/[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 211 by ph10, Thu Aug 9 09:52:43 2007 UTC revision 1567 by ph10, Fri Jun 12 16:24:58 2015 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) 1997-2014 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 */
# Line 72  setjmp and stdarg are used is when NO_RE Line 107  setjmp and stdarg are used is when NO_RE
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 132  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    #define PCRE_UINT16_MAX USHRT_MAX
200    #define PCRE_INT16_MAX SHRT_MAX
201  #elif UINT_MAX == 65535  #elif UINT_MAX == 65535
202    typedef unsigned int pcre_uint16;  typedef unsigned int pcre_uint16;
203    typedef int pcre_int16;
204    #define PCRE_UINT16_MAX UINT_MAX
205    #define PCRE_INT16_MAX INT_MAX
206  #else  #else
207    #error Cannot determine a type for 16-bit unsigned integers  #error Cannot determine a type for 16-bit integers
208  #endif  #endif
209    
210  #if UINT_MAX == 4294967295  #if UINT_MAX == 4294967295U
211    typedef unsigned int pcre_uint32;  typedef unsigned int pcre_uint32;
212  #elif ULONG_MAX == 4294967295  typedef int pcre_int32;
213    typedef unsigned long int pcre_uint32;  #define PCRE_UINT32_MAX UINT_MAX
214    #define PCRE_INT32_MAX INT_MAX
215    #elif ULONG_MAX == 4294967295UL
216    typedef unsigned long int pcre_uint32;
217    typedef long int pcre_int32;
218    #define PCRE_UINT32_MAX ULONG_MAX
219    #define PCRE_INT32_MAX LONG_MAX
220  #else  #else
221    #error Cannot determine a type for 32-bit unsigned integers  #error Cannot determine a type for 32-bit integers
222    #endif
223    
224    /* When checking for integer overflow in pcre_compile(), we need to handle
225    large integers. If a 64-bit integer type is available, we can use that.
226    Otherwise we have to cast to double, which of course requires floating point
227    arithmetic. Handle this by defining a macro for the appropriate type. If
228    stdint.h is available, include it; it may define INT64_MAX. Systems that do not
229    have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
230    by "configure". */
231    
232    #if defined HAVE_STDINT_H
233    #include <stdint.h>
234    #elif defined HAVE_INTTYPES_H
235    #include <inttypes.h>
236    #endif
237    
238    #if defined INT64_MAX || defined int64_t
239    #define INT64_OR_DOUBLE int64_t
240    #else
241    #define INT64_OR_DOUBLE double
242  #endif  #endif
243    
244  /* All character handling must be done as unsigned characters. Otherwise there  /* All character handling must be done as unsigned characters. Otherwise there
245  are problems with top-bit-set characters and functions such as isspace().  are problems with top-bit-set characters and functions such as isspace().
246  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 *,
247  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
248  to save lots of typing. I tried "uchar", but it causes problems on Digital  called pcre_uchar.
249  Unix, where it is defined in sys/types, so use "uschar" instead. */  
250    The IN_UCHARS macro multiply its argument with the byte size of the current
251    pcre_uchar type. Useful for memcpy and such operations, whose require the
252    byte size of their input/output buffers.
253    
254    The MAX_255 macro checks whether its pcre_uchar input is less than 256.
255    
256    The TABLE_GET macro is designed for accessing elements of tables whose contain
257    exactly 256 items. When the character is able to contain more than 256
258    items, some check is needed before accessing these tables.
259    */
260    
261  typedef unsigned char uschar;  #if defined COMPILE_PCRE8
262    
263    typedef unsigned char pcre_uchar;
264    #define IN_UCHARS(x) (x)
265    #define MAX_255(c) 1
266    #define TABLE_GET(c, table, default) ((table)[c])
267    
268    #elif defined COMPILE_PCRE16
269    
270    #if USHRT_MAX != 65535
271    /* This is a warning message. Change PCRE_UCHAR16 to a 16 bit data type in
272    pcre.h(.in) and disable (comment out) this message. */
273    #error Warning: PCRE_UCHAR16 is not a 16 bit data type.
274    #endif
275    
276    typedef pcre_uint16 pcre_uchar;
277    #define UCHAR_SHIFT (1)
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    #elif defined COMPILE_PCRE32
283    
284    typedef pcre_uint32 pcre_uchar;
285    #define UCHAR_SHIFT (2)
286    #define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
287    #define MAX_255(c) ((c) <= 255u)
288    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
289    
290    #else
291    #error Unsupported compiling mode
292    #endif /* COMPILE_PCRE[8|16|32] */
293    
294  /* 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
295  characters only go up to 0x7fffffff (though Unicode doesn't go beyond  characters only go up to 0x7fffffff (though Unicode doesn't go beyond
# Line 184  start/end of string field names are. */ Line 312  start/end of string field names are. */
312  #define IS_NEWLINE(p) \  #define IS_NEWLINE(p) \
313    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
314      ((p) < NLBLOCK->PSEND && \      ((p) < NLBLOCK->PSEND && \
315       _pcre_is_newline((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\       PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
316         utf8)) \         &(NLBLOCK->nllen), utf)) \
317      : \      : \
318      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
319       (p)[0] == NLBLOCK->nl[0] && \       UCHAR21TEST(p) == NLBLOCK->nl[0] && \
320       (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \       (NLBLOCK->nllen == 1 || UCHAR21TEST(p+1) == NLBLOCK->nl[1])       \
321      ) \      ) \
322    )    )
323    
# Line 198  start/end of string field names are. */ Line 326  start/end of string field names are. */
326  #define WAS_NEWLINE(p) \  #define WAS_NEWLINE(p) \
327    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
328      ((p) > NLBLOCK->PSSTART && \      ((p) > NLBLOCK->PSSTART && \
329       _pcre_was_newline((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \       PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
330         &(NLBLOCK->nllen), utf8)) \         &(NLBLOCK->nllen), utf)) \
331      : \      : \
332      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
333       (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \       UCHAR21TEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] &&              \
334       (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \       (NLBLOCK->nllen == 1 || UCHAR21TEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \
335      ) \      ) \
336    )    )
337    
# Line 217  used for the external interface and appe Line 345  used for the external interface and appe
345  must begin with PCRE_. */  must begin with PCRE_. */
346    
347  #ifdef CUSTOM_SUBJECT_PTR  #ifdef CUSTOM_SUBJECT_PTR
348  #define PCRE_SPTR CUSTOM_SUBJECT_PTR  #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
 #define USPTR CUSTOM_SUBJECT_PTR  
349  #else  #else
350  #define PCRE_SPTR const char *  #define PCRE_PUCHAR const pcre_uchar *
 #define USPTR const unsigned char *  
351  #endif  #endif
352    
   
   
353  /* Include the public PCRE header and the definitions of UCP character property  /* Include the public PCRE header and the definitions of UCP character property
354  values. */  values. */
355    
356  #include "pcre.h"  #include "pcre.h"
357  #include "ucp.h"  #include "ucp.h"
358    
359    #ifdef COMPILE_PCRE32
360    /* Assert that the public PCRE_UCHAR32 is a 32-bit type */
361    typedef int __assert_pcre_uchar32_size[sizeof(PCRE_UCHAR32) == 4 ? 1 : -1];
362    #endif
363    
364  /* When compiling for use with the Virtual Pascal compiler, these functions  /* When compiling for use with the Virtual Pascal compiler, these functions
365  need to have their names changed. PCRE must be compiled with the -DVPCOMPAT  need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
366  option on the command line. */  option on the command line. */
# Line 293  The macros are controlled by the value o Line 422  The macros are controlled by the value o
422  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
423  is automated on Unix systems via the "configure" command. */  is automated on Unix systems via the "configure" command. */
424    
425    #if defined COMPILE_PCRE8
426    
427  #if LINK_SIZE == 2  #if LINK_SIZE == 2
428    
429  #define PUT(a,n,d)   \  #define PUT(a,n,d)   \
# Line 329  is automated on Unix systems via the "co Line 460  is automated on Unix systems via the "co
460  #define GET(a,n) \  #define GET(a,n) \
461    (((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])
462    
463  #define MAX_PATTERN_SIZE (1 << 30)   /* Keep it positive */  /* Keep it positive */
464    #define MAX_PATTERN_SIZE (1 << 30)
465    
466    #else
467    #error LINK_SIZE must be either 2, 3, or 4
468    #endif
469    
470    #elif defined COMPILE_PCRE16
471    
472    #if LINK_SIZE == 2
473    
474    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
475    #undef LINK_SIZE
476    #define LINK_SIZE 1
477    
478    #define PUT(a,n,d)   \
479      (a[n] = (d))
480    
481    #define GET(a,n) \
482      (a[n])
483    
484    #define MAX_PATTERN_SIZE (1 << 16)
485    
486    #elif LINK_SIZE == 3 || LINK_SIZE == 4
487    
488    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
489    #undef LINK_SIZE
490    #define LINK_SIZE 2
491    
492    #define PUT(a,n,d)   \
493      (a[n] = (d) >> 16), \
494      (a[(n)+1] = (d) & 65535)
495    
496    #define GET(a,n) \
497      (((a)[n] << 16) | (a)[(n)+1])
498    
499    /* Keep it positive */
500    #define MAX_PATTERN_SIZE (1 << 30)
501    
502  #else  #else
503  #error LINK_SIZE must be either 2, 3, or 4  #error LINK_SIZE must be either 2, 3, or 4
504  #endif  #endif
505    
506    #elif defined COMPILE_PCRE32
507    
508    /* Only supported LINK_SIZE is 4 */
509    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
510    #undef LINK_SIZE
511    #define LINK_SIZE 1
512    
513    #define PUT(a,n,d)   \
514      (a[n] = (d))
515    
516    #define GET(a,n) \
517      (a[n])
518    
519    /* Keep it positive */
520    #define MAX_PATTERN_SIZE (1 << 30)
521    
522    #else
523    #error Unsupported compiling mode
524    #endif /* COMPILE_PCRE[8|16|32] */
525    
526  /* Convenience macro defined in terms of the others */  /* Convenience macro defined in terms of the others */
527    
# Line 346  is automated on Unix systems via the "co Line 532  is automated on Unix systems via the "co
532  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
533  capturing parenthesis numbers in back references. */  capturing parenthesis numbers in back references. */
534    
535    #if defined COMPILE_PCRE8
536    
537    #define IMM2_SIZE 2
538    
539  #define PUT2(a,n,d)   \  #define PUT2(a,n,d)   \
540    a[n] = (d) >> 8; \    a[n] = (d) >> 8; \
541    a[(n)+1] = (d) & 255    a[(n)+1] = (d) & 255
542    
543    /* For reasons that I do not understand, the expression in this GET2 macro is
544    treated by gcc as a signed expression, even when a is declared as unsigned. It
545    seems that any kind of arithmetic results in a signed value. */
546    
547  #define GET2(a,n) \  #define GET2(a,n) \
548    (((a)[n] << 8) | (a)[(n)+1])    (unsigned int)(((a)[n] << 8) | (a)[(n)+1])
549    
550    #elif defined COMPILE_PCRE16
551    
552  #define PUT2INC(a,n,d)  PUT2(a,n,d), a += 2  #define IMM2_SIZE 1
553    
554    #define PUT2(a,n,d)   \
555       a[n] = d
556    
557  /* When UTF-8 encoding is being used, a character is no longer just a single  #define GET2(a,n) \
558  byte. The macros for character handling generate simple sequences when used in     a[n]
 byte-mode, and more complicated ones for UTF-8 characters. BACKCHAR should  
 never be called in byte mode. To make sure it can never even appear when UTF-8  
 support is omitted, we don't even define it. */  
559    
560  #ifndef SUPPORT_UTF8  #elif defined COMPILE_PCRE32
561    
562    #define IMM2_SIZE 1
563    
564    #define PUT2(a,n,d)   \
565       a[n] = d
566    
567    #define GET2(a,n) \
568       a[n]
569    
570    #else
571    #error Unsupported compiling mode
572    #endif /* COMPILE_PCRE[8|16|32] */
573    
574    #define PUT2INC(a,n,d)  PUT2(a,n,d), a += IMM2_SIZE
575    
576    /* The maximum length of a MARK name is currently one data unit; it may be
577    changed in future to be a fixed number of bytes or to depend on LINK_SIZE. */
578    
579    #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
580    #define MAX_MARK ((1u << 16) - 1)
581    #else
582    #define MAX_MARK ((1u << 8) - 1)
583    #endif
584    
585    /* There is a proposed future special "UTF-21" mode, in which only the lowest
586    21 bits of a 32-bit character are interpreted as UTF, with the remaining 11
587    high-order bits available to the application for other uses. In preparation for
588    the future implementation of this mode, there are macros that load a data item
589    and, if in this special mode, mask it to 21 bits. These macros all have names
590    starting with UCHAR21. In all other modes, including the normal 32-bit
591    library, the macros all have the same simple definitions. When the new mode is
592    implemented, it is expected that these definitions will be varied appropriately
593    using #ifdef when compiling the library that supports the special mode. */
594    
595    #define UCHAR21(eptr)        (*(eptr))
596    #define UCHAR21TEST(eptr)    (*(eptr))
597    #define UCHAR21INC(eptr)     (*(eptr)++)
598    #define UCHAR21INCTEST(eptr) (*(eptr)++)
599    
600    /* When UTF encoding is being used, a character is no longer just a single
601    byte in 8-bit mode or a single short in 16-bit mode. The macros for character
602    handling generate simple sequences when used in the basic mode, and more
603    complicated ones for UTF characters. GETCHARLENTEST and other macros are not
604    used when UTF is not supported. To make sure they can never even appear when
605    UTF support is omitted, we don't even define them. */
606    
607    #ifndef SUPPORT_UTF
608    
609    /* #define MAX_VALUE_FOR_SINGLE_CHAR */
610    /* #define HAS_EXTRALEN(c) */
611    /* #define GET_EXTRALEN(c) */
612    /* #define NOT_FIRSTCHAR(c) */
613  #define GETCHAR(c, eptr) c = *eptr;  #define GETCHAR(c, eptr) c = *eptr;
614  #define GETCHARTEST(c, eptr) c = *eptr;  #define GETCHARTEST(c, eptr) c = *eptr;
615  #define GETCHARINC(c, eptr) c = *eptr++;  #define GETCHARINC(c, eptr) c = *eptr++;
616  #define GETCHARINCTEST(c, eptr) c = *eptr++;  #define GETCHARINCTEST(c, eptr) c = *eptr++;
617  #define GETCHARLEN(c, eptr, len) c = *eptr;  #define GETCHARLEN(c, eptr, len) c = *eptr;
618    /* #define GETCHARLENTEST(c, eptr, len) */
619  /* #define BACKCHAR(eptr) */  /* #define BACKCHAR(eptr) */
620    /* #define FORWARDCHAR(eptr) */
621    /* #define ACROSSCHAR(condition, eptr, action) */
622    
623  #else   /* SUPPORT_UTF8 */  #else   /* SUPPORT_UTF */
624    
625  /* 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. */  
626    
627  #define GETCHAR(c, eptr) \  #define HASUTF8EXTRALEN(c) ((c) >= 0xc0)
628    c = *eptr; \  
629    if (c >= 0xc0) \  /* Base macro to pick up the remaining bytes of a UTF-8 character, not
630    advancing the pointer. */
631    
632    #define GETUTF8(c, eptr) \
633        { \
634        if ((c & 0x20) == 0) \
635          c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
636        else if ((c & 0x10) == 0) \
637          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
638        else if ((c & 0x08) == 0) \
639          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
640          ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
641        else if ((c & 0x04) == 0) \
642          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
643              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
644              (eptr[4] & 0x3f); \
645        else \
646          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
647              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
648              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
649        }
650    
651    /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
652    the pointer. */
653    
654    #define GETUTF8INC(c, eptr) \
655      { \      { \
656      int gcii; \      if ((c & 0x20) == 0) \
657      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \        c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
658      int gcss = 6*gcaa; \      else if ((c & 0x10) == 0) \
659      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        { \
660      for (gcii = 1; gcii <= gcaa; gcii++) \        c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
661          eptr += 2; \
662          } \
663        else if ((c & 0x08) == 0) \
664          { \
665          c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
666              ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
667          eptr += 3; \
668          } \
669        else if ((c & 0x04) == 0) \
670        { \        { \
671        gcss -= 6; \        c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
672        c |= (eptr[gcii] & 0x3f) << gcss; \            ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
673              (eptr[3] & 0x3f); \
674          eptr += 4; \
675          } \
676        else \
677          { \
678          c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
679              ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
680              ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
681          eptr += 5; \
682        } \        } \
683      }      }
684    
685    #if defined COMPILE_PCRE8
686    
687    /* These macros were originally written in the form of loops that used data
688    from the tables whose names start with PRIV(utf8_table). They were rewritten by
689    a user so as not to use loops, because in some environments this gives a
690    significant performance advantage, and it seems never to do any harm. */
691    
692    /* Tells the biggest code point which can be encoded as a single character. */
693    
694    #define MAX_VALUE_FOR_SINGLE_CHAR 127
695    
696    /* Tests whether the code point needs extra characters to decode. */
697    
698    #define HAS_EXTRALEN(c) ((c) >= 0xc0)
699    
700    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
701    Otherwise it has an undefined behaviour. */
702    
703    #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
704    
705    /* Returns TRUE, if the given character is not the first character
706    of a UTF sequence. */
707    
708    #define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80)
709    
710    /* Get the next UTF-8 character, not advancing the pointer. This is called when
711    we know we are in UTF-8 mode. */
712    
713    #define GETCHAR(c, eptr) \
714      c = *eptr; \
715      if (c >= 0xc0) GETUTF8(c, eptr);
716    
717  /* 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
718  pointer. */  pointer. */
719    
720  #define GETCHARTEST(c, eptr) \  #define GETCHARTEST(c, eptr) \
721    c = *eptr; \    c = *eptr; \
722    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; \  
       } \  
     }  
723    
724  /* 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
725  know we are in UTF-8 mode. */  know we are in UTF-8 mode. */
726    
727  #define GETCHARINC(c, eptr) \  #define GETCHARINC(c, eptr) \
728    c = *eptr++; \    c = *eptr++; \
729    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; \  
       } \  
     }  
730    
731  /* 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.
