/[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 137 by ph10, Thu Mar 29 13:56:00 2007 UTC revision 1198 by chpe, Thu Nov 1 19:23:35 2012 UTC
# Line 7  Line 7 
7  and semantics are as close as possible to those of the Perl 5 language.  and semantics are as close as possible to those of the Perl 5 language.
8    
9                         Written by Philip Hazel                         Written by Philip Hazel
10             Copyright (c) 1997-2007 University of Cambridge             Copyright (c) 1997-2012 University of Cambridge
11    
12  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
13  Redistribution and use in source and binary forms, with or without  Redistribution and use in source and binary forms, with or without
# Line 40  POSSIBILITY OF SUCH DAMAGE. Line 40  POSSIBILITY OF SUCH DAMAGE.
40    
41  /* This header contains definitions that are shared between the different  /* This header contains definitions that are shared between the different
42  modules, but which are not relevant to the exported API. This includes some  modules, but which are not relevant to the exported API. This includes some
43  functions whose names all begin with "_pcre_". */  functions whose names all begin with "_pcre_", "_pcre16_" or "_pcre32_"
44    depending on the PRIV macro. */
45    
46  #ifndef PCRE_INTERNAL_H  #ifndef PCRE_INTERNAL_H
47  #define PCRE_INTERNAL_H  #define PCRE_INTERNAL_H
48    
49  /* Define DEBUG to get debugging output on stdout. */  /* Define PCRE_DEBUG to get debugging output on stdout. */
50    
51  #if 0  #if 0
52  #define DEBUG  #define PCRE_DEBUG
53    #endif
54    
55    /* PCRE is compiled as an 8 bit library if it is not requested otherwise. */
56    
57    #if !defined COMPILE_PCRE16 && !defined COMPILE_PCRE32
58    #define COMPILE_PCRE8
59    #endif
60    
61    /* If SUPPORT_UCP is defined, SUPPORT_UTF must also be defined. The
62    "configure" script ensures this, but not everybody uses "configure". */
63    
64    #if defined SUPPORT_UCP && !(defined SUPPORT_UTF)
65    #define SUPPORT_UTF 1
66    #endif
67    
68    /* We define SUPPORT_UTF if SUPPORT_UTF8 is enabled for compatibility
69    reasons with existing code. */
70    
71    #if defined SUPPORT_UTF8 && !(defined SUPPORT_UTF)
72    #define SUPPORT_UTF 1
73    #endif
74    
75    /* Fixme: SUPPORT_UTF8 should be eventually disappear from the code.
76    Until then we define it if SUPPORT_UTF is defined. */
77    
78    #if defined SUPPORT_UTF && !(defined SUPPORT_UTF8)
79    #define SUPPORT_UTF8 1
80    #endif
81    
82    /* We do not support both EBCDIC and UTF-8/16/32 at the same time. The "configure"
83    script prevents both being selected, but not everybody uses "configure". */
84    
85    #if defined EBCDIC && defined SUPPORT_UTF
86    #error The use of both EBCDIC and SUPPORT_UTF is not supported.
87  #endif  #endif
88    
89  /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef  /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
# Line 60  It turns out that the Mac Debugging.h he Line 95  It turns out that the Mac Debugging.h he
95  be absolutely sure we get our version. */  be absolutely sure we get our version. */
96    
97  #undef DPRINTF  #undef DPRINTF
98  #ifdef DEBUG  #ifdef PCRE_DEBUG
99  #define DPRINTF(p) printf p  #define DPRINTF(p) printf p
100  #else  #else
101  #define DPRINTF(p) /* Nothing */  #define DPRINTF(p) /* Nothing */
102  #endif  #endif
103    
104    
 /* Get the definitions provided by running "configure" */  
   
 #include "config.h"  
   
105  /* Standard C headers plus the external interface definition. The only time  /* Standard C headers plus the external interface definition. The only time
106  setjmp and stdarg are used is when NO_RECURSE is set. */  setjmp and stdarg are used is when NO_RECURSE is set. */
107    
108  #include <ctype.h>  #include <ctype.h>
109  #include <limits.h>  #include <limits.h>
 #include <setjmp.h>  
 #include <stdarg.h>  
110  #include <stddef.h>  #include <stddef.h>
111  #include <stdio.h>  #include <stdio.h>
112  #include <stdlib.h>  #include <stdlib.h>
113  #include <string.h>  #include <string.h>
114    
115  #ifndef PCRE_SPY  /* Valgrind (memcheck) support */
116  #define PCRE_DEFINITION       /* Win32 __declspec(export) trigger for .dll */  
117    #ifdef SUPPORT_VALGRIND
118    #include <valgrind/memcheck.h>
119  #endif  #endif
120    
121  /* We need to have types that specify unsigned 16-bit and 32-bit integers. We  /* 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:
123    http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
124    information there, using __declspec(dllexport) without "extern" we have a
125    definition; with "extern" we have a declaration. The settings here override the
126    setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL,
127    which is all that is needed for applications (they just import the symbols). We
128    use:
129    
130      PCRE_EXP_DECL       for declarations
131      PCRE_EXP_DEFN       for definitions of exported functions
132      PCRE_EXP_DATA_DEFN  for definitions of exported variables
133    
134    The reason for the two DEFN macros is that in non-Windows environments, one
135    does not want to have "extern" before variable definitions because it leads to
136    compiler warnings. So we distinguish between functions and variables. In
137    Windows, the two should always be the same.
138    
139    The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest,
140    which is an application, but needs to import this file in order to "peek" at
141    internals, can #include pcre.h first to get an application's-eye view.
142    
143    In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
144    special-purpose environments) might want to stick other stuff in front of
145    exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and
146    PCRE_EXP_DATA_DEFN only if they are not already set. */
147    
148    #ifndef PCRE_EXP_DECL
149    #  ifdef _WIN32
150    #    ifndef PCRE_STATIC
151    #      define PCRE_EXP_DECL       extern __declspec(dllexport)
152    #      define PCRE_EXP_DEFN       __declspec(dllexport)
153    #      define PCRE_EXP_DATA_DEFN  __declspec(dllexport)
154    #    else
155    #      define PCRE_EXP_DECL       extern
156    #      define PCRE_EXP_DEFN
157    #      define PCRE_EXP_DATA_DEFN
158    #    endif
159    #  else
160    #    ifdef __cplusplus
161    #      define PCRE_EXP_DECL       extern "C"
162    #    else
163    #      define PCRE_EXP_DECL       extern
164    #    endif
165    #    ifndef PCRE_EXP_DEFN
166    #      define PCRE_EXP_DEFN       PCRE_EXP_DECL
167    #    endif
168    #    ifndef PCRE_EXP_DATA_DEFN
169    #      define PCRE_EXP_DATA_DEFN
170    #    endif
171    #  endif
172    #endif
173    
174    /* When compiling with the MSVC compiler, it is sometimes necessary to include
175    a "calling convention" before exported function names. (This is secondhand
176    information; I know nothing about MSVC myself). For example, something like
177    
178      void __cdecl function(....)
179    
180    might be needed. In order so make this easy, all the exported functions have
181    PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
182    set, we ensure here that it has no effect. */
183    
184    #ifndef PCRE_CALL_CONVENTION
185    #define PCRE_CALL_CONVENTION
186    #endif
187    
188    /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
189  cannot determine these outside the compilation (e.g. by running a program as  cannot determine these outside the compilation (e.g. by running a program as
190  part of "configure") because PCRE is often cross-compiled for use on other  part of "configure") because PCRE is often cross-compiled for use on other
191  systems. Instead we make use of the maximum sizes that are available at  systems. Instead we make use of the maximum sizes that are available at
192  preprocessor time in standard C environments. */  preprocessor time in standard C environments. */
193    
194    typedef unsigned char pcre_uint8;
195    
196  #if USHRT_MAX == 65535  #if USHRT_MAX == 65535
197    typedef unsigned short pcre_uint16;    typedef unsigned short pcre_uint16;
198      typedef short pcre_int16;
199  #elif UINT_MAX == 65535  #elif UINT_MAX == 65535
200    typedef unsigned int pcre_uint16;    typedef unsigned int pcre_uint16;
201      typedef int pcre_int16;
202  #else  #else
203    #error Cannot determine a type for 16-bit unsigned integers    #error Cannot determine a type for 16-bit unsigned integers
204  #endif  #endif
205    
206  #if UINT_MAX == 4294967295  #if UINT_MAX == 4294967295
207    typedef unsigned int pcre_uint32;    typedef unsigned int pcre_uint32;
208      typedef int pcre_int32;
209  #elif ULONG_MAX == 4294967295  #elif ULONG_MAX == 4294967295
210    typedef unsigned long int pcre_uint32;    typedef unsigned long int pcre_uint32;
211      typedef long int pcre_int32;
212  #else  #else
213    #error Cannot determine a type for 32-bit unsigned integers    #error Cannot determine a type for 32-bit unsigned integers
214  #endif  #endif
215    
216    /* When checking for integer overflow in pcre_compile(), we need to handle
217    large integers. If a 64-bit integer type is available, we can use that.
218    Otherwise we have to cast to double, which of course requires floating point
219    arithmetic. Handle this by defining a macro for the appropriate type. If
220    stdint.h is available, include it; it may define INT64_MAX. Systems that do not
221    have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
222    by "configure". */
223    
224    #if defined HAVE_STDINT_H
225    #include <stdint.h>
226    #elif defined HAVE_INTTYPES_H
227    #include <inttypes.h>
228    #endif
229    
230    #if defined INT64_MAX || defined int64_t
231    #define INT64_OR_DOUBLE int64_t
232    #else
233    #define INT64_OR_DOUBLE double
234    #endif
235    
236  /* All character handling must be done as unsigned characters. Otherwise there  /* All character handling must be done as unsigned characters. Otherwise there
237  are problems with top-bit-set characters and functions such as isspace().  are problems with top-bit-set characters and functions such as isspace().
238  However, we leave the interface to the outside world as char *, because that  However, we leave the interface to the outside world as char * or short *,
239  should make things easier for callers. We define a short type for unsigned char  because that should make things easier for callers. This character type is
240  to save lots of typing. I tried "uchar", but it causes problems on Digital  called pcre_uchar.
241  Unix, where it is defined in sys/types, so use "uschar" instead. */  
242    The IN_UCHARS macro multiply its argument with the byte size of the current
243    pcre_uchar type. Useful for memcpy and such operations, whose require the
244    byte size of their input/output buffers.
245    
246    The MAX_255 macro checks whether its pcre_uchar input is less than 256.
247    
248    The TABLE_GET macro is designed for accessing elements of tables whose contain
249    exactly 256 items. When the character is able to contain more than 256
250    items, some check is needed before accessing these tables.
251    */
252    
253    #if defined COMPILE_PCRE8
254    
255  typedef unsigned char uschar;  typedef unsigned char pcre_uchar;
256    #define IN_UCHARS(x) (x)
257    #define MAX_255(c) 1
258    #define TABLE_GET(c, table, default) ((table)[c])
259    
260    #elif defined COMPILE_PCRE16
261    
262    #if USHRT_MAX != 65535
263    /* This is a warning message. Change PCRE_UCHAR16 to a 16 bit data type in
264    pcre.h(.in) and disable (comment out) this message. */
265    #error Warning: PCRE_UCHAR16 is not a 16 bit data type.
