_uint32, const pcre_uchar *, BOOL);
2706    
2707    #ifdef SUPPORT_JIT
2708    extern void              PRIV(jit_compile)(const REAL_PCRE *,
2709                               PUBL(extra) *, int);
2710    extern int               PRIV(jit_exec)(const PUBL(extra) *,
2711                               const pcre_uchar *, int, int, int, int *, int);
2712    extern void              PRIV(jit_free)(void *);
2713    extern int               PRIV(jit_get_size)(void *);
2714    extern const char*       PRIV(jit_get_target)(void);
2715    #endif
2716    
2717    /* Unicode character database (UCD) */
2718    
2719    typedef struct {
2720      pcre_uint8 script;     /* ucp_Arabic, etc. */
2721      pcre_uint8 chartype;   /* ucp_Cc, etc. (general categories) */
2722      pcre_uint8 gbprop;     /* ucp_gbControl, etc. (grapheme break property) */
2723      pcre_uint8 caseset;    /* offset to multichar other cases or zero */
2724      pcre_int32 other_case; /* offset to other case, or zero if none */
2725    } ucd_record;
2726    
2727    extern const pcre_uint32 PRIV(ucd_caseless_sets)[];
2728    extern const ucd_record  PRIV(ucd_records)[];
2729    extern const pcre_uint8  PRIV(ucd_stage1)[];
2730    extern const pcre_uint16 PRIV(ucd_stage2)[];
2731    extern const pcre_uint32 PRIV(ucp_gentype)[];
2732    extern const pcre_uint32 PRIV(ucp_gbtable)[];
2733    #ifdef SUPPORT_JIT
2734    extern const int         PRIV(ucp_typerange)[];
2735    #endif
2736    
2737    #ifdef SUPPORT_UCP
2738    /* UCD access macros */
2739    
2740    #define UCD_BLOCK_SIZE 128
2741    #define GET_UCD(ch) (PRIV(ucd_records) + \
2742            PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \
2743            UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE])
2744    
2745    #define UCD_CHARTYPE(ch)    GET_UCD(ch)->chartype
2746    #define UCD_SCRIPT(ch)      GET_UCD(ch)->script
2747    #define UCD_CATEGORY(ch)    PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2748    #define UCD_GRAPHBREAK(ch)  GET_UCD(ch)->gbprop
2749    #define UCD_CASESET(ch)     GET_UCD(ch)->caseset
2750    #define UCD_OTHERCASE(ch)   ((pcre_uint32)((int)ch + (int)(GET_UCD(ch)->other_case)))
2751    
2752    #endif /* SUPPORT_UCP */
2753    
2754  #endif  #endif
2755    

Legend:
Removed from v.192  
changed lines
  Added in v.1197

  ViewVC Help
Powered by ViewVC 1.1.5