732    This is called when we don't know if we are in UTF-8 mode. */
733    
734  #define GETCHARINCTEST(c, eptr) \  #define GETCHARINCTEST(c, eptr) \
735    c = *eptr++; \    c = *eptr++; \
736    if (utf8 && c >= 0xc0) \    if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
737    
738    /* Base macro to pick up the remaining bytes of a UTF-8 character, not
739    advancing the pointer, incrementing the length. */
740    
741    #define GETUTF8LEN(c, eptr, len) \
742      { \      { \
743      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \      if ((c & 0x20) == 0) \
744      int gcss = 6*gcaa; \        { \
745      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
746      while (gcaa-- > 0) \        len++; \
747          } \
748        else if ((c & 0x10)  == 0) \
749        { \        { \
750        gcss -= 6; \        c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
751        c |= (*eptr++ & 0x3f) << gcss; \        len += 2; \
752          } \
753        else if ((c & 0x08)  == 0) \
754          {\
755          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
756              ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
757          len += 3; \
758          } \
759        else if ((c & 0x04)  == 0) \
760          { \
761          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
762              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
763              (eptr[4] & 0x3f); \
764          len += 4; \
765          } \
766        else \
767          {\
768          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
769              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
770              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
771          len += 5; \
772        } \        } \
773      }      }
774    
# Line 446  if there are extra bytes. This is called Line 777  if there are extra bytes. This is called
777    
778  #define GETCHARLEN(c, eptr, len) \  #define GETCHARLEN(c, eptr, len) \
779    c = *eptr; \    c = *eptr; \
780    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
781      { \  
782      int gcii; \  /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
783      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  pointer, incrementing length if there are extra bytes. This is called when we
784      int gcss = 6*gcaa; \  do not know if we are in UTF-8 mode. */
785      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
786      for (gcii = 1; gcii <= gcaa; gcii++) \  #define GETCHARLENTEST(c, eptr, len) \
787        { \    c = *eptr; \
788        gcss -= 6; \    if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
       c |= (eptr[gcii] & 0x3f) << gcss; \  
       } \  
     len += gcaa; \  
     }  
789    
790  /* 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
791  it is. This is called only in UTF-8 mode - we don't put a test within the macro  it is. This is called only in UTF-8 mode - we don't put a test within the macro
# Line 466  because almost all calls are already wit Line 793  because almost all calls are already wit
793    
794  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
795    
796  #endif  /* Same as above, just in the other direction. */
797    #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
798    
799    /* Same as above, but it allows a fully customizable form. */
800    #define ACROSSCHAR(condition, eptr, action) \
801      while((condition) && ((eptr) & 0xc0) == 0x80) action
802    
803  /* In case there is no definition of offsetof() provided - though any proper  #elif defined COMPILE_PCRE16
 Standard C system should have one. */  
804    
805  #ifndef offsetof  /* Tells the biggest code point which can be encoded as a single character. */
806  #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))  
807  #endif  #define MAX_VALUE_FOR_SINGLE_CHAR 65535
808    
809    /* Tests whether the code point needs extra characters to decode. */
810    
811    #define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
812    
813  /* These are the public options that can change during matching. */  /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
814    Otherwise it has an undefined behaviour. */
815    
816  #define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)  #define GET_EXTRALEN(c) 1
817    
818  /* Private options flags start at the most significant end of the four bytes.  /* Returns TRUE, if the given character is not the first character
819  The public options defined in pcre.h start at the least significant end. Make  of a UTF sequence. */
 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 */  #define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00)
 #define PCRE_FIRSTSET      0x40000000  /* first_byte is set */  
 #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 */  
822    
823  /* Options for the "extra" block produced by pcre_study(). */  /* Base macro to pick up the low surrogate of a UTF-16 character, not
824    advancing the pointer. */
825    
826  #define PCRE_STUDY_MAPPED   0x01     /* a map of starting chars exists */  #define GETUTF16(c, eptr) \
827       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
828    
829    /* Get the next UTF-16 character, not advancing the pointer. This is called when
830    we know we are in UTF-16 mode. */
831    
832    #define GETCHAR(c, eptr) \
833      c = *eptr; \
834      if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
835    
836    /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
837    pointer. */
838    
839    #define GETCHARTEST(c, eptr) \
840      c = *eptr; \
841      if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
842    
843    /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
844    the pointer. */
845    
846    #define GETUTF16INC(c, eptr) \
847       { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
848    
849    /* Get the next UTF-16 character, advancing the pointer. This is called when we
850    know we are in UTF-16 mode. */
851    
852    #define GETCHARINC(c, eptr) \
853      c = *eptr++; \
854      if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
855    
856    /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
857    This is called when we don't know if we are in UTF-16 mode. */
858    
859    #define GETCHARINCTEST(c, eptr) \
860      c = *eptr++; \
861      if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
862    
863    /* Base macro to pick up the low surrogate of a UTF-16 character, not
864    advancing the pointer, incrementing the length. */
865    
866    #define GETUTF16LEN(c, eptr, len) \
867       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
868    
869    /* Get the next UTF-16 character, not advancing the pointer, incrementing
870    length if there is a low surrogate. This is called when we know we are in
871    UTF-16 mode. */
872    
873    #define GETCHARLEN(c, eptr, len) \
874      c = *eptr; \
875      if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
876    
877    /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
878    pointer, incrementing length if there is a low surrogate. This is called when
879    we do not know if we are in UTF-16 mode. */
880    
881    #define GETCHARLENTEST(c, eptr, len) \
882      c = *eptr; \
883      if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
884    
885    /* If the pointer is not at the start of a character, move it back until
886    it is. This is called only in UTF-16 mode - we don't put a test within the
887    macro because almost all calls are already within a block of UTF-16 only
888    code. */
889    
890    #define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
891    
892    /* Same as above, just in the other direction. */
893    #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
894    
895    /* Same as above, but it allows a fully customizable form. */
896    #define ACROSSCHAR(condition, eptr, action) \
897      if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
898    
899    #elif defined COMPILE_PCRE32
900    
901    /* These are trivial for the 32-bit library, since all UTF-32 characters fit
902    into one pcre_uchar unit. */
903    #define MAX_VALUE_FOR_SINGLE_CHAR (0x10ffffu)
904    #define HAS_EXTRALEN(c) (0)
905    #define GET_EXTRALEN(c) (0)
906    #define NOT_FIRSTCHAR(c) (0)
907    
908    /* Get the next UTF-32 character, not advancing the pointer. This is called when
909    we know we are in UTF-32 mode. */
910    
911    #define GETCHAR(c, eptr) \
912      c = *(eptr);
913    
914    /* Get the next UTF-32 character, testing for UTF-32 mode, and not advancing the
915    pointer. */
916    
917    #define GETCHARTEST(c, eptr) \
918      c = *(eptr);
919    
920    /* Get the next UTF-32 character, advancing the pointer. This is called when we
921    know we are in UTF-32 mode. */
922    
923    #define GETCHARINC(c, eptr) \
924      c = *((eptr)++);
925    
926    /* Get the next character, testing for UTF-32 mode, and advancing the pointer.
927    This is called when we don't know if we are in UTF-32 mode. */
928    
929    #define GETCHARINCTEST(c, eptr) \
930      c = *((eptr)++);
931    
932    /* Get the next UTF-32 character, not advancing the pointer, not incrementing
933    length (since all UTF-32 is of length 1). This is called when we know we are in
934    UTF-32 mode. */
935    
936    #define GETCHARLEN(c, eptr, len) \
937      GETCHAR(c, eptr)
938    
939    /* Get the next UTF-32character, testing for UTF-32 mode, not advancing the
940    pointer, not incrementing the length (since all UTF-32 is of length 1).
941    This is called when we do not know if we are in UTF-32 mode. */
942    
943    #define GETCHARLENTEST(c, eptr, len) \
944      GETCHARTEST(c, eptr)
945    
946    /* If the pointer is not at the start of a character, move it back until
947    it is. This is called only in UTF-32 mode - we don't put a test within the
948    macro because almost all calls are already within a block of UTF-32 only
949    code.
950    These are all no-ops since all UTF-32 characters fit into one pcre_uchar. */
951    
952    #define BACKCHAR(eptr) do { } while (0)
953    
954    /* Same as above, just in the other direction. */
955    #define FORWARDCHAR(eptr) do { } while (0)
956    
957    /* Same as above, but it allows a fully customizable form. */
958    #define ACROSSCHAR(condition, eptr, action) do { } while (0)
959    
960    #else
961    #error Unsupported compiling mode
962    #endif /* COMPILE_PCRE[8|16|32] */
963    
964    #endif  /* SUPPORT_UTF */
965    
966    /* Tests for Unicode horizontal and vertical whitespace characters must check a
967    number of different values. Using a switch statement for this generates the
968    fastest code (no loop, no memory access), and there are several places in the
969    interpreter code where this happens. In order to ensure that all the case lists
970    remain in step, we use macros so that there is only one place where the lists
971    are defined.
972    
973    These values are also required as lists in pcre_compile.c when processing \h,
974    \H, \v and \V in a character class. The lists are defined in pcre_tables.c, but