266    #endif
267    
268    typedef pcre_uint16 pcre_uchar;
269    #define UCHAR_SHIFT (1)
270    #define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
271    #define MAX_255(c) ((c) <= 255u)
272    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
273    
274    #elif defined COMPILE_PCRE32
275    
276    typedef pcre_uint32 pcre_uchar;
277    #define UCHAR_SHIFT (2)
278    #define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
279    #define MAX_255(c) ((c) <= 255u)
280    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
281    
282    /* Assert that pcre_uchar32 is a 32-bit type */
283    typedef int __assert_pcre_uchar32_size[sizeof(pcre_uchar) == 4 ? 1 : -1];
284    
285    #else
286    #error Unsupported compiling mode
287    #endif /* COMPILE_PCRE[8|16|32] */
288    
289  /* This is an unsigned int value that no character can ever have. UTF-8  /* This is an unsigned int value that no character can ever have. UTF-8
290  characters only go up to 0x7fffffff (though Unicode doesn't go beyond  characters only go up to 0x7fffffff (though Unicode doesn't go beyond
# Line 125  characters only go up to 0x7fffffff (tho Line 293  characters only go up to 0x7fffffff (tho
293  #define NOTACHAR 0xffffffff  #define NOTACHAR 0xffffffff
294    
295  /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,  /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
296  and "all" at present). The following macros are used to package up testing for  "any" and "anycrlf" at present). The following macros are used to package up
297  newlines. NLBLOCK, PSSTART, and PSEND are defined in the various modules to  testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
298  indicate in which datablock the parameters exist, and what the start/end of  modules to indicate in which datablock the parameters exist, and what the
299  string field names are. */  start/end of string field names are. */
300    
301  #define NLTYPE_FIXED   0     /* Newline is a fixed length string */  #define NLTYPE_FIXED    0     /* Newline is a fixed length string */
302  #define NLTYPE_ANY     1     /* Newline is any Unicode line ending */  #define NLTYPE_ANY      1     /* Newline is any Unicode line ending */
303    #define NLTYPE_ANYCRLF  2     /* Newline is CR, LF, or CRLF */
304    
305  /* This macro checks for a newline at the given position */  /* This macro checks for a newline at the given position */
306    
307  #define IS_NEWLINE(p) \  #define IS_NEWLINE(p) \
308    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
309      ((p) < NLBLOCK->PSEND && \      ((p) < NLBLOCK->PSEND && \
310       _pcre_is_newline((p), NLBLOCK->PSEND, &(NLBLOCK->nllen), utf8) \       PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
311      ) \         &(NLBLOCK->nllen), utf)) \
312      : \      : \
313      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
314       (p)[0] == NLBLOCK->nl[0] && \       RAWUCHARTEST(p) == NLBLOCK->nl[0] && \
315       (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \       (NLBLOCK->nllen == 1 || RAWUCHARTEST(p+1) == NLBLOCK->nl[1])       \
316      ) \      ) \
317    )    )
318    
# Line 152  string field names are. */ Line 321  string field names are. */
321  #define WAS_NEWLINE(p) \  #define WAS_NEWLINE(p) \
322    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
323      ((p) > NLBLOCK->PSSTART && \      ((p) > NLBLOCK->PSSTART && \
324       _pcre_was_newline((p), NLBLOCK->PSSTART, &(NLBLOCK->nllen), utf8) \       PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
325      ) \         &(NLBLOCK->nllen), utf)) \
326      : \      : \
327      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
328       (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \       RAWUCHARTEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] &&              \
329       (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \       (NLBLOCK->nllen == 1 || RAWUCHARTEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \
330      ) \      ) \
331    )    )
332    
# Line 171  used for the external interface and appe Line 340  used for the external interface and appe
340  must begin with PCRE_. */  must begin with PCRE_. */
341    
342  #ifdef CUSTOM_SUBJECT_PTR  #ifdef CUSTOM_SUBJECT_PTR
343  #define PCRE_SPTR CUSTOM_SUBJECT_PTR  #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
 #define USPTR CUSTOM_SUBJECT_PTR  
344  #else  #else
345  #define PCRE_SPTR const char *  #define PCRE_PUCHAR const pcre_uchar *
 #define USPTR const unsigned char *  
346  #endif  #endif
347    
348  /* Include the public PCRE header and the definitions of UCP character property  /* Include the public PCRE header and the definitions of UCP character property
349  values. */  values. */
350    
351  #include <pcre.h>  #include "pcre.h"
352  #include "ucp.h"  #include "ucp.h"
353    
354  /* When compiling for use with the Virtual Pascal compiler, these functions  /* When compiling for use with the Virtual Pascal compiler, these functions
# Line 245  The macros are controlled by the value o Line 412  The macros are controlled by the value o
412  the config.h file, but can be overridden by using -D on the command line. This  the config.h file, but can be overridden by using -D on the command line. This
413  is automated on Unix systems via the "configure" command. */  is automated on Unix systems via the "configure" command. */
414    
415    #if defined COMPILE_PCRE8
416    
417  #if LINK_SIZE == 2  #if LINK_SIZE == 2
418    
419  #define PUT(a,n,d)   \  #define PUT(a,n,d)   \
# Line 281  is automated on Unix systems via the "co Line 450  is automated on Unix systems via the "co
450  #define GET(a,n) \  #define GET(a,n) \
451    (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])    (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
452    
453  #define MAX_PATTERN_SIZE (1 << 30)   /* Keep it positive */  /* Keep it positive */
454    #define MAX_PATTERN_SIZE (1 << 30)
455    
456    #else
457    #error LINK_SIZE must be either 2, 3, or 4
458    #endif
459    
460    #elif defined COMPILE_PCRE16
461    
462    #if LINK_SIZE == 2
463    
464    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
465    #undef LINK_SIZE
466    #define LINK_SIZE 1
467    
468    #define PUT(a,n,d)   \
469      (a[n] = (d))
470    
471    #define GET(a,n) \
472      (a[n])
473    
474    #define MAX_PATTERN_SIZE (1 << 16)
475    
476    #elif LINK_SIZE == 3 || LINK_SIZE == 4
477    
478    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
479    #undef LINK_SIZE
480    #define LINK_SIZE 2
481    
482    #define PUT(a,n,d)   \
483      (a[n] = (d) >> 16), \
484      (a[(n)+1] = (d) & 65535)
485    
486    #define GET(a,n) \
487      (((a)[n] << 16) | (a)[(n)+1])
488    
489    /* Keep it positive */
490    #define MAX_PATTERN_SIZE (1 << 30)
491    
492  #else  #else
493  #error LINK_SIZE must be either 2, 3, or 4  #error LINK_SIZE must be either 2, 3, or 4
494  #endif  #endif
495    
496    #elif defined COMPILE_PCRE32
497    
498    /* Only supported LINK_SIZE is 4 */
499    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
500    #undef LINK_SIZE
501    #define LINK_SIZE 1
502    
503    #define PUT(a,n,d)   \
504      (a[n] = (d))
505    
506    #define GET(a,n) \
507      (a[n])
508    
509    /* Keep it positive */
510    #define MAX_PATTERN_SIZE (1 << 30)
511    
512    #else
513    #error Unsupported compiling mode
514    #endif /* COMPILE_PCRE[8|16|32] */
515    
516  /* Convenience macro defined in terms of the others */  /* Convenience macro defined in terms of the others */
517    
# Line 298  is automated on Unix systems via the "co Line 522  is automated on Unix systems via the "co
522  offsets changes. There are used for repeat counts and for other things such as  offsets changes. There are used for repeat counts and for other things such as
523  capturing parenthesis numbers in back references. */  capturing parenthesis numbers in back references. */
524    
525    #if defined COMPILE_PCRE8
526    
527    #define IMM2_SIZE 2
528    
529  #define PUT2(a,n,d)   \  #define PUT2(a,n,d)   \
530    a[n] = (d) >> 8; \    a[n] = (d) >> 8; \
531    a[(n)+1] = (d) & 255    a[(n)+1] = (d) & 255
532    
533    /* For reasons that I do not understand, the expression in this GET2 macro is
534    treated by gcc as a signed expression, even when a is declared as unsigned. It
535    seems that any kind of arithmetic results in a signed value. */
536    
537  #define GET2(a,n) \  #define GET2(a,n) \
538    (((a)[n] << 8) | (a)[(n)+1])    (unsigned int)(((a)[n] << 8) | (a)[(n)+1])
539    
540  #define PUT2INC(a,n,d)  PUT2(a,n,d), a += 2  #elif defined COMPILE_PCRE16
541    
542    #define IMM2_SIZE 1
543    
544  /* When UTF-8 encoding is being used, a character is no longer just a single  #define PUT2(a,n,d)   \
545  byte. The macros for character handling generate simple sequences when used in     a[n] = d
546  byte-mode, and more complicated ones for UTF-8 characters. */  
547    #define GET2(a,n) \
548       a[n]
549    
550    #elif defined COMPILE_PCRE32
551    
552    #define IMM2_SIZE 1
553    
554    #define PUT2(a,n,d)   \
555       a[n] = d
556    
557    #define GET2(a,n) \
558       a[n]
559    
560  #ifndef SUPPORT_UTF8  #else
561    #error Unsupported compiling mode
562    #endif /* COMPILE_PCRE[8|16|32] */
563    
564    #define PUT2INC(a,n,d)  PUT2(a,n,d), a += IMM2_SIZE
565    
566    /* The maximum length of a MARK name is currently one data unit; it may be
567    changed in future to be a fixed number of bytes or to depend on LINK_SIZE. */
568    
569    #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
570    #define MAX_MARK ((1u << 16) - 1)
571    #else
572    #define MAX_MARK ((1u << 8) - 1)
573    #endif
574    
575    /* When UTF encoding is being used, a character is no longer just a single
576    byte. The macros for character handling generate simple sequences when used in
577    character-mode, and more complicated ones for UTF characters. GETCHARLENTEST
578    and other macros are not used when UTF is not supported, so they are not
579    defined. To make sure they can never even appear when UTF support is omitted,
580    we don't even define them. */
581    
582    #ifndef SUPPORT_UTF
583    
584    /* #define MAX_VALUE_FOR_SINGLE_CHAR */
585    /* #define HAS_EXTRALEN(c) */
586    /* #define GET_EXTRALEN(c) */
587    /* #define NOT_FIRSTCHAR(c) */
588  #define GETCHAR(c, eptr) c = *eptr;  #define GETCHAR(c, eptr) c = *eptr;
589  #define GETCHARTEST(c, eptr) c = *eptr;  #define GETCHARTEST(c, eptr) c = *eptr;
590  #define GETCHARINC(c, eptr) c = *eptr++;  #define GETCHARINC(c, eptr) c = *eptr++;
591  #define GETCHARINCTEST(c, eptr) c = *eptr++;  #define GETCHARINCTEST(c, eptr) c = *eptr++;
592  #define GETCHARLEN(c, eptr, len) c = *eptr;  #define GETCHARLEN(c, eptr, len) c = *eptr;
593  #define BACKCHAR(eptr)  #define RAWUCHAR(eptr) (*(eptr))
594    #define RAWUCHARINC(eptr) (*(eptr)++)
595    #define RAWUCHARTEST(eptr) (*(eptr))
596    #define RAWUCHARINCTEST(eptr) (*(eptr)++)
597    /* #define GETCHARLENTEST(c, eptr, len) */
598    /* #define BACKCHAR(eptr) */
599    /* #define FORWARDCHAR(eptr) */
600    /* #define ACROSSCHAR(condition, eptr, action) */
601    
602  #else   /* SUPPORT_UTF8 */  #else   /* SUPPORT_UTF */
603    
604  /* Get the next UTF-8 character, not advancing the pointer. This is called when  /* Tests whether the code point needs extra characters to decode. */
 we know we are in UTF-8 mode. */  
605    
606  #define GETCHAR(c, eptr) \  #define HASUTF8EXTRALEN(c) ((c) >= 0xc0)
607    c = *eptr; \  
608    if (c >= 0xc0) \  /* Base macro to pick up the remaining bytes of a UTF-8 character, not
609    advancing the pointer. */
610    
611    #define GETUTF8(c, eptr) \
612        { \
613        if ((c & 0x20) == 0) \
614          c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
615        else if ((c & 0x10) == 0) \
616          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
617        else if ((c & 0x08) == 0) \
618          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
619          ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
620        else if ((c & 0x04) == 0) \
621          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
622              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
623              (eptr[4] & 0x3f); \
624        else \
625          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
626              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
627              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
628        }
629    
630    /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
631    the pointer. */
632    
633    #define GETUTF8INC(c, eptr) \
634      { \      { \
635      int gcii; \      if ((c & 0x20) == 0) \
636      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \        c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
637      int gcss = 6*gcaa; \      else if ((c & 0x10) == 0) \
638      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        { \
639      for (gcii = 1; gcii <= gcaa; gcii++) \        c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
640          eptr += 2; \
641          } \
642        else if ((c & 0x08) == 0) \
643        { \        { \
644        gcss -= 6; \        c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
645        c |= (eptr[gcii] & 0x3f) << gcss; \            ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
646          eptr += 3; \
647          } \
648        else if ((c & 0x04) == 0) \
649          { \
650          c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
651              ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
652              (eptr[3] & 0x3f); \
653          eptr += 4; \
654          } \
655        else \
656          { \
657          c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
658              ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
659              ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
660          eptr += 5; \
661        } \        } \
662      }      }
663    
664    #if defined COMPILE_PCRE8
665    
666    /* These macros were originally written in the form of loops that used data
667    from the tables whose names start with PRIV(utf8_table). They were rewritten by
668    a user so as not to use loops, because in some environments this gives a
669    significant performance advantage, and it seems never to do any harm. */
670    
671    /* Tells the biggest code point which can be encoded as a single character. */
672    
673    #define MAX_VALUE_FOR_SINGLE_CHAR 127
674    
675    /* Tests whether the code point needs extra characters to decode. */
676    
677    #define HAS_EXTRALEN(c) ((c) >= 0xc0)
678    
679    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
680    Otherwise it has an undefined behaviour. */
681    
682    #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
683    
684    /* Returns TRUE, if the given character is not the first character
685    of a UTF sequence. */
686    
687    #define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80)
688    
689    /* Get the next UTF-8 character, not advancing the pointer. This is called when
690    we know we are in UTF-8 mode. */
691    
692    #define GETCHAR(c, eptr) \
693      c = *eptr; \
694      if (c >= 0xc0) GETUTF8(c, eptr);
695    
696  /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the  /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
697  pointer. */  pointer. */
698    
699  #define GETCHARTEST(c, eptr) \  #define GETCHARTEST(c, eptr) \
700    c = *eptr; \    c = *eptr; \
701    if (utf8 && c >= 0xc0) \    if (utf && c >= 0xc0) GETUTF8(c, eptr);
     { \  
     int gcii; \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     for (gcii = 1; gcii <= gcaa; gcii++) \  
       { \  
       gcss -= 6; \  
       c |= (eptr[gcii] & 0x3f) << gcss; \  
       } \  
     }  
702    
703  /* Get the next UTF-8 character, advancing the pointer. This is called when we  /* Get the next UTF-8 character, advancing the pointer. This is called when we
704  know we are in UTF-8 mode. */  know we are in UTF-8 mode. */
705    
706  #define GETCHARINC(c, eptr) \  #define GETCHARINC(c, eptr) \
707    c = *eptr++; \    c = *eptr++; \
708    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8INC(c, eptr);
     { \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     while (gcaa-- > 0) \  
       { \  
       gcss -= 6; \  
       c |= (*eptr++ & 0x3f) << gcss; \  
       } \  
     }  
709    
710  /* Get the next character, testing for UTF-8 mode, and advancing the pointer */  /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
711    This is called when we don't know if we are in UTF-8 mode. */
712    
713  #define GETCHARINCTEST(c, eptr) \  #define GETCHARINCTEST(c, eptr) \
714    c = *eptr++; \    c = *eptr++; \
715    if (utf8 && c >= 0xc0) \    if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
716    
717    /* Base macro to pick up the remaining bytes of a UTF-8 character, not
718    advancing the pointer, incrementing the length. */
719    
720    #define GETUTF8LEN(c, eptr, len) \
721      { \      { \
722      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \      if ((c & 0x20) == 0) \
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     while (gcaa-- > 0) \  
723        { \        { \
724        gcss -= 6; \        c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
725        c |= (*eptr++ & 0x3f) << gcss; \        len++; \
726          } \
727        else if ((c & 0x10)  == 0) \
728          { \
729          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
730          len += 2; \
731          } \
732        else if ((c & 0x08)  == 0) \
733          {\
734          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
735              ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
736          len += 3; \
737          } \
738        else if ((c & 0x04)  == 0) \
739          { \
740          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
741              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
742              (eptr[4] & 0x3f); \
743          len += 4; \
744          } \
745        else \
746          {\
747          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
748              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
749              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
750          len += 5; \
751        } \        } \
752      }      }
753    
# Line 396  if there are extra bytes. This is called Line 756  if there are extra bytes. This is called
756    
757  #define GETCHARLEN(c, eptr, len) \  #define GETCHARLEN(c, eptr, len) \
758    c = *eptr; \    c = *eptr; \
759    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
760      { \  
761      int gcii; \  /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
762      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  pointer, incrementing length if there are extra bytes. This is called when we
763      int gcss = 6*gcaa; \  do not know if we are in UTF-8 mode. */
764      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
765      for (gcii = 1; gcii <= gcaa; gcii++) \  #define GETCHARLENTEST(c, eptr, len) \
766        { \    c = *eptr; \
767        gcss -= 6; \    if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
768        c |= (eptr[gcii] & 0x3f) << gcss; \  
769        } \  /* Returns the next uchar, not advancing the pointer. This is called when
770      len += gcaa; \  we know we are in UTF mode. */
771      }  
772    #define RAWUCHAR(eptr) \
773      (*(eptr))
774    
775    /* Returns the next uchar, advancing the pointer. This is called when
776    we know we are in UTF mode. */
777    
778    #define RAWUCHARINC(eptr) \
779      (*((eptr)++))
780    
781    /* Returns the next uchar, testing for UTF mode, and not advancing the
782    pointer. */
783    
784    #define RAWUCHARTEST(eptr) \
785      (*(eptr))
786    
787    /* Returns the next uchar, testing for UTF mode, advancing the
788    pointer. */
789    
790    #define RAWUCHARINCTEST(eptr) \
791      (*((eptr)++))
792    
793  /* If the pointer is not at the start of a character, move it back until  /* If the pointer is not at the start of a character, move it back until
794  it is. Called only in UTF-8 mode. */  it is. This is called only in UTF-8 mode - we don't put a test within the macro
795    because almost all calls are already within a block of UTF-8 only code. */
796    
797  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--;  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
798    
799  #endif  /* Same as above, just in the other direction. */
800    #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
801    
802    /* Same as above, but it allows a fully customizable form. */
803    #define ACROSSCHAR(condition, eptr, action) \
804      while((condition) && ((eptr) & 0xc0) == 0x80) action
805    
806  /* In case there is no definition of offsetof() provided - though any proper  #elif defined COMPILE_PCRE16
 Standard C system should have one. */  
807    
808  #ifndef offsetof  /* Tells the biggest code point which can be encoded as a single character. */
809  #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))  
810  #endif  #define MAX_VALUE_FOR_SINGLE_CHAR 65535
811    
812    /* Tests whether the code point needs extra characters to decode. */
813    
814    #define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
815    
816    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
817    Otherwise it has an undefined behaviour. */
818    
819    #define GET_EXTRALEN(c) 1
820    
821    /* Returns TRUE, if the given character is not the first character
822    of a UTF sequence. */
823    
824  /* These are the public options that can change during matching. */  #define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00)
825    
826  #define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)  /* Base macro to pick up the low surrogate of a UTF-16 character, not
827    advancing the pointer. */
828    
829  /* Private options flags start at the most significant end of the four bytes.  #define GETUTF16(c, eptr) \
830  The public options defined in pcre.h start at the least significant end. Make     { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
 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. */  
831    
832  #define PCRE_NOPARTIAL     0x80000000  /* can't use partial with this regex */  /* Get the next UTF-16 character, not advancing the pointer. This is called when
833  #define PCRE_FIRSTSET      0x40000000  /* first_byte is set */  we know we are in UTF-16 mode. */
 #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 */  
834    
835  /* Options for the "extra" block produced by pcre_study(). */  #define GETCHAR(c, eptr) \
836      c = *eptr; \
837      if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
838    
839    /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
840    pointer. */
841    
842    #define GETCHARTEST(c, eptr) \
843      c = *eptr; \
844      if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
845    
846    /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
847    the pointer. */
848    
849    #define GETUTF16INC(c, eptr) \
850       { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
851    
852    /* Get the next UTF-16 character, advancing the pointer. This is called when we
853    know we are in UTF-16 mode. */
854    
855    #define GETCHARINC(c, eptr) \
856      c = *eptr++; \
857      if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
858    
859    /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
860    This is called when we don't know if we are in UTF-16 mode. */
861    
862    #define GETCHARINCTEST(c, eptr) \
863      c = *eptr++; \
864      if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
865    
866    /* Base macro to pick up the low surrogate of a UTF-16 character, not
867    advancing the pointer, incrementing the length. */
868    
869    #define GETUTF16LEN(c, eptr, len) \
870       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
871    
872    /* Get the next UTF-16 character, not advancing the pointer, incrementing
873    length if there is a low surrogate. This is called when we know we are in
874    UTF-16 mode. */
875    
876    #define GETCHARLEN(c, eptr, len) \
877      c = *eptr; \
878      if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
879    
880    /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
881    pointer, incrementing length if there is a low surrogate. This is called when
882    we do not know if we are in UTF-16 mode. */
883    
884    #define GETCHARLENTEST(c, eptr, len) \
885      c = *eptr; \
886      if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
887    
888    /* Returns the next uchar, not advancing the pointer. This is called when
889    we know we are in UTF mode. */
890    
891    #define RAWUCHAR(eptr) \
892      (*(eptr))
893    
894    /* Returns the next uchar, advancing the pointer. This is called when
895    we know we are in UTF mode. */
896    
897    #define RAWUCHARINC(eptr) \
898      (*((eptr)++))
899    
900    /* Returns the next uchar, testing for UTF mode, and not advancing the
901    pointer. */
902    
903    #define RAWUCHARTEST(eptr) \
904      (*(eptr))
905    
906    /* Returns the next uchar, testing for UTF mode, advancing the
907    pointer. */
908    
909    #define RAWUCHARINCTEST(eptr) \
910      (*((eptr)++))
911    
912    /* If the pointer is not at the start of a character, move it back until
913    it is. This is called only in UTF-16 mode - we don't put a test within the
914    macro because almost all calls are already within a block of UTF-16 only
915    code. */
916    
917    #define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
918    
919    /* Same as above, just in the other direction. */
920    #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
921    
922    /* Same as above, but it allows a fully customizable form. */
923    #define ACROSSCHAR(condition, eptr, action) \
924      if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
925    
926    #elif defined COMPILE_PCRE32
927    
928    /* These are trivial for the 32-bit library, since all UTF-32 characters fit
929    into one pcre_uchar unit. */
930    #define MAX_VALUE_FOR_SINGLE_CHAR (0x10ffffu)
931    #define HAS_EXTRALEN(c) (0)
932    #define GET_EXTRALEN(c) (0)
933    #define NOT_FIRSTCHAR(c) (0)
934    
935    #define UTF32_MASK (0x1fffffu)
936    
937    /* Base macro to pick up an UTF-32 character out of a uint32 */
938    
939    #define MASKHIGHBITS(c) ((c) & UTF32_MASK)
940    
941    /* Base macro to pick up an UTF-32 character, not advancing the pointer */
942    
943    #define GETUTF32(eptr) (MASKHIGHBITS(*(eptr)))
944    
945    /* Base macro to pick up an UTF-32 character, advancing the pointer */
946    
947    #define GETUTF32INC(eptr) (MASKHIGHBITS(*((eptr)++)))
948    
949    /* Get the next UTF-32 character, not advancing the pointer. This is called when
950    we know we are in UTF-32 mode. */
951    
952    #define GETCHAR(c, eptr) \
953      c = GETUTF32(eptr);
954    
955    /* Get the next UTF-32 character, testing for UTF-32 mode, and not advancing the
956    pointer. */
957    
958    #define GETCHARTEST(c, eptr) \
959      c = (utf ? GETUTF32(eptr) : *(eptr));
960    
961    /* Get the next UTF-32 character, advancing the pointer. This is called when we
962    know we are in UTF-32 mode. */
963    
964    #define GETCHARINC(c, eptr) \
965      c = GETUTF32INC(eptr);
966    
967    /* Get the next character, testing for UTF-32 mode, and advancing the pointer.