975    macros that define the values are here so that all the definitions are
976    together. The lists must be in ascending character order, terminated by
977    NOTACHAR (which is 0xffffffff).
978    
979    Any changes should ensure that the various macros are kept in step with each
980    other. NOTE: The values also appear in pcre_jit_compile.c. */
981    
982    /* ------ ASCII/Unicode environments ------ */
983    
984    #ifndef EBCDIC
985    
986    #define HSPACE_LIST \
987      CHAR_HT, CHAR_SPACE, 0xa0, \
988      0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \
989      0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \
990      NOTACHAR
991    
992    #define HSPACE_MULTIBYTE_CASES \
993      case 0x1680:  /* OGHAM SPACE MARK */ \
994      case 0x180e:  /* MONGOLIAN VOWEL SEPARATOR */ \
995      case 0x2000:  /* EN QUAD */ \
996      case 0x2001:  /* EM QUAD */ \
997      case 0x2002:  /* EN SPACE */ \
998      case 0x2003:  /* EM SPACE */ \
999      case 0x2004:  /* THREE-PER-EM SPACE */ \
1000      case 0x2005:  /* FOUR-PER-EM SPACE */ \
1001      case 0x2006:  /* SIX-PER-EM SPACE */ \
1002      case 0x2007:  /* FIGURE SPACE */ \
1003      case 0x2008:  /* PUNCTUATION SPACE */ \
1004      case 0x2009:  /* THIN SPACE */ \
1005      case 0x200A:  /* HAIR SPACE */ \
1006      case 0x202f:  /* NARROW NO-BREAK SPACE */ \
1007      case 0x205f:  /* MEDIUM MATHEMATICAL SPACE */ \
1008      case 0x3000   /* IDEOGRAPHIC SPACE */
1009    
1010    #define HSPACE_BYTE_CASES \
1011      case CHAR_HT: \
1012      case CHAR_SPACE: \
1013      case 0xa0     /* NBSP */
1014    
1015    #define HSPACE_CASES \
1016      HSPACE_BYTE_CASES: \
1017      HSPACE_MULTIBYTE_CASES
1018    
1019    #define VSPACE_LIST \
1020      CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR
1021    
1022    #define VSPACE_MULTIBYTE_CASES \
1023      case 0x2028:    /* LINE SEPARATOR */ \
1024      case 0x2029     /* PARAGRAPH SEPARATOR */
1025    
1026    #define VSPACE_BYTE_CASES \
1027      case CHAR_LF: \
1028      case CHAR_VT: \
1029      case CHAR_FF: \
1030      case CHAR_CR: \
1031      case CHAR_NEL
1032    
1033    #define VSPACE_CASES \
1034      VSPACE_BYTE_CASES: \
1035      VSPACE_MULTIBYTE_CASES
1036    
1037    /* ------ EBCDIC environments ------ */
1038    
1039    #else
1040    #define HSPACE_LIST CHAR_HT, CHAR_SPACE
1041    
1042    #define HSPACE_BYTE_CASES \
1043      case CHAR_HT: \
1044      case CHAR_SPACE
1045    
1046    #define HSPACE_CASES HSPACE_BYTE_CASES
1047    
1048    #ifdef EBCDIC_NL25
1049    #define VSPACE_LIST \
1050      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR
1051    #else
1052    #define VSPACE_LIST \
1053      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR
1054    #endif
1055    
1056    #define VSPACE_BYTE_CASES \
1057      case CHAR_LF: \
1058      case CHAR_VT: \
1059      case CHAR_FF: \
1060      case CHAR_CR: \
1061      case CHAR_NEL
1062    
1063    #define VSPACE_CASES VSPACE_BYTE_CASES
1064    #endif  /* EBCDIC */
1065    
1066    /* ------ End of whitespace macros ------ */
1067    
1068    
1069    
1070    /* Private flags containing information about the compiled regex. They used to
1071    live at the top end of the options word, but that got almost full, so they were
1072    moved to a 16-bit flags word - which got almost full, so now they are in a
1073    32-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as the
1074    restrictions on partial matching have been lifted. It remains for backwards
1075    compatibility. */
1076    
1077    #define PCRE_MODE8         0x00000001  /* compiled in 8 bit mode */
1078    #define PCRE_MODE16        0x00000002  /* compiled in 16 bit mode */
1079    #define PCRE_MODE32        0x00000004  /* compiled in 32 bit mode */
1080    #define PCRE_FIRSTSET      0x00000010  /* first_char is set */
1081    #define PCRE_FCH_CASELESS  0x00000020  /* caseless first char */
1082    #define PCRE_REQCHSET      0x00000040  /* req_byte is set */
1083    #define PCRE_RCH_CASELESS  0x00000080  /* caseless requested char */
1084    #define PCRE_STARTLINE     0x00000100  /* start after \n for multiline */
1085    #define PCRE_NOPARTIAL     0x00000200  /* can't use partial with this regex */
1086    #define PCRE_JCHANGED      0x00000400  /* j option used in regex */
1087    #define PCRE_HASCRORLF     0x00000800  /* explicit \r or \n in pattern */
1088    #define PCRE_HASTHEN       0x00001000  /* pattern contains (*THEN) */
1089    #define PCRE_MLSET         0x00002000  /* match limit set by regex */
1090    #define PCRE_RLSET         0x00004000  /* recursion limit set by regex */
1091    #define PCRE_MATCH_EMPTY   0x00008000  /* pattern can match empty string */
1092    
1093    #if defined COMPILE_PCRE8
1094    #define PCRE_MODE          PCRE_MODE8
1095    #elif defined COMPILE_PCRE16
1096    #define PCRE_MODE          PCRE_MODE16
1097    #elif defined COMPILE_PCRE32
1098    #define PCRE_MODE          PCRE_MODE32
1099    #endif
1100    #define PCRE_MODE_MASK     (PCRE_MODE8 | PCRE_MODE16 | PCRE_MODE32)
1101    
1102    /* Flags for the "extra" block produced by pcre_study(). */
1103    
1104    #define PCRE_STUDY_MAPPED  0x0001  /* a map of starting chars exists */
1105    #define PCRE_STUDY_MINLEN  0x0002  /* a minimum length field exists */
1106    
1107  /* Masks for identifying the public options that are permitted at compile  /* Masks for identifying the public options that are permitted at compile
1108  time, run time, or study time, respectively. */  time, run time, or study time, respectively. */
# Line 503  time, run time, or study time, respectiv Line 1110  time, run time, or study time, respectiv
1110  #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| \
1111                             PCRE_NEWLINE_ANYCRLF)                             PCRE_NEWLINE_ANYCRLF)
1112    
1113  #define PUBLIC_OPTIONS \  #define PUBLIC_COMPILE_OPTIONS \
1114    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
1115     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
1116     PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \     PCRE_NO_AUTO_CAPTURE|PCRE_NO_AUTO_POSSESS| \
1117     PCRE_DUPNAMES|PCRE_NEWLINE_BITS)     PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
1118       PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1119       PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE|PCRE_NEVER_UTF)
1120    
1121  #define PUBLIC_EXEC_OPTIONS \  #define PUBLIC_EXEC_OPTIONS \
1122    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1123     PCRE_PARTIAL|PCRE_NEWLINE_BITS)     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
1124       PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
1125    
1126  #define PUBLIC_DFA_EXEC_OPTIONS \  #define PUBLIC_DFA_EXEC_OPTIONS \
1127    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1128     PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_BITS)     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
1129       PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1130       PCRE_NO_START_OPTIMIZE)
1131    
1132    #define PUBLIC_STUDY_OPTIONS \
1133       (PCRE_STUDY_JIT_COMPILE|PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE| \
1134        PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE|PCRE_STUDY_EXTRA_NEEDED)
1135    
1136    #define PUBLIC_JIT_EXEC_OPTIONS \
1137       (PCRE_NO_UTF8_CHECK|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|\
1138        PCRE_NOTEMPTY_ATSTART|PCRE_PARTIAL_SOFT|PCRE_PARTIAL_HARD)
1139    
1140  #define PUBLIC_STUDY_OPTIONS 0   /* None defined */  /* Magic number to provide a small check against being handed junk. */
   
 /* Magic number to provide a small check against being handed junk. Also used  
 to detect whether a pattern was compiled on a host of different endianness. */  
1141    
1142  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */
1143    
1144  /* Negative values for the firstchar and reqchar variables */  /* This variable is used to detect a loaded regular expression
1145    in different endianness. */
1146    
1147  #define REQ_UNSET (-2)  #define REVERSED_MAGIC_NUMBER  0x45524350UL   /* 'ERCP' */
 #define REQ_NONE  (-1)  
1148    
1149  /* 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
1150  req_byte match. */  req_byte match. */
1151    
1152  #define REQ_BYTE_MAX 1000  #define REQ_BYTE_MAX 1000
1153    
1154  /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a  /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
1155  variable-length repeat, or a anything other than literal characters. */  environments where these macros are defined elsewhere. Unfortunately, there
1156    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 */  
1157    
1158  typedef int BOOL;  typedef int BOOL;
1159    
1160    #ifndef FALSE
1161  #define FALSE   0  #define FALSE   0
1162  #define TRUE    1  #define TRUE    1
1163    #endif
1164    
1165    /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
1166    character constants like '*' because the compiler would emit their EBCDIC code,
1167    which is different from their ASCII/UTF-8 code. Instead we define macros for
1168    the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
1169    is enabled. When UTF-8 support is not enabled, the definitions use character
1170    literals. Both character and string versions of each character are needed, and
1171    there are some longer strings as well.
1172    
1173    This means that, on EBCDIC platforms, the PCRE library can handle either
1174    EBCDIC, or UTF-8, but not both. To support both in the same compiled library
1175    would need different lookups depending on whether PCRE_UTF8 was set or not.
1176    This would make it impossible to use characters in switch/case statements,
1177    which would reduce performance. For a theoretical use (which nobody has asked
1178    for) in a minority area (EBCDIC platforms), this is not sensible. Any
1179    application that did need both could compile two versions of the library, using
1180    macros to give the functions distinct names. */
1181    
1182    #ifndef SUPPORT_UTF
1183    
1184    /* UTF-8 support is not enabled; use the platform-dependent character literals
1185    so that PCRE works in both ASCII and EBCDIC environments, but only in non-UTF
1186    mode. Newline characters are problematic in EBCDIC. Though it has CR and LF
1187    characters, a common practice has been to use its NL (0x15) character as the
1188    line terminator in C-like processing environments. However, sometimes the LF
1189    (0x25) character is used instead, according to this Unicode document:
1190    
1191    http://unicode.org/standard/reports/tr13/tr13-5.html
1192    
1193    PCRE defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25
1194    instead. Whichever is *not* chosen is defined as NEL.
1195    
1196    In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the
1197    same code point. */
1198    
1199    #ifdef EBCDIC
1200    
1201    #ifndef EBCDIC_NL25
1202    #define CHAR_NL                     '\x15'
1203    #define CHAR_NEL                    '\x25'
1204    #define STR_NL                      "\x15"
1205    #define STR_NEL                     "\x25"
1206    #else
1207    #define CHAR_NL                     '\x25'
1208    #define CHAR_NEL                    '\x15'
1209    #define STR_NL                      "\x25"
1210    #define STR_NEL                     "\x15"
1211    #endif
1212    
1213    #define CHAR_LF                     CHAR_NL
1214    #define STR_LF                      STR_NL
1215    
1216    #define CHAR_ESC                    '\047'
1217    #define CHAR_DEL                    '\007'
1218    #define STR_ESC                     "\047"
1219    #define STR_DEL                     "\007"
1220    
1221    #else  /* Not EBCDIC */
1222    
1223    /* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for
1224    compatibility. NEL is the Unicode newline character; make sure it is
1225    a positive value. */
1226    
1227    #define CHAR_LF                     '\n'
1228    #define CHAR_NL                     CHAR_LF
1229    #define CHAR_NEL                    ((unsigned char)'\x85')
1230    #define CHAR_ESC                    '\033'
1231    #define CHAR_DEL                    '\177'
1232    
1233    #define STR_LF                      "\n"
1234    #define STR_NL                      STR_LF
1235    #define STR_NEL                     "\x85"
1236    #define STR_ESC                     "\033"
1237    #define STR_DEL                     "\177"
1238    
1239    #endif  /* EBCDIC */
1240    
1241    /* The remaining definitions work in both environments. */
1242    
1243    #define CHAR_NULL                   '\0'
1244    #define CHAR_HT                     '\t'
1245    #define CHAR_VT                     '\v'
1246    #define CHAR_FF                     '\f'
1247    #define CHAR_CR                     '\r'
1248    #define CHAR_BS                     '\b'
1249    #define CHAR_BEL                    '\a'
1250    
1251    #define CHAR_SPACE                  ' '
1252    #define CHAR_EXCLAMATION_MARK       '!'
1253    #define CHAR_QUOTATION_MARK         '"'
1254    #define CHAR_NUMBER_SIGN            '#'
1255    #define CHAR_DOLLAR_SIGN            '$'
1256    #define CHAR_PERCENT_SIGN           '%'
1257    #define CHAR_AMPERSAND              '&'
1258    #define CHAR_APOSTROPHE             '\''
1259    #define CHAR_LEFT_PARENTHESIS       '('
1260    #define CHAR_RIGHT_PARENTHESIS      ')'
1261    #define CHAR_ASTERISK               '*'
1262    #define CHAR_PLUS                   '+'
1263    #define CHAR_COMMA                  ','
1264    #define CHAR_MINUS                  '-'
1265    #define CHAR_DOT                    '.'
1266    #define CHAR_SLASH                  '/'
1267    #define CHAR_0                      '0'
1268    #define CHAR_1                      '1'
1269    #define CHAR_2                      '2'
1270    #define CHAR_3                      '3'
1271    #define CHAR_4                      '4'
1272    #define CHAR_5                      '5'
1273    #define CHAR_6                      '6'
1274    #define CHAR_7                      '7'
1275    #define CHAR_8                      '8'
1276    #define CHAR_9                      '9'
1277    #define CHAR_COLON                  ':'
1278    #define CHAR_SEMICOLON              ';'
1279    #define CHAR_LESS_THAN_SIGN         '<'
1280    #define CHAR_EQUALS_SIGN            '='
1281    #define CHAR_GREATER_THAN_SIGN      '>'
1282    #define CHAR_QUESTION_MARK          '?'
1283    #define CHAR_COMMERCIAL_AT          '@'
1284    #define CHAR_A                      'A'
1285    #define CHAR_B                      'B'
1286    #define CHAR_C                      'C'
1287    #define CHAR_D                      'D'
1288    #define CHAR_E                      'E'
1289    #define CHAR_F                      'F'
1290    #define CHAR_G                      'G'
1291    #define CHAR_H                      'H'
1292    #define CHAR_I                      'I'
1293    #define CHAR_J                      'J'
1294    #define CHAR_K                      'K'
1295    #define CHAR_L                      'L'
1296    #define CHAR_M                      'M'
1297    #define CHAR_N                      'N'
1298    #define CHAR_O                      'O'
1299    #define CHAR_P                      'P'
1300    #define CHAR_Q                      'Q'
1301    #define CHAR_R                      'R'
1302    #define CHAR_S                      'S'
1303    #define CHAR_T                      'T'
1304    #define CHAR_U                      'U'
1305    #define CHAR_V                      'V'
1306    #define CHAR_W                      'W'
1307    #define CHAR_X                      'X'
1308    #define CHAR_Y                      'Y'
1309    #define CHAR_Z                      'Z'
1310    #define CHAR_LEFT_SQUARE_BRACKET    '['
1311    #define CHAR_BACKSLASH              '\\'
1312    #define CHAR_RIGHT_SQUARE_BRACKET   ']'
1313    #define CHAR_CIRCUMFLEX_ACCENT      '^'
1314    #define CHAR_UNDERSCORE             '_'
1315    #define CHAR_GRAVE_ACCENT           '`'
1316    #define CHAR_a                      'a'
1317    #define CHAR_b                      'b'
1318    #define CHAR_c                      'c'
1319    #define CHAR_d                      'd'
1320    #define CHAR_e                      'e'
1321    #define CHAR_f                      'f'
1322    #define CHAR_g                      'g'
1323    #define CHAR_h                      'h'
1324    #define CHAR_i                      'i'
1325    #define CHAR_j                      'j'
1326    #define CHAR_k                      'k'
1327    #define CHAR_l                      'l'
1328    #define CHAR_m                      'm'
1329    #define CHAR_n                      'n'
1330    #define CHAR_o                      'o'
1331    #define CHAR_p                      'p'
1332    #define CHAR_q                      'q'
1333    #define CHAR_r                      'r'
1334    #define CHAR_s                      's'
1335    #define CHAR_t                      't'
1336    #define CHAR_u                      'u'
1337    #define CHAR_v                      'v'
1338    #define CHAR_w                      'w'
1339    #define CHAR_x                      'x'
1340    #define CHAR_y                      'y'
1341    #define CHAR_z                      'z'
1342    #define CHAR_LEFT_CURLY_BRACKET     '{'
1343    #define CHAR_VERTICAL_LINE          '|'
1344    #define CHAR_RIGHT_CURLY_BRACKET    '}'
1345    #define CHAR_TILDE                  '~'
1346    
1347    #define STR_HT                      "\t"
1348    #define STR_VT                      "\v"
1349    #define STR_FF                      "\f"
1350    #define STR_CR                      "\r"
1351    #define STR_BS                      "\b"
1352    #define STR_BEL                     "\a"
1353    
1354    #define STR_SPACE                   " "
1355    #define STR_EXCLAMATION_MARK        "!"
1356    #define STR_QUOTATION_MARK          "\""
1357    #define STR_NUMBER_SIGN             "#"
1358    #define STR_DOLLAR_SIGN             "$"
1359    #define STR_PERCENT_SIGN            "%"
1360    #define STR_AMPERSAND               "&"
1361    #define STR_APOSTROPHE              "'"
1362    #define STR_LEFT_PARENTHESIS        "("
1363    #define STR_RIGHT_PARENTHESIS       ")"
1364    #define STR_ASTERISK                "*"
1365    #define STR_PLUS                    "+"
1366    #define STR_COMMA                   ","
1367    #define STR_MINUS                   "-"
1368    #define STR_DOT                     "."
1369    #define STR_SLASH                   "/"
1370    #define STR_0                       "0"
1371    #define STR_1                       "1"
1372    #define STR_2                       "2"
1373    #define STR_3                       "3"
1374    #define STR_4                       "4"
1375    #define STR_5                       "5"
1376    #define STR_6                       "6"
1377    #define STR_7                       "7"
1378    #define STR_8                       "8"
1379    #define STR_9                       "9"
1380    #define STR_COLON                   ":"
1381    #define STR_SEMICOLON               ";"
1382    #define STR_LESS_THAN_SIGN          "<"
1383    #define STR_EQUALS_SIGN             "="
1384    #define STR_GREATER_THAN_SIGN       ">"
1385    #define STR_QUESTION_MARK           "?"