968    This is called when we don't know if we are in UTF-32 mode. */
969    
970    #define GETCHARINCTEST(c, eptr) \
971      c = (utf ? GETUTF32INC(eptr) : *((eptr)++));
972    
973    /* Get the next UTF-32 character, not advancing the pointer, not incrementing
974    length (since all UTF-32 is of length 1). This is called when we know we are in
975    UTF-32 mode. */
976    
977    #define GETCHARLEN(c, eptr, len) \
978      GETCHAR(c, eptr)
979    
980    /* Get the next UTF-32character, testing for UTF-32 mode, not advancing the
981    pointer, not incrementing the length (since all UTF-32 is of length 1).
982    This is called when we do not know if we are in UTF-32 mode. */
983    
984    #define GETCHARLENTEST(c, eptr, len) \
985      GETCHARTEST(c, eptr)
986    
987    /* Returns the next uchar, not advancing the pointer. This is called when
988    we know we are in UTF mode. */
989    
990    #define RAWUCHAR(eptr) \
991      (MASKHIGHBITS(*(eptr)))
992    
993    /* Returns the next uchar, advancing the pointer. This is called when
994    we know we are in UTF mode. */
995    
996    #define RAWUCHARINC(eptr) \
997      (MASKHIGHBITS(*((eptr)++)))
998    
999    /* Returns the next uchar, testing for UTF mode, and not advancing the
1000    pointer. */
1001    
1002    #define RAWUCHARTEST(eptr) \
1003      (utf ? (MASKHIGHBITS(*(eptr))) : *(eptr))
1004    
1005    /* Returns the next uchar, testing for UTF mode, advancing the
1006    pointer. */
1007    
1008    #define RAWUCHARINCTEST(eptr) \
1009      (utf ? (MASKHIGHBITS(*((eptr)++))) : *((eptr)++))
1010    
1011    /* If the pointer is not at the start of a character, move it back until
1012    it is. This is called only in UTF-32 mode - we don't put a test within the
1013    macro because almost all calls are already within a block of UTF-32 only
1014    code.
1015    These are all no-ops since all UTF-32 characters fit into one pcre_uchar. */
1016    
1017    #define BACKCHAR(eptr) do { } while (0)
1018    
1019  #define PCRE_STUDY_MAPPED   0x01     /* a map of starting chars exists */  /* Same as above, just in the other direction. */
1020    #define FORWARDCHAR(eptr) do { } while (0)
1021    
1022    /* Same as above, but it allows a fully customizable form. */
1023    #define ACROSSCHAR(condition, eptr, action) do { } while (0)
1024    
1025    #else
1026    #error Unsupported compiling mode
1027    #endif /* COMPILE_PCRE[8|16|32] */
1028    
1029    #endif  /* SUPPORT_UTF */
1030    
1031    /* Tests for Unicode horizontal and vertical whitespace characters must check a
1032    number of different values. Using a switch statement for this generates the
1033    fastest code (no loop, no memory access), and there are several places in the
1034    interpreter code where this happens. In order to ensure that all the case lists
1035    remain in step, we use macros so that there is only one place where the lists
1036    are defined.
1037    
1038    These values are also required as lists in pcre_compile.c when processing \h,
1039    \H, \v and \V in a character class. The lists are defined in pcre_tables.c, but
1040    macros that define the values are here so that all the definitions are
1041    together. The lists must be in ascending character order, terminated by
1042    NOTACHAR (which is 0xffffffff).
1043    
1044    Any changes should ensure that the various macros are kept in step with each
1045    other. NOTE: The values also appear in pcre_jit_compile.c. */
1046    
1047    /* ------ ASCII/Unicode environments ------ */
1048    
1049    #ifndef EBCDIC
1050    
1051    #define HSPACE_LIST \
1052      CHAR_HT, CHAR_SPACE, 0xa0, \
1053      0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \
1054      0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \
1055      NOTACHAR
1056    
1057    #define HSPACE_MULTIBYTE_CASES \
1058      case 0x1680:  /* OGHAM SPACE MARK */ \
1059      case 0x180e:  /* MONGOLIAN VOWEL SEPARATOR */ \
1060      case 0x2000:  /* EN QUAD */ \
1061      case 0x2001:  /* EM QUAD */ \
1062      case 0x2002:  /* EN SPACE */ \
1063      case 0x2003:  /* EM SPACE */ \
1064      case 0x2004:  /* THREE-PER-EM SPACE */ \
1065      case 0x2005:  /* FOUR-PER-EM SPACE */ \
1066      case 0x2006:  /* SIX-PER-EM SPACE */ \
1067      case 0x2007:  /* FIGURE SPACE */ \
1068      case 0x2008:  /* PUNCTUATION SPACE */ \
1069      case 0x2009:  /* THIN SPACE */ \
1070      case 0x200A:  /* HAIR SPACE */ \
1071      case 0x202f:  /* NARROW NO-BREAK SPACE */ \
1072      case 0x205f:  /* MEDIUM MATHEMATICAL SPACE */ \
1073      case 0x3000   /* IDEOGRAPHIC SPACE */
1074    
1075    #define HSPACE_BYTE_CASES \
1076      case CHAR_HT: \
1077      case CHAR_SPACE: \
1078      case 0xa0     /* NBSP */
1079    
1080    #define HSPACE_CASES \
1081      HSPACE_BYTE_CASES: \
1082      HSPACE_MULTIBYTE_CASES
1083    
1084    #define VSPACE_LIST \
1085      CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR
1086    
1087    #define VSPACE_MULTIBYTE_CASES \
1088      case 0x2028:    /* LINE SEPARATOR */ \
1089      case 0x2029     /* PARAGRAPH SEPARATOR */
1090    
1091    #define VSPACE_BYTE_CASES \
1092      case CHAR_LF: \
1093      case CHAR_VT: \
1094      case CHAR_FF: \
1095      case CHAR_CR: \
1096      case CHAR_NEL
1097    
1098    #define VSPACE_CASES \
1099      VSPACE_BYTE_CASES: \
1100      VSPACE_MULTIBYTE_CASES
1101    
1102    /* ------ EBCDIC environments ------ */
1103    
1104    #else
1105    #define HSPACE_LIST CHAR_HT, CHAR_SPACE
1106    
1107    #define HSPACE_BYTE_CASES \
1108      case CHAR_HT: \
1109      case CHAR_SPACE
1110    
1111    #define HSPACE_CASES HSPACE_BYTE_CASES
1112    
1113    #ifdef EBCDIC_NL25
1114    #define VSPACE_LIST \
1115      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR
1116    #else
1117    #define VSPACE_LIST \
1118      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR
1119    #endif
1120    
1121    #define VSPACE_BYTE_CASES \
1122      case CHAR_LF: \
1123      case CHAR_VT: \
1124      case CHAR_FF: \
1125      case CHAR_CR: \
1126      case CHAR_NEL
1127    
1128    #define VSPACE_CASES VSPACE_BYTE_CASES
1129    #endif  /* EBCDIC */
1130    
1131    /* ------ End of whitespace macros ------ */
1132    
1133    
1134    
1135    /* Private flags containing information about the compiled regex. They used to
1136    live at the top end of the options word, but that got almost full, so now they
1137    are in a 16-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as
1138    the restrictions on partial matching have been lifted. It remains for backwards
1139    compatibility. */
1140    
1141    #define PCRE_MODE8         0x0001  /* compiled in 8 bit mode */
1142    #define PCRE_MODE16        0x0002  /* compiled in 16 bit mode */
1143    #define PCRE_MODE32        0x0004  /* compiled in 32 bit mode */
1144    #define PCRE_FIRSTSET      0x0010  /* first_char is set */
1145    #define PCRE_FCH_CASELESS  0x0020  /* caseless first char */
1146    #define PCRE_REQCHSET      0x0040  /* req_byte is set */
1147    #define PCRE_RCH_CASELESS  0x0080  /* caseless requested char */
1148    #define PCRE_STARTLINE     0x0100  /* start after \n for multiline */
1149    #define PCRE_NOPARTIAL     0x0200  /* can't use partial with this regex */
1150    #define PCRE_JCHANGED      0x0400  /* j option used in regex */
1151    #define PCRE_HASCRORLF     0x0800  /* explicit \r or \n in pattern */
1152    #define PCRE_HASTHEN       0x1000  /* pattern contains (*THEN) */
1153    
1154    #if defined COMPILE_PCRE8
1155    #define PCRE_MODE          PCRE_MODE8
1156    #elif defined COMPILE_PCRE16
1157    #define PCRE_MODE          PCRE_MODE16
1158    #elif defined COMPILE_PCRE32
1159    #define PCRE_MODE          PCRE_MODE32
1160    #endif
1161    #define PCRE_MODE_MASK     (PCRE_MODE8 | PCRE_MODE16 | PCRE_MODE32)
1162    
1163    /* Flags for the "extra" block produced by pcre_study(). */
1164    
1165    #define PCRE_STUDY_MAPPED  0x0001  /* a map of starting chars exists */
1166    #define PCRE_STUDY_MINLEN  0x0002  /* a minimum length field exists */
1167    
1168  /* Masks for identifying the public options that are permitted at compile  /* Masks for identifying the public options that are permitted at compile
1169  time, run time, or study time, respectively. */  time, run time, or study time, respectively. */
1170    
1171  #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| \
1172                               PCRE_NEWLINE_ANYCRLF)
1173    
1174  #define PUBLIC_OPTIONS \  #define PUBLIC_COMPILE_OPTIONS \
1175    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
1176     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
1177     PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \     PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
1178     PCRE_DUPNAMES|PCRE_NEWLINE_BITS)     PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1179       PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE)
1180    
1181  #define PUBLIC_EXEC_OPTIONS \  #define PUBLIC_EXEC_OPTIONS \
1182    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1183     PCRE_PARTIAL|PCRE_NEWLINE_BITS)     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
1184       PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
1185    
1186  #define PUBLIC_DFA_EXEC_OPTIONS \  #define PUBLIC_DFA_EXEC_OPTIONS \
1187    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1188     PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_BITS)     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
1189       PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1190       PCRE_NO_START_OPTIMIZE)
1191    
1192    #define PUBLIC_STUDY_OPTIONS \
1193       (PCRE_STUDY_JIT_COMPILE|PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE| \
1194        PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE|PCRE_STUDY_EXTRA_NEEDED)
1195    
1196    #define PUBLIC_JIT_EXEC_OPTIONS \
1197       (PCRE_NO_UTF8_CHECK|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|\
1198        PCRE_NOTEMPTY_ATSTART|PCRE_PARTIAL_SOFT|PCRE_PARTIAL_HARD)
1199    
1200  #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. */  
1201    
1202  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */
1203    
1204  /* Negative values for the firstchar and reqchar variables */  /* This variable is used to detect a loaded regular expression
1205    in different endianness. */
1206    
1207  #define REQ_UNSET (-2)  #define REVERSED_MAGIC_NUMBER  0x45524350UL   /* 'ERCP' */
 #define REQ_NONE  (-1)  
1208    
1209  /* 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
1210  req_byte match. */  req_byte match. */
1211    
1212  #define REQ_BYTE_MAX 1000  #define REQ_BYTE_MAX 1000
1213    
1214  /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a  /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
1215  variable-length repeat, or a anything other than literal characters. */  environments where these macros are defined elsewhere. Unfortunately, there
1216    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 */  
1217    
1218  typedef int BOOL;  typedef int BOOL;
1219    
1220    #ifndef FALSE
1221  #define FALSE   0  #define FALSE   0
1222  #define TRUE    1  #define TRUE    1
1223    #endif
1224    
1225    /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
1226    character constants like '*' because the compiler would emit their EBCDIC code,
1227    which is different from their ASCII/UTF-8 code. Instead we define macros for
1228    the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
1229    is enabled. When UTF-8 support is not enabled, the definitions use character
1230    literals. Both character and string versions of each character are needed, and
1231    there are some longer strings as well.
1232    
1233    This means that, on EBCDIC platforms, the PCRE library can handle either
1234    EBCDIC, or UTF-8, but not both. To support both in the same compiled library
1235    would need different lookups depending on whether PCRE_UTF8 was set or not.
1236    This would make it impossible to use characters in switch/case statements,
1237    which would reduce performance. For a theoretical use (which nobody has asked
1238    for) in a minority area (EBCDIC platforms), this is not sensible. Any
1239    application that did need both could compile two versions of the library, using
1240    macros to give the functions distinct names. */
1241    
1242    #ifndef SUPPORT_UTF
1243    
1244    /* UTF-8 support is not enabled; use the platform-dependent character literals
1245    so that PCRE works in both ASCII and EBCDIC environments, but only in non-UTF
1246    mode. Newline characters are problematic in EBCDIC. Though it has CR and LF
1247    characters, a common practice has been to use its NL (0x15) character as the
1248    line terminator in C-like processing environments. However, sometimes the LF
1249    (0x25) character is used instead, according to this Unicode document:
1250    
1251    http://unicode.org/standard/reports/tr13/tr13-5.html
1252    
1253    PCRE defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25
1254    instead. Whichever is *not* chosen is defined as NEL.
1255    
1256    In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the
1257    same code point. */
1258    
1259    #ifdef EBCDIC
1260    
1261    #ifndef EBCDIC_NL25
1262    #define CHAR_NL                     '\x15'
1263    #define CHAR_NEL                    '\x25'
1264    #define STR_NL                      "\x15"
1265    #define STR_NEL                     "\x25"
1266    #else
1267    #define CHAR_NL                     '\x25'
1268    #define CHAR_NEL                    '\x15'
1269    #define STR_NL                      "\x25"
1270    #define STR_NEL                     "\x15"
1271    #endif
1272    
1273    #define CHAR_LF                     CHAR_NL
1274    #define STR_LF                      STR_NL
1275    
1276    #define CHAR_ESC                    '\047'
1277    #define CHAR_DEL                    '\007'
1278    #define STR_ESC                     "\047"
1279    #define STR_DEL                     "\007"
1280    
1281    #else  /* Not EBCDIC */
1282    
1283    /* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for
1284    compatibility. NEL is the Unicode newline character; make sure it is
1285    a positive value. */
1286    
1287    #define CHAR_LF                     '\n'
1288    #define CHAR_NL                     CHAR_LF
1289    #define CHAR_NEL                    ((unsigned char)'\x85')
1290    #define CHAR_ESC                    '\033'
1291    #define CHAR_DEL                    '\177'
1292    
1293    #define STR_LF                      "\n"
1294    #define STR_NL                      STR_LF
1295    #define STR_NEL                     "\x85"
1296    #define STR_ESC                     "\033"
1297    #define STR_DEL                     "\177"
1298    
1299    #endif  /* EBCDIC */
1300    
1301    /* The remaining definitions work in both environments. */
1302    
1303    #define CHAR_NULL                   '\0'
1304    #define CHAR_HT                     '\t'
1305    #define CHAR_VT                     '\v'
1306    #define CHAR_FF                     '\f'
1307    #define CHAR_CR                     '\r'
1308    #define CHAR_BS                     '\b'
1309    #define CHAR_BEL                    '\a'
1310    
1311    #define CHAR_SPACE                  ' '
1312    #define CHAR_EXCLAMATION_MARK       '!'
1313    #define CHAR_QUOTATION_MARK         '"'
1314    #define CHAR_NUMBER_SIGN            '#'
1315    #define CHAR_DOLLAR_SIGN            '$'
1316    #define CHAR_PERCENT_SIGN           '%'
1317    #define CHAR_AMPERSAND              '&'
1318    #define CHAR_APOSTROPHE             '\''
1319    #define CHAR_LEFT_PARENTHESIS       '('
1320    #define CHAR_RIGHT_PARENTHESIS      ')'
1321    #define CHAR_ASTERISK               '*'
1322    #define CHAR_PLUS                   '+'
1323    #define CHAR_COMMA                  ','
1324    #define CHAR_MINUS                  '-'
1325    #define CHAR_DOT                    '.'
1326    #define CHAR_SLASH                  '/'
1327    #define CHAR_0                      '0'
1328    #define CHAR_1                      '1'
1329    #define CHAR_2                      '2'
1330    #define CHAR_3                      '3'
1331    #define CHAR_4                      '4'
1332    #define CHAR_5                      '5'
1333    #define CHAR_6                      '6'
1334    #define CHAR_7                      '7'
1335    #define CHAR_8                      '8'
1336    #define CHAR_9                      '9'
1337    #define CHAR_COLON                  ':'
1338    #define CHAR_SEMICOLON              ';'
1339    #define CHAR_LESS_THAN_SIGN         '<'
1340    #define CHAR_EQUALS_SIGN            '='
1341    #define CHAR_GREATER_THAN_SIGN      '>'
1342    #define CHAR_QUESTION_MARK          '?'
1343    #define CHAR_COMMERCIAL_AT          '@'
1344    #define CHAR_A                      'A'
1345    #define CHAR_B                      'B'
1346    #define CHAR_C                      'C'
1347    #define CHAR_D                      'D'
1348    #define CHAR_E                      'E'
1349    #define CHAR_F                      'F'
1350    #define CHAR_G                      'G'
1351    #define CHAR_H                      'H'
1352    #define CHAR_I                      'I'
1353    #define CHAR_J                      'J'
1354    #define CHAR_K                      'K'
1355    #define CHAR_L                      'L'
1356    #define CHAR_M                      'M'
1357    #define CHAR_N                      'N'
1358    #define CHAR_O                      'O'
1359    #define CHAR_P                      'P'
1360    #define CHAR_Q                      'Q'
1361    #define CHAR_R                      'R'
1362    #define CHAR_S                      'S'
1363    #define CHAR_T                      'T'
1364    #define CHAR_U                      'U'
1365    #define CHAR_V                      'V'
1366    #define CHAR_W                      'W'
1367    #define CHAR_X                      'X'
1368    #define CHAR_Y                      'Y'
1369    #define CHAR_Z                      'Z'
1370    #define CHAR_LEFT_SQUARE_BRACKET    '['
1371    #define CHAR_BACKSLASH              '\\'
1372    #define CHAR_RIGHT_SQUARE_BRACKET   ']'
1373    #define CHAR_CIRCUMFLEX_ACCENT      '^'
1374    #define CHAR_UNDERSCORE             '_'
1375    #define CHAR_GRAVE_ACCENT           '`'
1376    #define CHAR_a                      'a'
1377    #define CHAR_b                      'b'
1378    #define CHAR_c                      'c'
1379    #define CHAR_d                      'd'
1380    #define CHAR_e                      'e'
1381    #define CHAR_f                      'f'
1382    #define CHAR_g                      'g'
1383    #define CHAR_h                      'h'
1384    #define CHAR_i                      'i'
1385    #define CHAR_j                      'j'
1386    #define CHAR_k                      'k'
1387    #define CHAR_l                      'l'
1388    #define CHAR_m                      'm'
1389    #define CHAR_n                      'n'
1390    #define CHAR_o                      'o'
1391    #define CHAR_p                      'p'
1392    #define CHAR_q                      'q'
1393    #define CHAR_r                      'r'
1394    #define CHAR_s                      's'
1395    #define CHAR_t                      't'
1396    #define CHAR_u                      'u'
1397    #define CHAR_v                      'v'
1398    #define CHAR_w                      'w'
1399    #define CHAR_x                      'x'
1400    #define CHAR_y                      'y'
1401    #define CHAR_z                      'z'
1402    #define CHAR_LEFT_CURLY_BRACKET     '{'
1403    #define CHAR_VERTICAL_LINE          '|'
1404    #define CHAR_RIGHT_CURLY_BRACKET    '}'
1405    #define CHAR_TILDE                  '~'
1406    
1407    #define STR_HT                      "\t"
1408    #define STR_VT                      "\v"
1409    #define STR_FF                      "\f"
1410    #define STR_CR                      "\r"
1411    #define STR_BS                      "\b"
1412    #define STR_BEL                     "\a"
1413    
1414    #define STR_SPACE                   " "
1415    #define STR_EXCLAMATION_MARK        "!"