1386    #define STR_COMMERCIAL_AT           "@"
1387    #define STR_A                       "A"
1388    #define STR_B                       "B"
1389    #define STR_C                       "C"
1390    #define STR_D                       "D"
1391    #define STR_E                       "E"
1392    #define STR_F                       "F"
1393    #define STR_G                       "G"
1394    #define STR_H                       "H"
1395    #define STR_I                       "I"
1396    #define STR_J                       "J"
1397    #define STR_K                       "K"
1398    #define STR_L                       "L"
1399    #define STR_M                       "M"
1400    #define STR_N                       "N"
1401    #define STR_O                       "O"
1402    #define STR_P                       "P"
1403    #define STR_Q                       "Q"
1404    #define STR_R                       "R"
1405    #define STR_S                       "S"
1406    #define STR_T                       "T"
1407    #define STR_U                       "U"
1408    #define STR_V                       "V"
1409    #define STR_W                       "W"
1410    #define STR_X                       "X"
1411    #define STR_Y                       "Y"
1412    #define STR_Z                       "Z"
1413    #define STR_LEFT_SQUARE_BRACKET     "["
1414    #define STR_BACKSLASH               "\\"
1415    #define STR_RIGHT_SQUARE_BRACKET    "]"
1416    #define STR_CIRCUMFLEX_ACCENT       "^"
1417    #define STR_UNDERSCORE              "_"
1418    #define STR_GRAVE_ACCENT            "`"
1419    #define STR_a                       "a"
1420    #define STR_b                       "b"
1421    #define STR_c                       "c"
1422    #define STR_d                       "d"
1423    #define STR_e                       "e"
1424    #define STR_f                       "f"
1425    #define STR_g                       "g"
1426    #define STR_h                       "h"
1427    #define STR_i                       "i"
1428    #define STR_j                       "j"
1429    #define STR_k                       "k"
1430    #define STR_l                       "l"
1431    #define STR_m                       "m"
1432    #define STR_n                       "n"
1433    #define STR_o                       "o"
1434    #define STR_p                       "p"
1435    #define STR_q                       "q"
1436    #define STR_r                       "r"
1437    #define STR_s                       "s"
1438    #define STR_t                       "t"
1439    #define STR_u                       "u"
1440    #define STR_v                       "v"
1441    #define STR_w                       "w"
1442    #define STR_x                       "x"
1443    #define STR_y                       "y"
1444    #define STR_z                       "z"
1445    #define STR_LEFT_CURLY_BRACKET      "{"
1446    #define STR_VERTICAL_LINE           "|"
1447    #define STR_RIGHT_CURLY_BRACKET     "}"
1448    #define STR_TILDE                   "~"
1449    
1450    #define STRING_ACCEPT0              "ACCEPT\0"
1451    #define STRING_COMMIT0              "COMMIT\0"
1452    #define STRING_F0                   "F\0"
1453    #define STRING_FAIL0                "FAIL\0"
1454    #define STRING_MARK0                "MARK\0"
1455    #define STRING_PRUNE0               "PRUNE\0"
1456    #define STRING_SKIP0                "SKIP\0"
1457    #define STRING_THEN                 "THEN"
1458    
1459    #define STRING_alpha0               "alpha\0"
1460    #define STRING_lower0               "lower\0"
1461    #define STRING_upper0               "upper\0"
1462    #define STRING_alnum0               "alnum\0"
1463    #define STRING_ascii0               "ascii\0"
1464    #define STRING_blank0               "blank\0"
1465    #define STRING_cntrl0               "cntrl\0"
1466    #define STRING_digit0               "digit\0"
1467    #define STRING_graph0               "graph\0"
1468    #define STRING_print0               "print\0"
1469    #define STRING_punct0               "punct\0"
1470    #define STRING_space0               "space\0"
1471    #define STRING_word0                "word\0"
1472    #define STRING_xdigit               "xdigit"
1473    
1474    #define STRING_DEFINE               "DEFINE"
1475    #define STRING_WEIRD_STARTWORD      "[:<:]]"
1476    #define STRING_WEIRD_ENDWORD        "[:>:]]"
1477    
1478    #define STRING_CR_RIGHTPAR              "CR)"
1479    #define STRING_LF_RIGHTPAR              "LF)"
1480    #define STRING_CRLF_RIGHTPAR            "CRLF)"
1481    #define STRING_ANY_RIGHTPAR             "ANY)"
1482    #define STRING_ANYCRLF_RIGHTPAR         "ANYCRLF)"
1483    #define STRING_BSR_ANYCRLF_RIGHTPAR     "BSR_ANYCRLF)"
1484    #define STRING_BSR_UNICODE_RIGHTPAR     "BSR_UNICODE)"
1485    #define STRING_UTF8_RIGHTPAR            "UTF8)"
1486    #define STRING_UTF16_RIGHTPAR           "UTF16)"
1487    #define STRING_UTF32_RIGHTPAR           "UTF32)"
1488    #define STRING_UTF_RIGHTPAR             "UTF)"
1489    #define STRING_UCP_RIGHTPAR             "UCP)"
1490    #define STRING_NO_AUTO_POSSESS_RIGHTPAR "NO_AUTO_POSSESS)"
1491    #define STRING_NO_START_OPT_RIGHTPAR    "NO_START_OPT)"
1492    #define STRING_LIMIT_MATCH_EQ           "LIMIT_MATCH="
1493    #define STRING_LIMIT_RECURSION_EQ       "LIMIT_RECURSION="
1494    
1495    #else  /* SUPPORT_UTF */
1496    
1497    /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1498    works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1499    only. */
1500    
1501    #define CHAR_HT                     '\011'
1502    #define CHAR_VT                     '\013'
1503    #define CHAR_FF                     '\014'
1504    #define CHAR_CR                     '\015'
1505    #define CHAR_LF                     '\012'
1506    #define CHAR_NL                     CHAR_LF
1507    #define CHAR_NEL                    ((unsigned char)'\x85')
1508    #define CHAR_BS                     '\010'
1509    #define CHAR_BEL                    '\007'
1510    #define CHAR_ESC                    '\033'
1511    #define CHAR_DEL                    '\177'
1512    
1513    #define CHAR_NULL                   '\0'
1514    #define CHAR_SPACE                  '\040'
1515    #define CHAR_EXCLAMATION_MARK       '\041'
1516    #define CHAR_QUOTATION_MARK         '\042'
1517    #define CHAR_NUMBER_SIGN            '\043'
1518    #define CHAR_DOLLAR_SIGN            '\044'
1519    #define CHAR_PERCENT_SIGN           '\045'
1520    #define CHAR_AMPERSAND              '\046'
1521    #define CHAR_APOSTROPHE             '\047'
1522    #define CHAR_LEFT_PARENTHESIS       '\050'
1523    #define CHAR_RIGHT_PARENTHESIS      '\051'
1524    #define CHAR_ASTERISK               '\052'
1525    #define CHAR_PLUS                   '\053'
1526    #define CHAR_COMMA                  '\054'
1527    #define CHAR_MINUS                  '\055'
1528    #define CHAR_DOT                    '\056'
1529    #define CHAR_SLASH                  '\057'
1530    #define CHAR_0                      '\060'
1531    #define CHAR_1                      '\061'
1532    #define CHAR_2                      '\062'
1533    #define CHAR_3                      '\063'
1534    #define CHAR_4                      '\064'
1535    #define CHAR_5                      '\065'
1536    #define CHAR_6                      '\066'
1537    #define CHAR_7                      '\067'
1538    #define CHAR_8                      '\070'
1539    #define CHAR_9                      '\071'
1540    #define CHAR_COLON                  '\072'
1541    #define CHAR_SEMICOLON              '\073'
1542    #define CHAR_LESS_THAN_SIGN         '\074'
1543    #define CHAR_EQUALS_SIGN            '\075'
1544    #define CHAR_GREATER_THAN_SIGN      '\076'
1545    #define CHAR_QUESTION_MARK          '\077'
1546    #define CHAR_COMMERCIAL_AT          '\100'
1547    #define CHAR_A                      '\101'
1548    #define CHAR_B                      '\102'
1549    #define CHAR_C                      '\103'
1550    #define CHAR_D                      '\104'
1551    #define CHAR_E                      '\105'
1552    #define CHAR_F                      '\106'
1553    #define CHAR_G                      '\107'
1554    #define CHAR_H                      '\110'
1555    #define CHAR_I                      '\111'
1556    #define CHAR_J                      '\112'
1557    #define CHAR_K                      '\113'
1558    #define CHAR_L                      '\114'
1559    #define CHAR_M                      '\115'
1560    #define CHAR_N                      '\116'
1561    #define CHAR_O                      '\117'
1562    #define CHAR_P                      '\120'
1563    #define CHAR_Q                      '\121'
1564    #define CHAR_R                      '\122'
1565    #define CHAR_S                      '\123'
1566    #define CHAR_T                      '\124'
1567    #define CHAR_U                      '\125'
1568    #define CHAR_V                      '\126'
1569    #define CHAR_W                      '\127'
1570    #define CHAR_X                      '\130'
1571    #define CHAR_Y                      '\131'
1572    #define CHAR_Z                      '\132'
1573    #define CHAR_LEFT_SQUARE_BRACKET    '\133'
1574    #define CHAR_BACKSLASH              '\134'
1575    #define CHAR_RIGHT_SQUARE_BRACKET   '\135'
1576    #define CHAR_CIRCUMFLEX_ACCENT      '\136'
1577    #define CHAR_UNDERSCORE             '\137'
1578    #define CHAR_GRAVE_ACCENT           '\140'
1579    #define CHAR_a                      '\141'
1580    #define CHAR_b                      '\142'
1581    #define CHAR_c                      '\143'
1582    #define CHAR_d                      '\144'
1583    #define CHAR_e                      '\145'
1584    #define CHAR_f                      '\146'
1585    #define CHAR_g                      '\147'
1586    #define CHAR_h                      '\150'
1587    #define CHAR_i                      '\151'
1588    #define CHAR_j                      '\152'
1589    #define CHAR_k                      '\153'
1590    #define CHAR_l                      '\154'
1591    #define CHAR_m                      '\155'
1592    #define CHAR_n                      '\156'
1593    #define CHAR_o                      '\157'
1594    #define CHAR_p                      '\160'
1595    #define CHAR_q                      '\161'
1596    #define CHAR_r                      '\162'
1597    #define CHAR_s                      '\163'
1598    #define CHAR_t                      '\164'
1599    #define CHAR_u                      '\165'
1600    #define CHAR_v                      '\166'
1601    #define CHAR_w                      '\167'
1602    #define CHAR_x                      '\170'
1603    #define CHAR_y                      '\171'
1604    #define CHAR_z                      '\172'
1605    #define CHAR_LEFT_CURLY_BRACKET     '\173'
1606    #define CHAR_VERTICAL_LINE          '\174'
1607    #define CHAR_RIGHT_CURLY_BRACKET    '\175'
1608    #define CHAR_TILDE                  '\176'
1609    
1610    #define STR_HT                      "\011"
1611    #define STR_VT                      "\013"
1612    #define STR_FF                      "\014"
1613    #define STR_CR                      "\015"
1614    #define STR_NL                      "\012"
1615    #define STR_BS                      "\010"
1616    #define STR_BEL                     "\007"
1617    #define STR_ESC                     "\033"
1618    #define STR_DEL                     "\177"
1619    
1620    #define STR_SPACE                   "\040"
1621    #define STR_EXCLAMATION_MARK        "\041"
1622    #define STR_QUOTATION_MARK          "\042"
1623    #define STR_NUMBER_SIGN             "\043"
1624    #define STR_DOLLAR_SIGN             "\044"
1625    #define STR_PERCENT_SIGN            "\045"
1626    #define STR_AMPERSAND               "\046"
1627    #define STR_APOSTROPHE              "\047"
1628    #define STR_LEFT_PARENTHESIS        "\050"
1629    #define STR_RIGHT_PARENTHESIS       "\051"
1630    #define STR_ASTERISK                "\052"
1631    #define STR_PLUS                    "\053"
1632    #define STR_COMMA                   "\054"
1633    #define STR_MINUS                   "\055"
1634    #define STR_DOT                     "\056"
1635    #define STR_SLASH                   "\057"
1636    #define STR_0                       "\060"
1637    #define STR_1                       "\061"
1638    #define STR_2                       "\062"
1639    #define STR_3                       "\063"
1640    #define STR_4                       "\064"
1641    #define STR_5                       "\065"
1642    #define STR_6                       "\066"
1643    #define STR_7                       "\067"
1644    #define STR_8                       "\070"
1645    #define STR_9                       "\071"
1646    #define STR_COLON                   "\072"
1647    #define STR_SEMICOLON               "\073"
1648    #define STR_LESS_THAN_SIGN          "\074"
1649    #define STR_EQUALS_SIGN             "\075"
1650    #define STR_GREATER_THAN_SIGN       "\076"
1651    #define STR_QUESTION_MARK           "\077"
1652    #define STR_COMMERCIAL_AT           "\100"
1653    #define STR_A                       "\101"
1654    #define STR_B                       "\102"
1655    #define STR_C                       "\103"
1656    #define STR_D                       "\104"
1657    #define STR_E                       "\105"
1658    #define STR_F                       "\106"
1659    #define STR_G                       "\107"
1660    #define STR_H                       "\110"
1661    #define STR_I                       "\111"
1662    #define STR_J                       "\112"
1663    #define STR_K                       "\113"
1664    #define STR_L                       "\114"
1665    #define STR_M                       "\115"
1666    #define STR_N                       "\116"
1667    #define STR_O                       "\117"
1668    #define STR_P                       "\120"
1669    #define STR_Q                       "\121"
1670    #define STR_R                       "\122"
1671    #define STR_S                       "\123"
1672    #define STR_T                       "\124"
1673    #define STR_U                       "\125"
1674    #define STR_V                       "\126"
1675    #define STR_W                       "\127"
1676    #define STR_X                       "\130"
1677    #define STR_Y                       "\131"
1678    #define STR_Z                       "\132"
1679    #define STR_LEFT_SQUARE_BRACKET     "\133"
1680    #define STR_BACKSLASH               "\134"
1681    #define STR_RIGHT_SQUARE_BRACKET    "\135"
1682    #define STR_CIRCUMFLEX_ACCENT       "\136"
1683    #define STR_UNDERSCORE              "\137"
1684    #define STR_GRAVE_ACCENT            "\140"
1685    #define STR_a                       "\141"
1686    #define STR_b                       "\142"
1687    #define STR_c                       "\143"
1688    #define STR_d                       "\144"
1689    #define STR_e                       "\145"
1690    #define STR_f                       "\146"
1691    #define STR_g                       "\147"
1692    #define STR_h                       "\150"
1693    #define STR_i                       "\151"
1694    #define STR_j                       "\152"
1695    #define STR_k                       "\153"
1696    #define STR_l                       "\154"
1697    #define STR_m                       "\155"
1698    #define STR_n                       "\156"
1699    #define STR_o                       "\157"
1700    #define STR_p                       "\160"
1701    #define STR_q                       "\161"
1702    #define STR_r                       "\162"
1703    #define STR_s                       "\163"
1704    #define STR_t                       "\164"
1705    #define STR_u                       "\165"
1706    #define STR_v                       "\166"
1707    #define STR_w                       "\167"
1708    #define STR_x                       "\170"
1709    #define STR_y                       "\171"
1710    #define STR_z                       "\172"
1711    #define STR_LEFT_CURLY_BRACKET      "\173"
1712    #define STR_VERTICAL_LINE           "\174"
1713    #define STR_RIGHT_CURLY_BRACKET     "\175"
1714    #define STR_TILDE                   "\176"
1715    
1716    #define STRING_ACCEPT0              STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1717    #define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1718    #define STRING_F0                   STR_F "\0"
1719    #define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"
1720    #define STRING_MARK0                STR_M STR_A STR_R STR_K "\0"
1721    #define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"
1722    #define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"
1723    #define STRING_THEN                 STR_T STR_H STR_E STR_N
1724    
1725    #define STRING_alpha0               STR_a STR_l STR_p STR_h STR_a "\0"
1726    #define STRING_lower0               STR_l STR_o STR_w STR_e STR_r "\0"
1727    #define STRING_upper0               STR_u STR_p STR_p STR_e STR_r "\0"
1728    #define STRING_alnum0               STR_a STR_l STR_n STR_u STR_m "\0"