1416    #define STR_QUOTATION_MARK          "\""
1417    #define STR_NUMBER_SIGN             "#"
1418    #define STR_DOLLAR_SIGN             "$"
1419    #define STR_PERCENT_SIGN            "%"
1420    #define STR_AMPERSAND               "&"
1421    #define STR_APOSTROPHE              "'"
1422    #define STR_LEFT_PARENTHESIS        "("
1423    #define STR_RIGHT_PARENTHESIS       ")"
1424    #define STR_ASTERISK                "*"
1425    #define STR_PLUS                    "+"
1426    #define STR_COMMA                   ","
1427    #define STR_MINUS                   "-"
1428    #define STR_DOT                     "."
1429    #define STR_SLASH                   "/"
1430    #define STR_0                       "0"
1431    #define STR_1                       "1"
1432    #define STR_2                       "2"
1433    #define STR_3                       "3"
1434    #define STR_4                       "4"
1435    #define STR_5                       "5"
1436    #define STR_6                       "6"
1437    #define STR_7                       "7"
1438    #define STR_8                       "8"
1439    #define STR_9                       "9"
1440    #define STR_COLON                   ":"
1441    #define STR_SEMICOLON               ";"
1442    #define STR_LESS_THAN_SIGN          "<"
1443    #define STR_EQUALS_SIGN             "="
1444    #define STR_GREATER_THAN_SIGN       ">"
1445    #define STR_QUESTION_MARK           "?"
1446    #define STR_COMMERCIAL_AT           "@"
1447    #define STR_A                       "A"
1448    #define STR_B                       "B"
1449    #define STR_C                       "C"
1450    #define STR_D                       "D"
1451    #define STR_E                       "E"
1452    #define STR_F                       "F"
1453    #define STR_G                       "G"
1454    #define STR_H                       "H"
1455    #define STR_I                       "I"
1456    #define STR_J                       "J"
1457    #define STR_K                       "K"
1458    #define STR_L                       "L"
1459    #define STR_M                       "M"
1460    #define STR_N                       "N"
1461    #define STR_O                       "O"
1462    #define STR_P                       "P"
1463    #define STR_Q                       "Q"
1464    #define STR_R                       "R"
1465    #define STR_S                       "S"
1466    #define STR_T                       "T"
1467    #define STR_U                       "U"
1468    #define STR_V                       "V"
1469    #define STR_W                       "W"
1470    #define STR_X                       "X"
1471    #define STR_Y                       "Y"
1472    #define STR_Z                       "Z"
1473    #define STR_LEFT_SQUARE_BRACKET     "["
1474    #define STR_BACKSLASH               "\\"
1475    #define STR_RIGHT_SQUARE_BRACKET    "]"
1476    #define STR_CIRCUMFLEX_ACCENT       "^"
1477    #define STR_UNDERSCORE              "_"
1478    #define STR_GRAVE_ACCENT            "`"
1479    #define STR_a                       "a"
1480    #define STR_b                       "b"
1481    #define STR_c                       "c"
1482    #define STR_d                       "d"
1483    #define STR_e                       "e"
1484    #define STR_f                       "f"
1485    #define STR_g                       "g"
1486    #define STR_h                       "h"
1487    #define STR_i                       "i"
1488    #define STR_j                       "j"
1489    #define STR_k                       "k"
1490    #define STR_l                       "l"
1491    #define STR_m                       "m"
1492    #define STR_n                       "n"
1493    #define STR_o                       "o"
1494    #define STR_p                       "p"
1495    #define STR_q                       "q"
1496    #define STR_r                       "r"
1497    #define STR_s                       "s"
1498    #define STR_t                       "t"
1499    #define STR_u                       "u"
1500    #define STR_v                       "v"
1501    #define STR_w                       "w"
1502    #define STR_x                       "x"
1503    #define STR_y                       "y"
1504    #define STR_z                       "z"
1505    #define STR_LEFT_CURLY_BRACKET      "{"
1506    #define STR_VERTICAL_LINE           "|"
1507    #define STR_RIGHT_CURLY_BRACKET     "}"
1508    #define STR_TILDE                   "~"
1509    
1510    #define STRING_ACCEPT0              "ACCEPT\0"
1511    #define STRING_COMMIT0              "COMMIT\0"
1512    #define STRING_F0                   "F\0"
1513    #define STRING_FAIL0                "FAIL\0"
1514    #define STRING_MARK0                "MARK\0"
1515    #define STRING_PRUNE0               "PRUNE\0"
1516    #define STRING_SKIP0                "SKIP\0"
1517    #define STRING_THEN                 "THEN"
1518    
1519    #define STRING_alpha0               "alpha\0"
1520    #define STRING_lower0               "lower\0"
1521    #define STRING_upper0               "upper\0"
1522    #define STRING_alnum0               "alnum\0"
1523    #define STRING_ascii0               "ascii\0"
1524    #define STRING_blank0               "blank\0"
1525    #define STRING_cntrl0               "cntrl\0"
1526    #define STRING_digit0               "digit\0"
1527    #define STRING_graph0               "graph\0"
1528    #define STRING_print0               "print\0"
1529    #define STRING_punct0               "punct\0"
1530    #define STRING_space0               "space\0"
1531    #define STRING_word0                "word\0"
1532    #define STRING_xdigit               "xdigit"
1533    
1534    #define STRING_DEFINE               "DEFINE"
1535    
1536    #define STRING_CR_RIGHTPAR             "CR)"
1537    #define STRING_LF_RIGHTPAR             "LF)"
1538    #define STRING_CRLF_RIGHTPAR           "CRLF)"
1539    #define STRING_ANY_RIGHTPAR            "ANY)"
1540    #define STRING_ANYCRLF_RIGHTPAR        "ANYCRLF)"
1541    #define STRING_BSR_ANYCRLF_RIGHTPAR    "BSR_ANYCRLF)"
1542    #define STRING_BSR_UNICODE_RIGHTPAR    "BSR_UNICODE)"
1543    #ifdef COMPILE_PCRE8
1544    #define STRING_UTF_RIGHTPAR            "UTF8)"
1545    #endif
1546    #ifdef COMPILE_PCRE16
1547    #define STRING_UTF_RIGHTPAR            "UTF16)"
1548    #endif
1549    #ifdef COMPILE_PCRE32
1550    #define STRING_UTF_RIGHTPAR            "UTF32)"
1551    #endif
1552    #define STRING_UCP_RIGHTPAR            "UCP)"
1553    #define STRING_NO_START_OPT_RIGHTPAR   "NO_START_OPT)"
1554    
1555    #else  /* SUPPORT_UTF */
1556    
1557    /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1558    works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1559    only. */
1560    
1561    #define CHAR_HT                     '\011'
1562    #define CHAR_VT                     '\013'
1563    #define CHAR_FF                     '\014'
1564    #define CHAR_CR                     '\015'
1565    #define CHAR_LF                     '\012'
1566    #define CHAR_NL                     CHAR_LF
1567    #define CHAR_NEL                    ((unsigned char)'\x85')
1568    #define CHAR_BS                     '\010'
1569    #define CHAR_BEL                    '\007'
1570    #define CHAR_ESC                    '\033'
1571    #define CHAR_DEL                    '\177'
1572    
1573    #define CHAR_NULL                   '\0'
1574    #define CHAR_SPACE                  '\040'
1575    #define CHAR_EXCLAMATION_MARK       '\041'
1576    #define CHAR_QUOTATION_MARK         '\042'
1577    #define CHAR_NUMBER_SIGN            '\043'
1578    #define CHAR_DOLLAR_SIGN            '\044'
1579    #define CHAR_PERCENT_SIGN           '\045'
1580    #define CHAR_AMPERSAND              '\046'
1581    #define CHAR_APOSTROPHE             '\047'
1582    #define CHAR_LEFT_PARENTHESIS       '\050'
1583    #define CHAR_RIGHT_PARENTHESIS      '\051'
1584    #define CHAR_ASTERISK               '\052'
1585    #define CHAR_PLUS                   '\053'
1586    #define CHAR_COMMA                  '\054'
1587    #define CHAR_MINUS                  '\055'
1588    #define CHAR_DOT                    '\056'
1589    #define CHAR_SLASH                  '\057'
1590    #define CHAR_0                      '\060'
1591    #define CHAR_1                      '\061'
1592    #define CHAR_2                      '\062'
1593    #define CHAR_3                      '\063'
1594    #define CHAR_4                      '\064'
1595    #define CHAR_5                      '\065'
1596    #define CHAR_6                      '\066'
1597    #define CHAR_7                      '\067'
1598    #define CHAR_8                      '\070'
1599    #define CHAR_9                      '\071'
1600    #define CHAR_COLON                  '\072'
1601    #define CHAR_SEMICOLON              '\073'
1602    #define CHAR_LESS_THAN_SIGN         '\074'
1603    #define CHAR_EQUALS_SIGN            '\075'
1604    #define CHAR_GREATER_THAN_SIGN      '\076'
1605    #define CHAR_QUESTION_MARK          '\077'
1606    #define CHAR_COMMERCIAL_AT          '\100'
1607    #define CHAR_A                      '\101'
1608    #define CHAR_B                      '\102'
1609    #define CHAR_C                      '\103'
1610    #define CHAR_D                      '\104'
1611    #define CHAR_E                      '\105'
1612    #define CHAR_F                      '\106'
1613    #define CHAR_G                      '\107'
1614    #define CHAR_H                      '\110'
1615    #define CHAR_I                      '\111'
1616    #define CHAR_J                      '\112'
1617    #define CHAR_K                      '\113'
1618    #define CHAR_L                      '\114'
1619    #define CHAR_M                      '\115'
1620    #define CHAR_N                      '\116'
1621    #define CHAR_O                      '\117'
1622    #define CHAR_P                      '\120'
1623    #define CHAR_Q                      '\121'
1624    #define CHAR_R                      '\122'
1625    #define CHAR_S                      '\123'
1626    #define CHAR_T                      '\124'
1627    #define CHAR_U                      '\125'
1628    #define CHAR_V                      '\126'
1629    #define CHAR_W                      '\127'
1630    #define CHAR_X                      '\130'
1631    #define CHAR_Y                      '\131'
1632    #define CHAR_Z                      '\132'
1633    #define CHAR_LEFT_SQUARE_BRACKET    '\133'
1634    #define CHAR_BACKSLASH              '\134'
1635    #define CHAR_RIGHT_SQUARE_BRACKET   '\135'
1636    #define CHAR_CIRCUMFLEX_ACCENT      '\136'
1637    #define CHAR_UNDERSCORE             '\137'
1638    #define CHAR_GRAVE_ACCENT           '\140'
1639    #define CHAR_a                      '\141'
1640    #define CHAR_b                      '\142'
1641    #define CHAR_c                      '\143'
1642    #define CHAR_d                      '\144'
1643    #define CHAR_e                      '\145'
1644    #define CHAR_f                      '\146'
1645    #define CHAR_g                      '\147'
1646    #define CHAR_h                      '\150'
1647    #define CHAR_i                      '\151'
1648    #define CHAR_j                      '\152'
1649    #define CHAR_k                      '\153'
1650    #define CHAR_l                      '\154'
1651    #define CHAR_m                      '\155'
1652    #define CHAR_n                      '\156'
1653    #define CHAR_o                      '\157'
1654    #define CHAR_p                      '\160'
1655    #define CHAR_q                      '\161'
1656    #define CHAR_r                      '\162'
1657    #define CHAR_s                      '\163'
1658    #define CHAR_t                      '\164'
1659    #define CHAR_u                      '\165'
1660    #define CHAR_v                      '\166'
1661    #define CHAR_w                      '\167'
1662    #define CHAR_x                      '\170'
1663    #define CHAR_y                      '\171'
1664    #define CHAR_z                      '\172'
1665    #define CHAR_LEFT_CURLY_BRACKET     '\173'