1729    #define STRING_ascii0               STR_a STR_s STR_c STR_i STR_i "\0"
1730    #define STRING_blank0               STR_b STR_l STR_a STR_n STR_k "\0"
1731    #define STRING_cntrl0               STR_c STR_n STR_t STR_r STR_l "\0"
1732    #define STRING_digit0               STR_d STR_i STR_g STR_i STR_t "\0"
1733    #define STRING_graph0               STR_g STR_r STR_a STR_p STR_h "\0"
1734    #define STRING_print0               STR_p STR_r STR_i STR_n STR_t "\0"
1735    #define STRING_punct0               STR_p STR_u STR_n STR_c STR_t "\0"
1736    #define STRING_space0               STR_s STR_p STR_a STR_c STR_e "\0"
1737    #define STRING_word0                STR_w STR_o STR_r STR_d       "\0"
1738    #define STRING_xdigit               STR_x STR_d STR_i STR_g STR_i STR_t
1739    
1740    #define STRING_DEFINE               STR_D STR_E STR_F STR_I STR_N STR_E
1741    #define STRING_WEIRD_STARTWORD      STR_LEFT_SQUARE_BRACKET STR_COLON STR_LESS_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET
1742    #define STRING_WEIRD_ENDWORD        STR_LEFT_SQUARE_BRACKET STR_COLON STR_GREATER_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET
1743    
1744    #define STRING_CR_RIGHTPAR              STR_C STR_R STR_RIGHT_PARENTHESIS
1745    #define STRING_LF_RIGHTPAR              STR_L STR_F STR_RIGHT_PARENTHESIS
1746    #define STRING_CRLF_RIGHTPAR            STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1747    #define STRING_ANY_RIGHTPAR             STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1748    #define STRING_ANYCRLF_RIGHTPAR         STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1749    #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
1750    #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
1751    #define STRING_UTF8_RIGHTPAR            STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1752    #define STRING_UTF16_RIGHTPAR           STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
1753    #define STRING_UTF32_RIGHTPAR           STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS
1754    #define STRING_UTF_RIGHTPAR             STR_U STR_T STR_F STR_RIGHT_PARENTHESIS
1755    #define STRING_UCP_RIGHTPAR             STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1756    #define STRING_NO_AUTO_POSSESS_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_A STR_U STR_T STR_O STR_UNDERSCORE STR_P STR_O STR_S STR_S STR_E STR_S STR_S STR_RIGHT_PARENTHESIS
1757    #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
1758    #define STRING_LIMIT_MATCH_EQ           STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_M STR_A STR_T STR_C STR_H STR_EQUALS_SIGN
1759    #define STRING_LIMIT_RECURSION_EQ       STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_R STR_E STR_C STR_U STR_R STR_S STR_I STR_O STR_N STR_EQUALS_SIGN
1760    
1761    #endif  /* SUPPORT_UTF */
1762    
1763  /* Escape items that are just an encoding of a particular data value. */  /* Escape items that are just an encoding of a particular data value. */
1764    
1765    #ifndef ESC_a
1766    #define ESC_a CHAR_BEL
1767    #endif
1768    
1769  #ifndef ESC_e  #ifndef ESC_e
1770  #define ESC_e 27  #define ESC_e CHAR_ESC
1771  #endif  #endif
1772    
1773  #ifndef ESC_f  #ifndef ESC_f
1774  #define ESC_f '\f'  #define ESC_f CHAR_FF
1775  #endif  #endif
1776    
1777  #ifndef ESC_n  #ifndef ESC_n
1778  #define ESC_n '\n'  #define ESC_n CHAR_LF
1779  #endif  #endif
1780    
1781  #ifndef ESC_r  #ifndef ESC_r
1782  #define ESC_r '\r'  #define ESC_r CHAR_CR
1783  #endif  #endif
1784    
1785  /* 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
1786  (presumably because of all the others like size_t). */  (presumably because of all the others like size_t). */
1787    
1788  #ifndef ESC_tee  #ifndef ESC_tee
1789  #define ESC_tee '\t'  #define ESC_tee CHAR_HT
1790  #endif  #endif
1791    
1792  /* Codes for different types of Unicode property */  /* Codes for different types of Unicode property */
1793    
1794  #define PT_ANY        0    /* Any property - matches all chars */  #define PT_ANY        0    /* Any property - matches all chars */
1795  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */
1796  #define PT_GC         2    /* General characteristic (e.g. L) */  #define PT_GC         2    /* Specified general characteristic (e.g. L) */
1797  #define PT_PC         3    /* Particular characteristic (e.g. Lu) */  #define PT_PC         3    /* Specified particular characteristic (e.g. Lu) */
1798  #define PT_SC         4    /* Script (e.g. Han) */  #define PT_SC         4    /* Script (e.g. Han) */
1799    #define PT_ALNUM      5    /* Alphanumeric - the union of L and N */
1800    #define PT_SPACE      6    /* Perl space - Z plus 9,10,12,13 */
1801    #define PT_PXSPACE    7    /* POSIX space - Z plus 9,10,11,12,13 */
1802    #define PT_WORD       8    /* Word - L plus N plus underscore */
1803    #define PT_CLIST      9    /* Pseudo-property: match character list */
1804    #define PT_UCNC      10    /* Universal Character nameable character */
1805    #define PT_TABSIZE   11    /* Size of square table for autopossessify tests */
1806    
1807    /* The following special properties are used only in XCLASS items, when POSIX
1808    classes are specified and PCRE_UCP is set - in other words, for Unicode
1809    handling of these classes. They are not available via the \p or \P escapes like
1810    those in the above list, and so they do not take part in the autopossessifying
1811    table. */
1812    
1813    #define PT_PXGRAPH   11    /* [:graph:] - characters that mark the paper */
1814    #define PT_PXPRINT   12    /* [:print:] - [:graph:] plus non-control spaces */
1815    #define PT_PXPUNCT   13    /* [:punct:] - punctuation characters */
1816    
1817  /* 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
1818  contain UTF-8 characters with values greater than 255. */  contain characters with values greater than 255. */
1819    
1820  #define XCL_NOT    0x01    /* Flag: this is a negative class */  #define XCL_NOT       0x01    /* Flag: this is a negative class */
1821  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */  #define XCL_MAP       0x02    /* Flag: a 32-byte map is present */
1822    #define XCL_HASPROP   0x04    /* Flag: property checks are present. */
1823    
1824  #define XCL_END       0    /* Marks end of individual items */  #define XCL_END       0    /* Marks end of individual items */
1825  #define XCL_SINGLE    1    /* Single item (one multibyte char) follows */  #define XCL_SINGLE    1    /* Single item (one multibyte char) follows */
# Line 593  contain UTF-8 characters with values gre Line 1828  contain UTF-8 characters with values gre
1828  #define XCL_NOTPROP   4    /* Unicode inverted property (ditto) */  #define XCL_NOTPROP   4    /* Unicode inverted property (ditto) */
1829    
1830  /* 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
1831  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 0
1832  their negation. Also, they must appear in the same order as in the opcode  for a data character.  Also, they must appear in the same order as in the
1833  definitions below, up to ESC_z. There's a dummy for OP_ANY because it  opcode definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
1834  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
1835  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
1836  There are two tests in the code for an escape greater than ESC_b and less than  non-DOTALL mode, "." behaves like \N.
1837  ESC_Z to detect the types that may be repeated. These are the types that  
1838  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.
1839  character, that code will have to change. */  when PCRE_UCP is set and replacement of \d etc by \p sequences is required.
1840    They must be contiguous, and remain in order so that the replacements can be
1841    looked up from a table.
1842    
1843    Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in
1844    check_escape(). There are two tests in the code for an escape
1845    greater than ESC_b and less than ESC_Z to detect the types that may be
1846    repeated. These are the types that consume characters. If any new escapes are
1847    put in between that don't consume a character, that code will have to change.
1848    */
1849    
1850  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,
1851         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,
1852         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,
1853           ESC_E, ESC_Q, ESC_g, ESC_k,
1854           ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu };
1855    
1856    
1857    /********************** Opcode definitions ******************/
1858    
1859    /****** NOTE NOTE NOTE ******
1860    
1861    Starting from 1 (i.e. after OP_END), the values up to OP_EOD must correspond in
1862    order to the list of escapes immediately above. Furthermore, values up to
1863    OP_DOLLM must not be changed without adjusting the table called autoposstab in
1864    pcre_compile.c
1865    
1866    Whenever this list is updated, the two macro definitions that follow must be
1867    updated to match. The possessification table called "opcode_possessify" in
1868    pcre_compile.c must also be updated, and also the tables called "coptable"
1869    and "poptable" in pcre_dfa_exec.c.
1870    
1871    ****** NOTE NOTE NOTE ******/
1872    
 /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to  
 OP_EOD must correspond in order to the list of escapes immediately above.  
1873    
1874  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions  /* The values between FIRST_AUTOTAB_OP and LAST_AUTOTAB_RIGHT_OP, inclusive,
1875  that follow must also be updated to match. There is also a table called  are used in a table for deciding whether a repeated character type can be
1876  "coptable" in pcre_dfa_exec.c that must be updated. */  auto-possessified. */
1877    
1878    #define FIRST_AUTOTAB_OP       OP_NOT_DIGIT
1879    #define LAST_AUTOTAB_LEFT_OP   OP_EXTUNI
1880    #define LAST_AUTOTAB_RIGHT_OP  OP_DOLLM
1881    
1882  enum {  enum {
1883    OP_END,            /* 0 End of pattern */    OP_END,            /* 0 End of pattern */
# Line 631  enum { Line 1895  enum {
1895    OP_WHITESPACE,         /*  9 \s */    OP_WHITESPACE,         /*  9 \s */
1896    OP_NOT_WORDCHAR,       /* 10 \W */    OP_NOT_WORDCHAR,       /* 10 \W */
1897    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 */  
1898    
1899    /* These three must follow the previous three, in the same order. There's a    OP_ANY,            /* 12 Match any character except newline (\N) */
1900    check for >= SBRA to distinguish the two sets. */    OP_ALLANY,         /* 13 Match any character */
1901      OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1902      OP_NOTPROP,        /* 15 \P (not Unicode property) */
1903      OP_PROP,           /* 16 \p (Unicode property) */
1904      OP_ANYNL,          /* 17 \R (any newline sequence) */
1905      OP_NOT_HSPACE,     /* 18 \H (not horizontal whitespace) */
1906      OP_HSPACE,         /* 19 \h (horizontal whitespace) */
1907      OP_NOT_VSPACE,     /* 20 \V (not vertical whitespace) */
1908      OP_VSPACE,         /* 21 \v (vertical whitespace) */
1909      OP_EXTUNI,         /* 22 \X (extended Unicode sequence */
1910      OP_EODN,           /* 23 End of data or \n at end of data (\Z) */
1911      OP_EOD,            /* 24 End of data (\z) */
1912    
1913      /* Line end assertions */
1914    
1915      OP_DOLL,           /* 25 End of line - not multiline */
1916      OP_DOLLM,          /* 26 End of line - multiline */
1917      OP_CIRC,           /* 27 Start of line - not multiline */
1918      OP_CIRCM,          /* 28 Start of line - multiline */
1919    
1920      /* Single characters; caseful must precede the caseless ones */
1921    
1922      OP_CHAR,           /* 29 Match one character, casefully */
1923      OP_CHARI,          /* 30 Match one character, caselessly */
1924      OP_NOT,            /* 31 Match one character, not the given one, casefully */
1925      OP_NOTI,           /* 32 Match one character, not the given one, caselessly */
1926    
1927      /* The following sets of 13 opcodes must always be kept in step because
1928      the offset from the first one is used to generate the others. */
1929    
1930      /* Repeated characters; caseful must precede the caseless ones */
1931    
1932      OP_STAR,           /* 33 The maximizing and minimizing versions of */
1933      OP_MINSTAR,        /* 34 these six opcodes must come in pairs, with */
1934      OP_PLUS,           /* 35 the minimizing one second. */
1935      OP_MINPLUS,        /* 36 */
1936      OP_QUERY,          /* 37 */
1937      OP_MINQUERY,       /* 38 */
1938    
1939      OP_UPTO,           /* 39 From 0 to n matches of one character, caseful*/
1940      OP_MINUPTO,        /* 40 */
1941      OP_EXACT,          /* 41 Exactly n matches */
1942    
1943      OP_POSSTAR,        /* 42 Possessified star, caseful */
1944      OP_POSPLUS,        /* 43 Possessified plus, caseful */
1945      OP_POSQUERY,       /* 44 Posesssified query, caseful */
1946      OP_POSUPTO,        /* 45 Possessified upto, caseful */
1947    
1948      /* Repeated characters; caseless must follow the caseful ones */
1949    
1950      OP_STARI,          /* 46 */
1951      OP_MINSTARI,       /* 47 */
1952      OP_PLUSI,          /* 48 */
1953      OP_MINPLUSI,       /* 49 */
1954      OP_QUERYI,         /* 50 */
1955      OP_MINQUERYI,      /* 51 */
1956    
1957      OP_UPTOI,          /* 52 From 0 to n matches of one character, caseless */
1958      OP_MINUPTOI,       /* 53 */
1959      OP_EXACTI,         /* 54 */
1960    
1961      OP_POSSTARI,       /* 55 Possessified star, caseless */
1962      OP_POSPLUSI,       /* 56 Possessified plus, caseless */
1963      OP_POSQUERYI,      /* 57 Posesssified query, caseless */
1964      OP_POSUPTOI,       /* 58 Possessified upto, caseless */
1965    
1966      /* The negated ones must follow the non-negated ones, and match them */
1967      /* Negated repeated character, caseful; must precede the caseless ones */
1968    
1969      OP_NOTSTAR,        /* 59 The maximizing and minimizing versions of */
1970      OP_NOTMINSTAR,     /* 60 these six opcodes must come in pairs, with */
1971      OP_NOTPLUS,        /* 61 the minimizing one second. They must be in */
1972      OP_NOTMINPLUS,     /* 62 exactly the same order as those above. */
1973      OP_NOTQUERY,       /* 63 */
1974      OP_NOTMINQUERY,    /* 64 */
1975    
1976      OP_NOTUPTO,        /* 65 From 0 to n matches, caseful */
1977      OP_NOTMINUPTO,     /* 66 */
1978      OP_NOTEXACT,       /* 67 Exactly n matches */
1979    
1980      OP_NOTPOSSTAR,     /* 68 Possessified versions, caseful */
1981      OP_NOTPOSPLUS,     /* 69 */
1982      OP_NOTPOSQUERY,    /* 70 */
1983      OP_NOTPOSUPTO,     /* 71 */
1984    
1985      /* Negated repeated character, caseless; must follow the caseful ones */
1986    
1987      OP_NOTSTARI,       /* 72 */
1988      OP_NOTMINSTARI,    /* 73 */
1989      OP_NOTPLUSI,       /* 74 */
1990      OP_NOTMINPLUSI,    /* 75 */
1991      OP_NOTQUERYI,      /* 76 */
1992      OP_NOTMINQUERYI,   /* 77 */
1993    
1994      OP_NOTUPTOI,       /* 78 From 0 to n matches, caseless */
1995      OP_NOTMINUPTOI,    /* 79 */
1996      OP_NOTEXACTI,      /* 80 Exactly n matches */
1997    
1998      OP_NOTPOSSTARI,    /* 81 Possessified versions, caseless */
1999      OP_NOTPOSPLUSI,    /* 82 */
2000      OP_NOTPOSQUERYI,   /* 83 */
2001      OP_NOTPOSUPTOI,    /* 84 */
2002    
2003      /* Character types */
2004    
2005      OP_TYPESTAR,       /* 85 The maximizing and minimizing versions of */
2006      OP_TYPEMINSTAR,    /* 86 these six opcodes must come in pairs, with */
2007      OP_TYPEPLUS,       /* 87 the minimizing one second. These codes must */
2008      OP_TYPEMINPLUS,    /* 88 be in exactly the same order as those above. */
2009      OP_TYPEQUERY,      /* 89 */
2010      OP_TYPEMINQUERY,   /* 90 */
2011    
2012      OP_TYPEUPTO,       /* 91 From 0 to n matches */
2013      OP_TYPEMINUPTO,    /* 92 */
2014      OP_TYPEEXACT,      /* 93 Exactly n matches */
2015    
2016      OP_TYPEPOSSTAR,    /* 94 Possessified versions */
2017      OP_TYPEPOSPLUS,    /* 95 */
2018      OP_TYPEPOSQUERY,   /* 96 */
2019      OP_TYPEPOSUPTO,    /* 97 */
2020    
2021      /* These are used for character classes and back references; only the
2022      first six are the same as the sets above. */
2023    
2024      OP_CRSTAR,         /* 98 The maximizing and minimizing versions of */
2025      OP_CRMINSTAR,      /* 99 all these opcodes must come in pairs, with */
2026      OP_CRPLUS,         /* 100 the minimizing one second. These codes must */
2027      OP_CRMINPLUS,      /* 101 be in exactly the same order as those above. */
2028      OP_CRQUERY,        /* 102 */
2029      OP_CRMINQUERY,     /* 103 */
2030    
2031      OP_CRRANGE,        /* 104 These are different to the three sets above. */
2032      OP_CRMINRANGE,     /* 105 */
2033    
2034      OP_CRPOSSTAR,      /* 106 Possessified versions */