1666    #define CHAR_VERTICAL_LINE          '\174'
1667    #define CHAR_RIGHT_CURLY_BRACKET    '\175'
1668    #define CHAR_TILDE                  '\176'
1669    
1670    #define STR_HT                      "\011"
1671    #define STR_VT                      "\013"
1672    #define STR_FF                      "\014"
1673    #define STR_CR                      "\015"
1674    #define STR_NL                      "\012"
1675    #define STR_BS                      "\010"
1676    #define STR_BEL                     "\007"
1677    #define STR_ESC                     "\033"
1678    #define STR_DEL                     "\177"
1679    
1680    #define STR_SPACE                   "\040"
1681    #define STR_EXCLAMATION_MARK        "\041"
1682    #define STR_QUOTATION_MARK          "\042"
1683    #define STR_NUMBER_SIGN             "\043"
1684    #define STR_DOLLAR_SIGN             "\044"
1685    #define STR_PERCENT_SIGN            "\045"
1686    #define STR_AMPERSAND               "\046"
1687    #define STR_APOSTROPHE              "\047"
1688    #define STR_LEFT_PARENTHESIS        "\050"
1689    #define STR_RIGHT_PARENTHESIS       "\051"
1690    #define STR_ASTERISK                "\052"
1691    #define STR_PLUS                    "\053"
1692    #define STR_COMMA                   "\054"
1693    #define STR_MINUS                   "\055"
1694    #define STR_DOT                     "\056"
1695    #define STR_SLASH                   "\057"
1696    #define STR_0                       "\060"
1697    #define STR_1                       "\061"
1698    #define STR_2                       "\062"
1699    #define STR_3                       "\063"
1700    #define STR_4                       "\064"
1701    #define STR_5                       "\065"
1702    #define STR_6                       "\066"
1703    #define STR_7                       "\067"
1704    #define STR_8                       "\070"
1705    #define STR_9                       "\071"
1706    #define STR_COLON                   "\072"
1707    #define STR_SEMICOLON               "\073"
1708    #define STR_LESS_THAN_SIGN          "\074"
1709    #define STR_EQUALS_SIGN             "\075"
1710    #define STR_GREATER_THAN_SIGN       "\076"
1711    #define STR_QUESTION_MARK           "\077"
1712    #define STR_COMMERCIAL_AT           "\100"
1713    #define STR_A                       "\101"
1714    #define STR_B                       "\102"
1715    #define STR_C                       "\103"
1716    #define STR_D                       "\104"
1717    #define STR_E                       "\105"
1718    #define STR_F                       "\106"
1719    #define STR_G                       "\107"
1720    #define STR_H                       "\110"
1721    #define STR_I                       "\111"
1722    #define STR_J                       "\112"
1723    #define STR_K                       "\113"
1724    #define STR_L                       "\114"
1725    #define STR_M                       "\115"
1726    #define STR_N                       "\116"
1727    #define STR_O                       "\117"
1728    #define STR_P                       "\120"
1729    #define STR_Q                       "\121"
1730    #define STR_R                       "\122"
1731    #define STR_S                       "\123"
1732    #define STR_T                       "\124"
1733    #define STR_U                       "\125"
1734    #define STR_V                       "\126"
1735    #define STR_W                       "\127"
1736    #define STR_X                       "\130"
1737    #define STR_Y                       "\131"
1738    #define STR_Z                       "\132"
1739    #define STR_LEFT_SQUARE_BRACKET     "\133"
1740    #define STR_BACKSLASH               "\134"
1741    #define STR_RIGHT_SQUARE_BRACKET    "\135"
1742    #define STR_CIRCUMFLEX_ACCENT       "\136"
1743    #define STR_UNDERSCORE              "\137"
1744    #define STR_GRAVE_ACCENT            "\140"
1745    #define STR_a                       "\141"
1746    #define STR_b                       "\142"
1747    #define STR_c                       "\143"
1748    #define STR_d                       "\144"
1749    #define STR_e                       "\145"
1750    #define STR_f                       "\146"
1751    #define STR_g                       "\147"
1752    #define STR_h                       "\150"
1753    #define STR_i                       "\151"
1754    #define STR_j                       "\152"
1755    #define STR_k                       "\153"
1756    #define STR_l                       "\154"
1757    #define STR_m                       "\155"
1758    #define STR_n                       "\156"
1759    #define STR_o                       "\157"
1760    #define STR_p                       "\160"
1761    #define STR_q                       "\161"
1762    #define STR_r                       "\162"
1763    #define STR_s                       "\163"
1764    #define STR_t                       "\164"
1765    #define STR_u                       "\165"
1766    #define STR_v                       "\166"
1767    #define STR_w                       "\167"
1768    #define STR_x                       "\170"
1769    #define STR_y                       "\171"
1770    #define STR_z                       "\172"
1771    #define STR_LEFT_CURLY_BRACKET      "\173"
1772    #define STR_VERTICAL_LINE           "\174"
1773    #define STR_RIGHT_CURLY_BRACKET     "\175"
1774    #define STR_TILDE                   "\176"
1775    
1776    #define STRING_ACCEPT0              STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1777    #define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1778    #define STRING_F0                   STR_F "\0"
1779    #define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"
1780    #define STRING_MARK0                STR_M STR_A STR_R STR_K "\0"
1781    #define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"
1782    #define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"
1783    #define STRING_THEN                 STR_T STR_H STR_E STR_N
1784    
1785    #define STRING_alpha0               STR_a STR_l STR_p STR_h STR_a "\0"
1786    #define STRING_lower0               STR_l STR_o STR_w STR_e STR_r "\0"
1787    #define STRING_upper0               STR_u STR_p STR_p STR_e STR_r "\0"
1788    #define STRING_alnum0               STR_a STR_l STR_n STR_u STR_m "\0"
1789    #define STRING_ascii0               STR_a STR_s STR_c STR_i STR_i "\0"
1790    #define STRING_blank0               STR_b STR_l STR_a STR_n STR_k "\0"
1791    #define STRING_cntrl0               STR_c STR_n STR_t STR_r STR_l "\0"
1792    #define STRING_digit0               STR_d STR_i STR_g STR_i STR_t "\0"
1793    #define STRING_graph0               STR_g STR_r STR_a STR_p STR_h "\0"
1794    #define STRING_print0               STR_p STR_r STR_i STR_n STR_t "\0"
1795    #define STRING_punct0               STR_p STR_u STR_n STR_c STR_t "\0"
1796    #define STRING_space0               STR_s STR_p STR_a STR_c STR_e "\0"
1797    #define STRING_word0                STR_w STR_o STR_r STR_d       "\0"
1798    #define STRING_xdigit               STR_x STR_d STR_i STR_g STR_i STR_t
1799    
1800    #define STRING_DEFINE               STR_D STR_E STR_F STR_I STR_N STR_E
1801    
1802    #define STRING_CR_RIGHTPAR             STR_C STR_R STR_RIGHT_PARENTHESIS
1803    #define STRING_LF_RIGHTPAR             STR_L STR_F STR_RIGHT_PARENTHESIS
1804    #define STRING_CRLF_RIGHTPAR           STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1805    #define STRING_ANY_RIGHTPAR            STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1806    #define STRING_ANYCRLF_RIGHTPAR        STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1807    #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
1808    #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
1809    #ifdef COMPILE_PCRE8
1810    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1811    #endif
1812    #ifdef COMPILE_PCRE16
1813    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
1814    #endif
1815    #ifdef COMPILE_PCRE32
1816    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS
1817    #endif
1818    #define STRING_UCP_RIGHTPAR            STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1819    #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
1820    
1821    #endif  /* SUPPORT_UTF */
1822    
1823  /* Escape items that are just an encoding of a particular data value. */  /* Escape items that are just an encoding of a particular data value. */
1824    
1825  #ifndef ESC_e  #ifndef ESC_e
1826  #define ESC_e 27  #define ESC_e CHAR_ESC
1827  #endif  #endif
1828    
1829  #ifndef ESC_f  #ifndef ESC_f
1830  #define ESC_f '\f'  #define ESC_f CHAR_FF
1831  #endif  #endif
1832    
1833  #ifndef ESC_n  #ifndef ESC_n
1834  #define ESC_n '\n'  #define ESC_n CHAR_LF
1835  #endif  #endif
1836    
1837  #ifndef ESC_r  #ifndef ESC_r
1838  #define ESC_r '\r'  #define ESC_r CHAR_CR
1839  #endif  #endif
1840    
1841  /* 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
1842  (presumably because of all the others like size_t). */  (presumably because of all the others like size_t). */
1843    
1844  #ifndef ESC_tee  #ifndef ESC_tee
1845  #define ESC_tee '\t'  #define ESC_tee CHAR_HT
1846  #endif  #endif
1847    
1848  /* Codes for different types of Unicode property */  /* Codes for different types of Unicode property */
1849    
1850  #define PT_ANY        0    /* Any property - matches all chars */  #define PT_ANY        0    /* Any property - matches all chars */
1851  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */
1852  #define PT_GC         2    /* General characteristic (e.g. L) */  #define PT_GC         2    /* Specified general characteristic (e.g. L) */
1853  #define PT_PC         3    /* Particular characteristic (e.g. Lu) */  #define PT_PC         3    /* Specified particular characteristic (e.g. Lu) */
1854  #define PT_SC         4    /* Script (e.g. Han) */  #define PT_SC         4    /* Script (e.g. Han) */
1855    #define PT_ALNUM      5    /* Alphanumeric - the union of L and N */
1856    #define PT_SPACE      6    /* Perl space - Z plus 9,10,12,13 */
1857    #define PT_PXSPACE    7    /* POSIX space - Z plus 9,10,11,12,13 */
1858    #define PT_WORD       8    /* Word - L plus N plus underscore */
1859    #define PT_CLIST      9    /* Pseudo-property: match character list */
1860    
1861  /* 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
1862  contain UTF-8 characters with values greater than 255. */  contain characters with values greater than 255. */
1863    
1864  #define XCL_NOT    0x01    /* Flag: this is a negative class */  #define XCL_NOT    0x01    /* Flag: this is a negative class */
1865  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */
# Line 542  contain UTF-8 characters with values gre Line 1872  contain UTF-8 characters with values gre
1872    
1873  /* 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
1874  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
1875  their negation. Also, they must appear in the same order as in the opcode  0 for a data character.  Also, they must appear in the same order as in the opcode
1876  definitions below, up to ESC_z. There's a dummy for OP_ANY because it  definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
1877  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
1878  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
1879  There are two tests in the code for an escape greater than ESC_b and less than  non-DOTALL mode, "." behaves like \N.
1880  ESC_Z to detect the types that may be repeated. These are the types that  
1881  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.
1882  character, that code will have to change. */  when PCRE_UCP is set and replacement of \d etc by \p sequences is required.
1883    They must be contiguous, and remain in order so that the replacements can be
1884  enum { ESC_A = 1, ESC_G, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s, ESC_W,  looked up from a table.