2035      OP_CRPOSPLUS,      /* 107 */
2036      OP_CRPOSQUERY,     /* 108 */
2037      OP_CRPOSRANGE,     /* 109 */
2038    
2039      /* End of quantifier opcodes */
2040    
2041      OP_CLASS,          /* 110 Match a character class, chars < 256 only */
2042      OP_NCLASS,         /* 111 Same, but the bitmap was created from a negative
2043                                  class - the difference is relevant only when a
2044                                  character > 255 is encountered. */
2045      OP_XCLASS,         /* 112 Extended class for handling > 255 chars within the
2046                                  class. This does both positive and negative. */
2047      OP_REF,            /* 113 Match a back reference, casefully */
2048      OP_REFI,           /* 114 Match a back reference, caselessly */
2049      OP_DNREF,          /* 115 Match a duplicate name backref, casefully */
2050      OP_DNREFI,         /* 116 Match a duplicate name backref, caselessly */
2051      OP_RECURSE,        /* 117 Match a numbered subpattern (possibly recursive) */
2052      OP_CALLOUT,        /* 118 Call out to external function if provided */
2053    
2054      OP_ALT,            /* 119 Start of alternation */
2055      OP_KET,            /* 120 End of group that doesn't have an unbounded repeat */
2056      OP_KETRMAX,        /* 121 These two must remain together and in this */
2057      OP_KETRMIN,        /* 122 order. They are for groups the repeat for ever. */
2058      OP_KETRPOS,        /* 123 Possessive unlimited repeat. */
2059    
2060      /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
2061      asserts must remain in order. */
2062    
2063      OP_REVERSE,        /* 124 Move pointer back - used in lookbehind assertions */
2064      OP_ASSERT,         /* 125 Positive lookahead */
2065      OP_ASSERT_NOT,     /* 126 Negative lookahead */
2066      OP_ASSERTBACK,     /* 127 Positive lookbehind */
2067      OP_ASSERTBACK_NOT, /* 128 Negative lookbehind */
2068    
2069      /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
2070      after the assertions, with ONCE first, as there's a test for >= ONCE for a
2071      subpattern that isn't an assertion. The POS versions must immediately follow
2072      the non-POS versions in each case. */
2073    
2074      OP_ONCE,           /* 129 Atomic group, contains captures */
2075      OP_ONCE_NC,        /* 130 Atomic group containing no captures */
2076      OP_BRA,            /* 131 Start of non-capturing bracket */
2077      OP_BRAPOS,         /* 132 Ditto, with unlimited, possessive repeat */
2078      OP_CBRA,           /* 133 Start of capturing bracket */
2079      OP_CBRAPOS,        /* 134 Ditto, with unlimited, possessive repeat */
2080      OP_COND,           /* 135 Conditional group */
2081    
2082    OP_SBRA,           /* 96 Start of non-capturing bracket, check empty  */    /* These five must follow the previous five, in the same order. There's a
2083    OP_SCBRA,          /* 97 Start of capturing bracket, check empty */    check for >= SBRA to distinguish the two sets. */
   OP_SCOND,          /* 98 Conditional group, check empty */  
   
   OP_CREF,           /* 99 Used to hold a capture number as condition */  
   OP_RREF,           /* 100 Used to hold a recursion number as condition */  
   OP_DEF,            /* 101 The DEFINE condition */  
2084    
2085    OP_BRAZERO,        /* 102 These two must remain together and in this */    OP_SBRA,           /* 136 Start of non-capturing bracket, check empty  */
2086    OP_BRAMINZERO,     /* 103 order. */    OP_SBRAPOS,        /* 137 Ditto, with unlimited, possessive repeat */
2087      OP_SCBRA,          /* 138 Start of capturing bracket, check empty */
2088      OP_SCBRAPOS,       /* 139 Ditto, with unlimited, possessive repeat */
2089      OP_SCOND,          /* 140 Conditional group, check empty */
2090    
2091      /* The next two pairs must (respectively) be kept together. */
2092    
2093      OP_CREF,           /* 141 Used to hold a capture number as condition */
2094      OP_DNCREF,         /* 142 Used to point to duplicate names as a condition */
2095      OP_RREF,           /* 143 Used to hold a recursion number as condition */
2096      OP_DNRREF,         /* 144 Used to point to duplicate names as a condition */
2097      OP_DEF,            /* 145 The DEFINE condition */
2098    
2099      OP_BRAZERO,        /* 146 These two must remain together and in this */
2100      OP_BRAMINZERO,     /* 147 order. */
2101      OP_BRAPOSZERO,     /* 148 */
2102    
2103    /* These are backtracking control verbs */    /* These are backtracking control verbs */
2104    
2105    OP_PRUNE,          /* 104 */    OP_MARK,           /* 149 always has an argument */
2106    OP_SKIP,           /* 105 */    OP_PRUNE,          /* 150 */
2107    OP_THEN,           /* 106 */    OP_PRUNE_ARG,      /* 151 same, but with argument */
2108    OP_COMMIT,         /* 107 */    OP_SKIP,           /* 152 */
2109      OP_SKIP_ARG,       /* 153 same, but with argument */
2110      OP_THEN,           /* 154 */
2111      OP_THEN_ARG,       /* 155 same, but with argument */
2112      OP_COMMIT,         /* 156 */
2113    
2114    /* These are forced failure and success verbs */    /* These are forced failure and success verbs */
2115    
2116    OP_FAIL,           /* 108 */    OP_FAIL,           /* 157 */
2117    OP_ACCEPT          /* 109 */    OP_ACCEPT,         /* 158 */
2118      OP_ASSERT_ACCEPT,  /* 159 Used inside assertions */
2119      OP_CLOSE,          /* 160 Used before OP_ACCEPT to close open captures */
2120    
2121      /* This is used to skip a subpattern with a {0} quantifier */
2122    
2123      OP_SKIPZERO,       /* 161 */
2124    
2125      /* This is not an opcode, but is used to check that tables indexed by opcode
2126      are the correct length, in order to catch updating errors - there have been
2127      some in the past. */
2128    
2129      OP_TABLE_LENGTH
2130  };  };
2131    
2132    /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
2133    definitions that follow must also be updated to match. There are also tables
2134    called "opcode_possessify" in pcre_compile.c and "coptable" and "poptable" in
2135    pcre_dfa_exec.c that must be updated. */
2136    
2137    
2138  /* 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
2139  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
2140    only in pcre_printint.c, which fills out the full names in many cases (and in
2141    some cases doesn't actually use these names at all). */
2142    
2143  #define OP_NAME_LIST \  #define OP_NAME_LIST \
2144    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \
2145    "\\S", "\\s", "\\W", "\\w", "Any", "Anybyte",                   \    "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte",         \
2146    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \
2147    "extuni",  "\\Z", "\\z",                                        \    "extuni",  "\\Z", "\\z",                                        \
2148    "Opt", "^", "$", "char", "charnc", "not",                       \    "$", "$", "^", "^", "char", "chari", "not", "noti",             \
2149    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2150      "{", "{", "{",                                                  \
2151    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2152    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2153      "{", "{", "{",                                                  \
2154      "*+","++", "?+", "{",                                           \
2155      "*", "*?", "+", "+?", "?", "??",                                \
2156      "{", "{", "{",                                                  \
2157      "*+","++", "?+", "{",                                           \
2158      "*", "*?", "+", "+?", "?", "??",                                \
2159      "{", "{", "{",                                                  \
2160    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2161    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \
2162    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2163    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \
2164    "class", "nclass", "xclass", "Ref", "Recurse", "Callout",       \    "*+","++", "?+", "{",                                           \
2165    "Alt", "Ket", "KetRmax", "KetRmin", "Assert", "Assert not",     \    "class", "nclass", "xclass", "Ref", "Refi", "DnRef", "DnRefi",  \
2166    "AssertB", "AssertB not", "Reverse",                            \    "Recurse", "Callout",                                           \
2167    "Once", "Bra", "CBra", "Cond", "SBra", "SCBra", "SCond",        \    "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos",                  \
2168    "Cond ref", "Cond rec", "Cond def", "Brazero", "Braminzero",    \    "Reverse", "Assert", "Assert not", "AssertB", "AssertB not",    \
2169    "*PRUNE", "*SKIP", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT"    "Once", "Once_NC",                                              \
2170      "Bra", "BraPos", "CBra", "CBraPos",                             \
2171      "Cond",                                                         \
2172      "SBra", "SBraPos", "SCBra", "SCBraPos",                         \
2173      "SCond",                                                        \
2174      "Cond ref", "Cond dnref", "Cond rec", "Cond dnrec", "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 806  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, 1, 1, 1+2*IMM2_SIZE,        /* Possessive *+, ++, ?+, CRPOSRANGE      */ \
2231   33,                             /* NCLASS                                 */ \    1+(32/sizeof(pcre_uchar)),     /* CLASS                                  */ \
2232      1+(32/sizeof(pcre_uchar)),     /* NCLASS                                 */ \
2233    0,                             /* XCLASS - variable length               */ \    0,                             /* XCLASS - variable length               */ \
2234    3,                             /* REF                                    */ \    1+IMM2_SIZE,                   /* REF                                    */ \
2235      1+IMM2_SIZE,                   /* REFI                                   */ \
2236      1+2*IMM2_SIZE,                 /* DNREF                                  */ \
2237      1+2*IMM2_SIZE,                 /* DNREFI                                 */ \
2238    1+LINK_SIZE,                   /* RECURSE                                */ \    1+LINK_SIZE,                   /* RECURSE                                */ \
2239    2+2*LINK_SIZE,                 /* CALLOUT                                */ \    2+2*LINK_SIZE,                 /* CALLOUT                                */ \
2240    1+LINK_SIZE,                   /* Alt                                    */ \    1+LINK_SIZE,                   /* Alt                                    */ \
2241    1+LINK_SIZE,                   /* Ket                                    */ \    1+LINK_SIZE,                   /* Ket                                    */ \
2242    1+LINK_SIZE,                   /* KetRmax                                */ \    1+LINK_SIZE,                   /* KetRmax                                */ \
2243    1+LINK_SIZE,                   /* KetRmin                                */ \    1+LINK_SIZE,                   /* KetRmin                                */ \
2244      1+LINK_SIZE,                   /* KetRpos                                */ \
2245      1+LINK_SIZE,                   /* Reverse                                */ \
2246    1+LINK_SIZE,                   /* Assert                                 */ \    1+LINK_SIZE,                   /* Assert                                 */ \
2247    1+LINK_SIZE,                   /* Assert not                             */ \    1+LINK_SIZE,                   /* Assert not                             */ \
2248    1+LINK_SIZE,                   /* Assert behind                          */ \    1+LINK_SIZE,                   /* Assert behind                          */ \
2249    1+LINK_SIZE,                   /* Assert behind not                      */ \    1+LINK_SIZE,                   /* Assert behind not                      */ \
   1+LINK_SIZE,                   /* Reverse                                */ \  
2250    1+LINK_SIZE,                   /* ONCE                                   */ \    1+LINK_SIZE,                   /* ONCE                                   */ \
2251      1+LINK_SIZE,                   /* ONCE_NC                                */ \
2252    1+LINK_SIZE,                   /* BRA                                    */ \    1+LINK_SIZE,                   /* BRA                                    */ \
2253    3+LINK_SIZE,                   /* CBRA                                   */ \    1+LINK_SIZE,                   /* BRAPOS                                 */ \
2254      1+LINK_SIZE+IMM2_SIZE,         /* CBRA                                   */ \
2255      1+LINK_SIZE+IMM2_SIZE,         /* CBRAPOS                                */ \
2256    1+LINK_SIZE,                   /* COND                                   */ \    1+LINK_SIZE,                   /* COND                                   */ \
2257    1+LINK_SIZE,                   /* SBRA                                   */ \    1+LINK_SIZE,                   /* SBRA                                   */ \
2258    3+LINK_SIZE,                   /* SCBRA                                  */ \    1+LINK_SIZE,                   /* SBRAPOS                                */ \
2259      1+LINK_SIZE+IMM2_SIZE,         /* SCBRA                                  */ \
2260      1+LINK_SIZE+IMM2_SIZE,         /* SCBRAPOS                               */ \
2261    1+LINK_SIZE,                   /* SCOND                                  */ \    1+LINK_SIZE,                   /* SCOND                                  */ \
2262    3,                             /* CREF                                   */ \    1+IMM2_SIZE, 1+2*IMM2_SIZE,    /* CREF, DNCREF                           */ \
2263    3,                             /* RREF                                   */ \    1+IMM2_SIZE, 1+2*IMM2_SIZE,    /* RREF, DNRREF                           */ \
2264    1,                             /* DEF                                    */ \    1,                             /* DEF                                    */ \
2265    1, 1,                          /* BRAZERO, BRAMINZERO                    */ \    1, 1, 1,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */ \
2266    1, 1, 1, 1,                    /* PRUNE, SKIP, THEN, COMMIT,             */ \    3, 1, 3,                       /* MARK, PRUNE, PRUNE_ARG                 */ \
2267    1, 1                           /* FAIL, ACCEPT                           */    1, 3,                          /* SKIP, SKIP_ARG                         */ \
2268      1, 3,                          /* THEN, THEN_ARG                         */ \
2269      1, 1, 1, 1,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */ \
2270      1+IMM2_SIZE, 1                 /* CLOSE, SKIPZERO                        */
2271    
2272  /* A magic value for OP_RREF to indicate the "any recursion" condition. */  /* A magic value for OP_RREF to indicate the "any recursion" condition. */
2273    
2274  #define RREF_ANY  0xffff  #define RREF_ANY  0xffff
2275    
2276  /* 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
2277  tracked. */  easily be tracked. When a new number is added, the table called eint in
2278    pcreposix.c must be updated. */
2279    
2280  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
2281         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
# Line 871  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4, Line 2283  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,
2283         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
2284         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
2285         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
2286         ERR60 };         ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
2287           ERR70, ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERR77, ERR78, ERR79,
2288           ERR80, ERR81, ERR82, ERR83, ERR84, ERR85, ERR86, ERRCOUNT };
2289    
2290    /* JIT compiling modes. The function list is indexed by them. */
2291    
2292    enum { JIT_COMPILE, JIT_PARTIAL_SOFT_COMPILE, JIT_PARTIAL_HARD_COMPILE,
2293           JIT_NUMBER_OF_COMPILE_MODES };
2294    
2295  /* 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
2296  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
2297  offset to the name table so that if a regex is compiled on one host, saved, and  offset to the name table so that if a regex is compiled on one host, saved, and
2298  then run on another where the size of pointers is different, all might still  then run on another where the size of pointers is different, all might still
2299  be well. For the case of compiled-on-4 and run-on-8, we include an extra  be well.
2300  pointer that is always NULL. For future-proofing, a few dummy fields were  
2301  originally included - even though you can never get this planning right - but  The size of the structure must be a multiple of 8 bytes. For the case of
2302  there is only one left now.  compiled-on-4 and run-on-8, we include an extra pointer that is always NULL so
2303    that there are an even number of pointers which therefore are a multiple of 8
2304  NOTE NOTE NOTE:  bytes.
2305  Because people can now save and re-use compiled patterns, any additions to this  
2306  structure should be made at the end, and something earlier (e.g. a new  It is necessary to fork the struct for the 32 bit library, since it needs to
2307  flag in the options or one of the dummy fields) should indicate that the new  use pcre_uint32 for first_char and req_char. We can't put an ifdef inside the
2308  fields are present. Currently PCRE always sets the dummy fields to zero.  typedef because pcretest needs access to the struct of the 8-, 16- and 32-bit
2309  NOTE NOTE NOTE:  variants.
2310    
2311    *** WARNING ***
2312    When new fields are added to these structures, remember to adjust the code in
2313    pcre_byte_order.c that is concerned with swapping the byte order of the fields
2314    when a compiled regex is reloaded on a host with different endianness.
2315    *** WARNING ***
2316    There is also similar byte-flipping code in pcretest.c, which is used for
2317    testing the byte-flipping features. It must also be kept in step.
2318    *** WARNING ***
2319  */  */
2320    
2321  typedef struct real_pcre {  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_uint32 flags;              /* Private flags */
2326      pcre_uint32 limit_match;        /* Limit set from regex */
2327      pcre_uint32 limit_recursion;    /* Limit set from regex */
2328      pcre_uint16 first_char;         /* Starting character */
2329      pcre_uint16 req_char;           /* This character must be seen */
2330      pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2331      pcre_uint16 top_bracket;        /* Highest numbered group */
2332      pcre_uint16 top_backref;        /* Highest numbered back reference */
2333      pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2334      pcre_uint16 name_entry_size;    /* Size of any name items */
2335      pcre_uint16 name_count;         /* Number of name items */
2336      pcre_uint16 ref_count;          /* Reference count */
2337      pcre_uint16 dummy1;             /* To ensure size is a multiple of 8 */
2338      pcre_uint16 dummy2;             /* To ensure size is a multiple of 8 */
2339      pcre_uint16 dummy3;             /* To ensure size is a multiple of 8 */
2340      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2341      void             *nullpad;      /* NULL padding */
2342    } real_pcre8_or_16;
2343    
2344    pcre_uint16 top_bracket;  typedef struct real_pcre8_or_16 real_pcre;
2345    pcre_uint16 top_backref;  typedef struct real_pcre8_or_16 real_pcre16;
2346    pcre_uint16 first_byte;  
2347    pcre_uint16 req_byte;  typedef struct real_pcre32 {
2348      pcre_uint32 magic_number;
2349      pcre_uint32 size;               /* Total that was malloced */
2350      pcre_uint32 options;            /* Public options */
2351      pcre_uint32 flags;              /* Private flags */
2352      pcre_uint32 limit_match;        /* Limit set from regex */
2353      pcre_uint32 limit_recursion;    /* Limit set from regex */
2354      pcre_uint32 first_char;         /* Starting character */
2355      pcre_uint32 req_char;           /* This character must be seen */
2356      pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2357      pcre_uint16 top_bracket;        /* Highest numbered group */
2358      pcre_uint16 top_backref;        /* Highest numbered back reference */
2359    pcre_uint16 name_table_offset;  /* Offset to name table that follows */    pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2360    pcre_uint16 name_entry_size;    /* Size of any name items */    pcre_uint16 name_entry_size;    /* Size of any name items */
2361    pcre_uint16 name_count;         /* Number of name items */    pcre_uint16 name_count;         /* Number of name items */
2362    pcre_uint16 ref_count;          /* Reference count */    pcre_uint16 ref_count;          /* Reference count */
2363      pcre_uint16 dummy;              /* To ensure size is a multiple of 8 */
2364      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2365      void             *nullpad;      /* NULL padding */
2366    } real_pcre32;
2367    
2368    #if defined COMPILE_PCRE8
2369    #define REAL_PCRE real_pcre
2370    #elif defined COMPILE_PCRE16
2371    #define REAL_PCRE real_pcre16
2372    #elif defined COMPILE_PCRE32
2373    #define REAL_PCRE real_pcre32
2374    #endif
2375    
2376    const unsigned char *tables;    /* Pointer to tables or NULL for std */  /* Assert that the size of REAL_PCRE is divisible by 8 */
2377    const unsigned char *nullpad;   /* NULL padding */  typedef int __assert_real_pcre_size_divisible_8[(sizeof(REAL_PCRE) % 8) == 0 ? 1 : -1];
2378  } real_pcre;  
2379    /* Needed in pcretest to access some fields in the real_pcre* structures
2380     * directly. They're unified for 8/16/32 bits since the structs only differ
2381     * after these fields; if that ever changes, need to fork those defines into
2382     * 8/16 and 32 bit versions. */
2383    #define REAL_PCRE_MAGIC(re)     (((REAL_PCRE*)re)->magic_number)
2384    #define REAL_PCRE_SIZE(re)      (((REAL_PCRE*)re)->size)
2385    #define REAL_PCRE_OPTIONS(re)   (((REAL_PCRE*)re)->options)
2386    #define REAL_PCRE_FLAGS(re)     (((REAL_PCRE*)re)->flags)
2387    
2388  /* 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
2389  remark (see NOTE above) about extending this structure applies. */  remark (see NOTE above) about extending this structure applies. */
2390    
2391  typedef struct pcre_study_data {  typedef struct pcre_study_data {
2392    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2393    pcre_uint32 options;    pcre_uint32 flags;              /* Private flags */
2394    uschar start_bits[32];    pcre_uint8 start_bits[32];      /* Starting char bits */
2395      pcre_uint32 minlength;          /* Minimum subject length */
2396  } pcre_study_data;  } pcre_study_data;
2397    
2398    /* Structure for building a chain of open capturing subpatterns during
2399    compiling, so that instructions to close them can be compiled when (*ACCEPT) is
2400    encountered. This is also used to identify subpatterns that contain recursive
2401    back references to themselves, so that they can be made atomic. */
2402    
2403    typedef struct open_capitem {
2404      struct open_capitem *next;    /* Chain link */
2405      pcre_uint16 number;           /* Capture number */
2406      pcre_uint16 flag;             /* Set TRUE if recursive back ref */
2407    } open_capitem;
2408    
2409    /* Structure for building a list of named groups during the first pass of
2410    compiling. */
2411    
2412    typedef struct named_group {
2413      const pcre_uchar  *name;          /* Points to the name in the pattern */
2414      int                length;        /* Length of the name */
2415      pcre_uint32        number;        /* Group number */
2416    } named_group;
2417    
2418  /* Structure for passing "static" information around between the functions  /* Structure for passing "static" information around between the functions
2419  doing the compiling, so that they are thread-safe. */  doing the compiling, so that they are thread-safe. */
2420    
2421  typedef struct compile_data {  typedef struct compile_data {
2422    const uschar *lcc;            /* Points to lower casing table */    const pcre_uint8 *lcc;            /* Points to lower casing table */
2423    const uschar *fcc;            /* Points to case-flipping table */    const pcre_uint8 *fcc;            /* Points to case-flipping table */
2424    const uschar *cbits;          /* Points to character type table */    const pcre_uint8 *cbits;          /* Points to character type table */
2425    const uschar *ctypes;         /* Points to table of type maps */    const pcre_uint8 *ctypes;         /* Points to table of type maps */
2426    const uschar *start_workspace;/* The start of working space */    const pcre_uchar *start_workspace;/* The start of working space */
2427    const uschar *start_code;     /* The start of the compiled code */    const pcre_uchar *start_code;     /* The start of the compiled code */
2428    const uschar *start_pattern;  /* The start of the pattern */    const pcre_uchar *start_pattern;  /* The start of the pattern */
2429    const uschar *end_pattern;    /* The end of the pattern */    const pcre_uchar *end_pattern;    /* The end of the pattern */
2430    uschar *hwm;                  /* High watermark of workspace */    pcre_uchar *hwm;                  /* High watermark of workspace */
2431    uschar *name_table;           /* The name/number table */    open_capitem *open_caps;          /* Chain of open capture items */
2432    int  names_found;             /* Number of entries so far */    named_group *named_groups;        /* Points to vector in pre-compile */
2433    int  name_entry_size;         /* Size of each entry */    pcre_uchar *name_table;           /* The name/number table */
2434    int  bracount;                /* Count of capturing parens */    int  names_found;                 /* Number of entries so far */
2435    int  top_backref;             /* Maximum back reference */    int  name_entry_size;             /* Size of each entry */
2436    unsigned int backref_map;     /* Bitmap of low back refs */    int  named_group_list_size;       /* Number of entries in the list */
2437    int  external_options;        /* External (initial) options */    int  workspace_size;              /* Size of workspace */
2438    int  req_varyopt;             /* "After variable item" flag for reqbyte */    unsigned int bracount;            /* Count of capturing parens as we compile */
2439    BOOL nopartial;               /* Set TRUE if partial won't work */    int  final_bracount;              /* Saved value after first pass */
2440    BOOL had_accept;              /* (*ACCEPT) encountered */    int  max_lookbehind;              /* Maximum lookbehind (characters) */
2441    int  nltype;                  /* Newline type */    int  top_backref;                 /* Maximum back reference */
2442    int  nllen;                   /* Newline string length */    unsigned int backref_map;         /* Bitmap of low back refs */
2443    uschar nl[4];                 /* Newline string when fixed length */    unsigned int namedrefcount;       /* Number of backreferences by name */
2444      int  parens_depth;                /* Depth of nested parentheses */
2445      int  assert_depth;                /* Depth of nested assertions */
2446      pcre_uint32 external_options;     /* External (initial) options */
2447      pcre_uint32 external_flags;       /* External flag bits to be set */
2448      int  req_varyopt;                 /* "After variable item" flag for reqbyte */
2449      BOOL had_accept;                  /* (*ACCEPT) encountered */
2450      BOOL had_pruneorskip;             /* (*PRUNE) or (*SKIP) encountered */
2451      BOOL check_lookbehind;            /* Lookbehinds need later checking */
2452      BOOL dupnames;                    /* Duplicate names exist */
2453      BOOL iscondassert;                /* Next assert is a condition */
2454      int  nltype;                      /* Newline type */
2455      int  nllen;                       /* Newline string length */
2456      pcre_uchar nl[4];                 /* Newline string when fixed length */
2457  } compile_data;  } compile_data;
2458    
2459  /* Structure for maintaining a chain of pointers to the currently incomplete  /* Structure for maintaining a chain of pointers to the currently incomplete
2460  branches, for testing for left recursion. */  branches, for testing for left recursion while compiling. */
2461    
2462  typedef struct branch_chain {  typedef struct branch_chain {
2463    struct branch_chain *outer;    struct branch_chain *outer;
2464    uschar *current;    pcre_uchar *current_branch;
2465  } branch_chain;  } branch_chain;
2466    
2467    /* Structure for mutual recursion detection. */
2468    
2469    typedef struct recurse_check {
2470      struct recurse_check *prev;
2471      const pcre_uchar *group;
2472    } recurse_check;
2473    
2474  /* Structure for items in a linked list that represents an explicit recursive  /* Structure for items in a linked list that represents an explicit recursive
2475  call within the pattern. */  call within the pattern; used by pcre_exec(). */
2476    
2477  typedef struct recursion_info {  typedef struct recursion_info {
2478    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
2479    int group_num;                /* Number of group that was called */    unsigned int group_num;         /* Number of group that was called */
2480    const uschar *after_call;     /* "Return value": points after the call in the expr */    int *offset_save;               /* Pointer to start of saved offsets */
2481    USPTR save_start;             /* Old value of mstart */    int saved_max;                  /* Number of saved offsets */
2482    int *offset_save;             /* Pointer to start of saved offsets */    int saved_capture_last;         /* Last capture number */
2483    int saved_max;                /* Number of saved offsets */    PCRE_PUCHAR subject_position;   /* Position at start of recursion */
2484  } recursion_info;  } recursion_info;
2485    
2486    /* A similar structure for pcre_dfa_exec(). */
2487    
2488    typedef struct dfa_recursion_info {
2489      struct dfa_recursion_info *prevrec;
2490      int group_num;
2491      PCRE_PUCHAR subject_position;
2492    } dfa_recursion_info;
2493    
2494  /* 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
2495  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
2496  has been matched by a subpattern - to break infinite loops. */  has been matched by a subpattern - to break infinite loops; used by
2497    pcre_exec(). */
2498    
2499  typedef struct eptrblock {  typedef struct eptrblock {
2500    struct eptrblock *epb_prev;    struct eptrblock *epb_prev;
2501    USPTR epb_saved_eptr;    PCRE_PUCHAR epb_saved_eptr;
2502  } eptrblock;  } eptrblock;
2503    
2504    
# Line 983  typedef struct match_data { Line 2509  typedef struct match_data {
2509    unsigned long int match_call_count;      /* As it says */    unsigned long int match_call_count;      /* As it says */
2510    unsigned long int match_limit;           /* As it says */    unsigned long int match_limit;           /* As it says */
2511    unsigned long int match_limit_recursion; /* As it says */    unsigned long int match_limit_recursion; /* As it says */
2512    int   *offset_vector;         /* Offset vector */    int   *offset_vector;           /* Offset vector */
2513    int    offset_end;            /* One past the end */    int    offset_end;              /* One past the end */
2514    int    offset_max;            /* The maximum usable for return data */    int    offset_max;              /* The maximum usable for return data */
2515    int    nltype;                /* Newline type */    int    nltype;                  /* Newline type */
2516    int    nllen;                 /* Newline string length */    int    nllen;                   /* Newline string length */
2517    uschar nl[4];                 /* Newline string when fixed */    int    name_count;              /* Number of names in name table */
2518    const uschar *lcc;            /* Points to lower casing table */    int    name_entry_size;         /* Size of entry in names table */
2519    const uschar *ctypes;         /* Points to table of type maps */    unsigned int skip_arg_count;    /* For counting SKIP_ARGs */
2520    BOOL   offset_overflow;       /* Set if too many extractions */    unsigned int ignore_skip_arg;   /* For re-run when SKIP arg name not found */
2521    BOOL   notbol;                /* NOTBOL flag */    pcre_uchar *name_table;         /* Table of names */
2522    BOOL   noteol;                /* NOTEOL flag */    pcre_uchar nl[4];               /* Newline string when fixed */
2523    BOOL   utf8;                  /* UTF8 flag */    const  pcre_uint8 *lcc;         /* Points to lower casing table */
2524    BOOL   endonly;               /* Dollar not before final \n */    const  pcre_uint8 *fcc;         /* Points to case-flipping table */
2525    BOOL   notempty;              /* Empty string match not wanted */    const  pcre_uint8 *ctypes;      /* Points to table of type maps */
2526    BOOL   partial;               /* PARTIAL flag */    BOOL   notbol;                  /* NOTBOL flag */
2527    BOOL   hitend;                /* Hit the end of the subject at some point */    BOOL   noteol;                  /* NOTEOL flag */
2528    const uschar *start_code;     /* For use when recursing */    BOOL   utf;                     /* UTF-8 / UTF-16 flag */
2529    USPTR  start_subject;         /* Start of the subject string */    BOOL   jscript_compat;          /* JAVASCRIPT_COMPAT flag */
2530    USPTR  end_subject;           /* End of the subject string */    BOOL   use_ucp;                 /* PCRE_UCP flag */
2531    USPTR  start_match_ptr;       /* Start of matched string */    BOOL   endonly;                 /* Dollar not before final \n */
2532    USPTR  end_match_ptr;         /* Subject position at end match */    BOOL   notempty;                /* Empty string match not wanted */
2533    int    end_offset_top;        /* Highwater mark at end of match */    BOOL   notempty_atstart;        /* Empty string match at start not wanted */
2534    int    capture_last;          /* Most recent capture number */    BOOL   hitend;                  /* Hit the end of the subject at some point */
2535    int    start_offset;          /* The start offset value */    BOOL   bsr_anycrlf;             /* \R is just any CRLF, not full Unicode */
2536    eptrblock *eptrchain;         /* Chain of eptrblocks for tail recursions */    BOOL   hasthen;                 /* Pattern contains (*THEN) */
2537    int    eptrn;                 /* Next free eptrblock */    const  pcre_uchar *start_code;  /* For use when recursing */
2538    recursion_info *recursive;    /* Linked list of recursion data */    PCRE_PUCHAR start_subject;      /* Start of the subject string */
2539    void  *callout_data;          /* To pass back to callouts */    PCRE_PUCHAR end_subject;        /* End of the subject string */
2540      PCRE_PUCHAR start_match_ptr;    /* Start of matched string */
2541      PCRE_PUCHAR end_match_ptr;      /* Subject position at end match */
2542      PCRE_PUCHAR start_used_ptr;     /* Earliest consulted character */
2543      int    partial;                 /* PARTIAL options */
2544      int    end_offset_top;          /* Highwater mark at end of match */
2545      pcre_int32 capture_last;        /* Most recent capture number + overflow flag */
2546      int    start_offset;            /* The start offset value */
2547      int    match_function_type;     /* Set for certain special calls of MATCH() */
2548      eptrblock *eptrchain;           /* Chain of eptrblocks for tail recursions */
2549      int    eptrn;                   /* Next free eptrblock */
2550      recursion_info *recursive;      /* Linked list of recursion data */
2551      void  *callout_data;            /* To pass back to callouts */
2552      const  pcre_uchar *mark;        /* Mark pointer to pass back on success */
2553      const  pcre_uchar *nomatch_mark;/* Mark pointer to pass back on failure */
2554      const  pcre_uchar *once_target; /* Where to back up to for atomic groups */
2555    #ifdef NO_RECURSE
2556      void  *match_frames_base;       /* For remembering malloc'd frames */
2557    #endif
2558  } match_data;  } match_data;
2559    
2560  /* 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
2561  functions. */  functions. */
2562    
2563  typedef struct dfa_match_data {  typedef struct dfa_match_data {
2564    const uschar *start_code;     /* Start of the compiled pattern */    const pcre_uchar *start_code;     /* Start of the compiled pattern */
2565    const uschar *start_subject;  /* Start of the subject string */    const pcre_uchar *start_subject ; /* Start of the subject string */
2566    const uschar *end_subject;    /* End of subject string */    const pcre_uchar *end_subject;    /* End of subject string */
2567    const uschar *tables;         /* Character tables */    const pcre_uchar *start_used_ptr; /* Earliest consulted character */
2568    int   moptions;               /* Match options */    const pcre_uint8 *tables;         /* Character tables */
2569    int   poptions;               /* Pattern options */    int   start_offset;               /* The start offset value */
2570    int    nltype;                /* Newline type */    int   moptions;                   /* Match options */
2571    int    nllen;                 /* Newline string length */    int   poptions;                   /* Pattern options */
2572    uschar nl[4];                 /* Newline string when fixed */    int   nltype;                     /* Newline type */
2573    void  *callout_data;          /* To pass back to callouts */    int   nllen;                      /* Newline string length */
2574      pcre_uchar nl[4];                 /* Newline string when fixed */
2575      void *callout_data;               /* To pass back to callouts */
2576      dfa_recursion_info *recursive;    /* Linked list of recursion data */
2577  } dfa_match_data;  } dfa_match_data;
2578    
2579  /* Bit definitions for entries in the pcre_ctypes table. */  /* Bit definitions for entries in the pcre_ctypes table. */
# Line 1035  typedef struct dfa_match_data { Line 2582  typedef struct dfa_match_data {
2582  #define ctype_letter  0x02  #define ctype_letter  0x02
2583  #define ctype_digit   0x04  #define ctype_digit   0x04
2584  #define ctype_xdigit  0x08  #define ctype_xdigit  0x08
2585  #define ctype_word    0x10   /* alphameric or '_' */  #define ctype_word    0x10   /* alphanumeric or '_' */
2586  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */
2587    
2588  /* 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 1062  total length. */ Line 2609  total length. */
2609  #define ctypes_offset (cbits_offset + cbit_length)  #define ctypes_offset (cbits_offset + cbit_length)
2610  #define tables_length (ctypes_offset + 256)  #define tables_length (ctypes_offset + 256)
2611    
2612    /* Internal function and data prefixes. */
2613    
2614    #if defined COMPILE_PCRE8
2615    #ifndef PUBL
2616    #define PUBL(name) pcre_##name
2617    #endif
2618    #ifndef PRIV
2619    #define PRIV(name) _pcre_##name
2620    #endif
2621    #elif defined COMPILE_PCRE16
2622    #ifndef PUBL
2623    #define PUBL(name) pcre16_##name
2624    #endif
2625    #ifndef PRIV
2626    #define PRIV(name) _pcre16_##name
2627    #endif
2628    #elif defined COMPILE_PCRE32
2629    #ifndef PUBL
2630    #define PUBL(name) pcre32_##name
2631    #endif
2632    #ifndef PRIV
2633    #define PRIV(name) _pcre32_##name
2634    #endif
2635    #else
2636    #error Unsupported compiling mode
2637    #endif /* COMPILE_PCRE[8|16|32] */
2638    
2639  /* 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
2640  codes. */  codes. Each entry used to point directly to a name, but to reduce the number of
2641    relocations in shared libraries, it now has an offset into a single string
2642    instead. */
2643    
2644  typedef struct {  typedef struct {
2645    const char *name;    pcre_uint16 name_offset;
2646    pcre_uint16 type;    pcre_uint16 type;
2647    pcre_uint16 value;    pcre_uint16 value;
2648  } ucp_type_table;  } ucp_type_table;
# Line 1077  of the exported public functions. They h Line 2653  of the exported public functions. They h
2653  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
2654  pcre_tables.c module. */  pcre_tables.c module. */
2655    
2656  extern const int    _pcre_utf8_table1[];  #ifdef COMPILE_PCRE8
2657  extern const int    _pcre_utf8_table2[];  extern const int            PRIV(utf8_table1)[];
2658  extern const int    _pcre_utf8_table3[];  extern const int            PRIV(utf8_table1_size);
2659  extern const uschar _pcre_utf8_table4[];  extern const int            PRIV(utf8_table2)[];
2660    extern const int            PRIV(utf8_table3)[];
2661  extern const int    _pcre_utf8_table1_size;  extern const pcre_uint8     PRIV(utf8_table4)[];
2662    #endif /* COMPILE_PCRE8 */
2663  extern const ucp_type_table _pcre_utt[];  
2664  extern const int _pcre_utt_size;  extern const char           PRIV(utt_names)[];
2665    extern const ucp_type_table PRIV(utt)[];
2666    extern const int            PRIV(utt_size);
2667    
2668  extern const uschar _pcre_default_tables[];  extern const pcre_uint8     PRIV(OP_lengths)[];
2669    extern const pcre_uint8     PRIV(default_tables)[];
2670    
2671  extern const uschar _pcre_OP_lengths[];  extern const pcre_uint32    PRIV(hspace_list)[];
2672    extern const pcre_uint32    PRIV(vspace_list)[];
2673    
2674    
2675  /* Internal shared functions. These are functions that are used by more than  /* Internal shared functions. These are functions that are used by more than
2676  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
2677  sense, but are not part of the PCRE public API. */  sense, but are not part of the PCRE public API. */
2678    
2679  extern BOOL         _pcre_is_newline(const uschar *, int, const uschar *,  /* String comparison functions. */
2680                        int *, BOOL);  #if defined COMPILE_PCRE8
2681  extern int          _pcre_ord2utf8(int, uschar *);  
2682  extern real_pcre   *_pcre_try_flipped(const real_pcre *, real_pcre *,  #define STRCMP_UC_UC(str1, str2) \
2683                        const pcre_study_data *, pcre_study_data *);    strcmp((char *)(str1), (char *)(str2))
2684  extern int          _pcre_ucp_findprop(const unsigned int, int *, int *);  #define STRCMP_UC_C8(str1, str2) \
2685  extern unsigned int _pcre_ucp_othercase(const unsigned int);    strcmp((char *)(str1), (str2))
2686  extern int          _pcre_valid_utf8(const uschar *, int);  #define STRNCMP_UC_UC(str1, str2, num) \
2687  extern BOOL         _pcre_was_newline(const uschar *, int, const uschar *,    strncmp((char *)(str1), (char *)(str2), (num))
2688                        int *, BOOL);  #define STRNCMP_UC_C8(str1, str2, num) \
2689  extern BOOL         _pcre_xclass(int, const uschar *);    strncmp((char *)(str1), (str2), (num))
2690    #define STRLEN_UC(str) strlen((const char *)str)
2691    
2692    #elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
2693    
2694    extern int               PRIV(strcmp_uc_uc)(const pcre_uchar *,
2695                               const pcre_uchar *);
2696    extern int               PRIV(strcmp_uc_c8)(const pcre_uchar *,
2697                               const char *);
2698    extern int               PRIV(strncmp_uc_uc)(const pcre_uchar *,
2699                               const pcre_uchar *, unsigned int num);
2700    extern int               PRIV(strncmp_uc_c8)(const pcre_uchar *,
2701                               const char *, unsigned int num);
2702    extern unsigned int      PRIV(strlen_uc)(const pcre_uchar *str);
2703    
2704    #define STRCMP_UC_UC(str1, str2) \
2705      PRIV(strcmp_uc_uc)((str1), (str2))
2706    #define STRCMP_UC_C8(str1, str2) \
2707      PRIV(strcmp_uc_c8)((str1), (str2))
2708    #define STRNCMP_UC_UC(str1, str2, num) \
2709      PRIV(strncmp_uc_uc)((str1), (str2), (num))
2710    #define STRNCMP_UC_C8(str1, str2, num) \
2711      PRIV(strncmp_uc_c8)((str1), (str2), (num))
2712    #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2713    
2714    #endif /* COMPILE_PCRE[8|16|32] */
2715    
2716    #if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
2717    
2718    #define STRCMP_UC_UC_TEST(str1, str2) STRCMP_UC_UC(str1, str2)
2719    #define STRCMP_UC_C8_TEST(str1, str2) STRCMP_UC_C8(str1, str2)
2720    
2721    #elif defined COMPILE_PCRE32
2722    
2723    extern int               PRIV(strcmp_uc_uc_utf)(const pcre_uchar *,
2724                               const pcre_uchar *);
2725    extern int               PRIV(strcmp_uc_c8_utf)(const pcre_uchar *,
2726                               const char *);
2727    
2728    #define STRCMP_UC_UC_TEST(str1, str2) \
2729      (utf ? PRIV(strcmp_uc_uc_utf)((str1), (str2)) : PRIV(strcmp_uc_uc)((str1), (str2)))
2730    #define STRCMP_UC_C8_TEST(str1, str2) \
2731      (utf ? PRIV(strcmp_uc_c8_utf)((str1), (str2)) : PRIV(strcmp_uc_c8)((str1), (str2)))
2732    
2733    #endif /* COMPILE_PCRE[8|16|32] */
2734    
2735    extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2736    extern BOOL              PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2737                               int *, BOOL);
2738    extern unsigned int      PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
2739    extern int               PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
2740    extern BOOL              PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2741                               int *, BOOL);
2742    extern BOOL              PRIV(xclass)(pcre_uint32, const pcre_uchar *, BOOL);
2743    
2744    #ifdef SUPPORT_JIT
2745    extern void              PRIV(jit_compile)(const REAL_PCRE *,
2746                               PUBL(extra) *, int);
2747    extern int               PRIV(jit_exec)(const PUBL(extra) *,
2748                               const pcre_uchar *, int, int, int, int *, int);
2749    extern void              PRIV(jit_free)(void *);
2750    extern int               PRIV(jit_get_size)(void *);
2751    extern const char*       PRIV(jit_get_target)(void);
2752    #endif
2753    
2754    /* Unicode character database (UCD) */
2755    
2756    typedef struct {
2757      pcre_uint8 script;     /* ucp_Arabic, etc. */
2758      pcre_uint8 chartype;   /* ucp_Cc, etc. (general categories) */
2759      pcre_uint8 gbprop;     /* ucp_gbControl, etc. (grapheme break property) */
2760      pcre_uint8 caseset;    /* offset to multichar other cases or zero */
2761      pcre_int32 other_case; /* offset to other case, or zero if none */
2762    } ucd_record;
2763    
2764    extern const pcre_uint32 PRIV(ucd_caseless_sets)[];
2765    extern const ucd_record  PRIV(ucd_records)[];
2766    extern const pcre_uint8  PRIV(ucd_stage1)[];
2767    extern const pcre_uint16 PRIV(ucd_stage2)[];
2768    extern const pcre_uint32 PRIV(ucp_gentype)[];
2769    extern const pcre_uint32 PRIV(ucp_gbtable)[];
2770    #ifdef SUPPORT_JIT
2771    extern const int         PRIV(ucp_typerange)[];
2772    #endif
2773    
2774    #ifdef SUPPORT_UCP
2775    /* UCD access macros */
2776    
2777    #define UCD_BLOCK_SIZE 128
2778    #define GET_UCD(ch) (PRIV(ucd_records) + \
2779            PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \
2780            UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE])
2781    
2782    #define UCD_CHARTYPE(ch)    GET_UCD(ch)->chartype
2783    #define UCD_SCRIPT(ch)      GET_UCD(ch)->script
2784    #define UCD_CATEGORY(ch)    PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2785    #define UCD_GRAPHBREAK(ch)  GET_UCD(ch)->gbprop
2786    #define UCD_CASESET(ch)     GET_UCD(ch)->caseset
2787    #define UCD_OTHERCASE(ch)   ((pcre_uint32)((int)ch + (int)(GET_UCD(ch)->other_case)))
2788    
2789    #endif /* SUPPORT_UCP */
2790    
2791  #endif  #endif
2792    

Legend:
Removed from v.211  
changed lines
  Added in v.1567

  ViewVC Help
Powered by ViewVC 1.1.5