1885         ESC_w, ESC_dum1, ESC_C, ESC_P, ESC_p, ESC_R, ESC_X, ESC_Z, ESC_z,  
1886         ESC_E, ESC_Q, ESC_k, ESC_REF };  Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in
1887    check_escape(). There are two tests in the code for an escape
1888    greater than ESC_b and less than ESC_Z to detect the types that may be
1889    repeated. These are the types that consume characters. If any new escapes are
1890    put in between that don't consume a character, that code will have to change.
1891    */
1892    
1893    enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1894           ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1895           ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
1896           ESC_E, ESC_Q, ESC_g, ESC_k,
1897           ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu };
1898    
1899  /* Opcode table: OP_BRA must be last, as all values >= it are used for brackets  /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
 that extract substrings. Starting from 1 (i.e. after OP_END), the values up to  
1900  OP_EOD must correspond in order to the list of escapes immediately above.  OP_EOD must correspond in order to the list of escapes immediately above.
1901    
 To keep stored, compiled patterns compatible, new opcodes should be added  
 immediately before OP_BRA, where (since release 7.0) a gap is left for this  
 purpose.  
   
1902  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1903  that follow must also be updated to match. There is also a table called  that follow must also be updated to match. There are also tables called
1904  "coptable" in pcre_dfa_exec.c that must be updated. */  "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1905    
1906  enum {  enum {
1907    OP_END,            /* 0 End of pattern */    OP_END,            /* 0 End of pattern */
# Line 575  enum { Line 1910  enum {
1910    
1911    OP_SOD,            /* 1 Start of data: \A */    OP_SOD,            /* 1 Start of data: \A */
1912    OP_SOM,            /* 2 Start of match (subject + offset): \G */    OP_SOM,            /* 2 Start of match (subject + offset): \G */
1913    OP_NOT_WORD_BOUNDARY,  /*  3 \B */    OP_SET_SOM,        /* 3 Set start of match (\K) */
1914    OP_WORD_BOUNDARY,      /*  4 \b */    OP_NOT_WORD_BOUNDARY,  /*  4 \B */
1915    OP_NOT_DIGIT,          /*  5 \D */    OP_WORD_BOUNDARY,      /*  5 \b */
1916    OP_DIGIT,              /*  6 \d */    OP_NOT_DIGIT,          /*  6 \D */
1917    OP_NOT_WHITESPACE,     /*  7 \S */    OP_DIGIT,              /*  7 \d */
1918    OP_WHITESPACE,         /*  8 \s */    OP_NOT_WHITESPACE,     /*  8 \S */
1919    OP_NOT_WORDCHAR,       /*  9 \W */    OP_WHITESPACE,         /*  9 \s */
1920    OP_WORDCHAR,           /* 10 \w */    OP_NOT_WORDCHAR,       /* 10 \W */
1921    OP_ANY,            /* 11 Match any character */    OP_WORDCHAR,           /* 11 \w */
1922    OP_ANYBYTE,        /* 12 Match any byte (\C); different to OP_ANY for UTF-8 */  
1923    OP_NOTPROP,        /* 13 \P (not Unicode property) */    OP_ANY,            /* 12 Match any character except newline (\N) */
1924    OP_PROP,           /* 14 \p (Unicode property) */    OP_ALLANY,         /* 13 Match any character */
1925    OP_ANYNL,          /* 15 \R (any newline sequence) */    OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1926    OP_EXTUNI,         /* 16 \X (extended Unicode sequence */    OP_NOTPROP,        /* 15 \P (not Unicode property) */
1927    OP_EODN,           /* 17 End of data or \n at end of data: \Z. */    OP_PROP,           /* 16 \p (Unicode property) */
1928    OP_EOD,            /* 18 End of data: \z */    OP_ANYNL,          /* 17 \R (any newline sequence) */
1929      OP_NOT_HSPACE,     /* 18 \H (not horizontal whitespace) */
1930    OP_OPT,            /* 19 Set runtime options */    OP_HSPACE,         /* 19 \h (horizontal whitespace) */
1931    OP_CIRC,           /* 20 Start of line - varies with multiline switch */    OP_NOT_VSPACE,     /* 20 \V (not vertical whitespace) */
1932    OP_DOLL,           /* 21 End of line - varies with multiline switch */    OP_VSPACE,         /* 21 \v (vertical whitespace) */
1933    OP_CHAR,           /* 22 Match one character, casefully */    OP_EXTUNI,         /* 22 \X (extended Unicode sequence */
1934    OP_CHARNC,         /* 23 Match one character, caselessly */    OP_EODN,           /* 23 End of data or \n at end of data (\Z) */
1935    OP_NOT,            /* 24 Match one character, not the following one */    OP_EOD,            /* 24 End of data (\z) */
1936    
1937    OP_STAR,           /* 25 The maximizing and minimizing versions of */    OP_CIRC,           /* 25 Start of line - not multiline */
1938    OP_MINSTAR,        /* 26 these six opcodes must come in pairs, with */    OP_CIRCM,          /* 26 Start of line - multiline */
1939    OP_PLUS,           /* 27 the minimizing one second. */    OP_DOLL,           /* 27 End of line - not multiline */
1940    OP_MINPLUS,        /* 28 This first set applies to single characters.*/    OP_DOLLM,          /* 28 End of line - multiline */
1941    OP_QUERY,          /* 29 */    OP_CHAR,           /* 29 Match one character, casefully */
1942    OP_MINQUERY,       /* 30 */    OP_CHARI,          /* 30 Match one character, caselessly */
1943      OP_NOT,            /* 31 Match one character, not the given one, casefully */
1944    OP_UPTO,           /* 31 From 0 to n matches */    OP_NOTI,           /* 32 Match one character, not the given one, caselessly */
1945    OP_MINUPTO,        /* 32 */  
1946    OP_EXACT,          /* 33 Exactly n matches */    /* The following sets of 13 opcodes must always be kept in step because
1947      the offset from the first one is used to generate the others. */
1948    OP_POSSTAR,        /* 34 Possessified star */  
1949    OP_POSPLUS,        /* 35 Possessified plus */    /**** Single characters, caseful, must precede the caseless ones ****/
1950    OP_POSQUERY,       /* 36 Posesssified query */  
1951    OP_POSUPTO,        /* 37 Possessified upto */    OP_STAR,           /* 33 The maximizing and minimizing versions of */
1952      OP_MINSTAR,        /* 34 these six opcodes must come in pairs, with */
1953    OP_NOTSTAR,        /* 38 The maximizing and minimizing versions of */    OP_PLUS,           /* 35 the minimizing one second. */
1954    OP_NOTMINSTAR,     /* 39 these six opcodes must come in pairs, with */    OP_MINPLUS,        /* 36 */
1955    OP_NOTPLUS,        /* 40 the minimizing one second. They must be in */    OP_QUERY,          /* 37 */
1956    OP_NOTMINPLUS,     /* 41 exactly the same order as those above. */    OP_MINQUERY,       /* 38 */
1957    OP_NOTQUERY,       /* 42 This set applies to "not" single characters. */  
1958    OP_NOTMINQUERY,    /* 43 */    OP_UPTO,           /* 39 From 0 to n matches of one character, caseful*/
1959      OP_MINUPTO,        /* 40 */
1960    OP_NOTUPTO,        /* 44 From 0 to n matches */    OP_EXACT,          /* 41 Exactly n matches */
1961    OP_NOTMINUPTO,     /* 45 */  
1962    OP_NOTEXACT,       /* 46 Exactly n matches */    OP_POSSTAR,        /* 42 Possessified star, caseful */
1963      OP_POSPLUS,        /* 43 Possessified plus, caseful */
1964    OP_NOTPOSSTAR,     /* 47 Possessified versions */    OP_POSQUERY,       /* 44 Posesssified query, caseful */
1965    OP_NOTPOSPLUS,     /* 48 */    OP_POSUPTO,        /* 45 Possessified upto, caseful */
1966    OP_NOTPOSQUERY,    /* 49 */  
1967    OP_NOTPOSUPTO,     /* 50 */    /**** Single characters, caseless, must follow the caseful ones */
1968    
1969    OP_TYPESTAR,       /* 51 The maximizing and minimizing versions of */    OP_STARI,          /* 46 */
1970    OP_TYPEMINSTAR,    /* 52 these six opcodes must come in pairs, with */    OP_MINSTARI,       /* 47 */
1971    OP_TYPEPLUS,       /* 53 the minimizing one second. These codes must */    OP_PLUSI,          /* 48 */
1972    OP_TYPEMINPLUS,    /* 54 be in exactly the same order as those above. */    OP_MINPLUSI,       /* 49 */
1973    OP_TYPEQUERY,      /* 55 This set applies to character types such as \d */    OP_QUERYI,         /* 50 */
1974    OP_TYPEMINQUERY,   /* 56 */    OP_MINQUERYI,      /* 51 */
1975    
1976    OP_TYPEUPTO,       /* 57 From 0 to n matches */    OP_UPTOI,          /* 52 From 0 to n matches of one character, caseless */
1977    OP_TYPEMINUPTO,    /* 58 */    OP_MINUPTOI,       /* 53 */
1978    OP_TYPEEXACT,      /* 59 Exactly n matches */    OP_EXACTI,         /* 54 */
1979    
1980    OP_TYPEPOSSTAR,    /* 60 Possessified versions */    OP_POSSTARI,       /* 55 Possessified star, caseless */
1981    OP_TYPEPOSPLUS,    /* 61 */    OP_POSPLUSI,       /* 56 Possessified plus, caseless */
1982    OP_TYPEPOSQUERY,   /* 62 */    OP_POSQUERYI,      /* 57 Posesssified query, caseless */
1983    OP_TYPEPOSUPTO,    /* 63 */    OP_POSUPTOI,       /* 58 Possessified upto, caseless */
1984    
1985    OP_CRSTAR,         /* 64 The maximizing and minimizing versions of */    /**** The negated ones must follow the non-negated ones, and match them ****/
1986    OP_CRMINSTAR,      /* 65 all these opcodes must come in pairs, with */    /**** Negated single character, caseful; must precede the caseless ones ****/
1987    OP_CRPLUS,         /* 66 the minimizing one second. These codes must */  
1988    OP_CRMINPLUS,      /* 67 be in exactly the same order as those above. */    OP_NOTSTAR,        /* 59 The maximizing and minimizing versions of */
1989    OP_CRQUERY,        /* 68 These are for character classes and back refs */    OP_NOTMINSTAR,     /* 60 these six opcodes must come in pairs, with */
1990    OP_CRMINQUERY,     /* 69 */    OP_NOTPLUS,        /* 61 the minimizing one second. They must be in */
1991    OP_CRRANGE,        /* 70 These are different to the three sets above. */    OP_NOTMINPLUS,     /* 62 exactly the same order as those above. */
1992    OP_CRMINRANGE,     /* 71 */    OP_NOTQUERY,       /* 63 */
1993      OP_NOTMINQUERY,    /* 64 */
1994    OP_CLASS,          /* 72 Match a character class, chars < 256 only */  
1995    OP_NCLASS,         /* 73 Same, but the bitmap was created from a negative    OP_NOTUPTO,        /* 65 From 0 to n matches, caseful */
1996                             class - the difference is relevant only when a UTF-8    OP_NOTMINUPTO,     /* 66 */
1997                             character > 255 is encountered. */    OP_NOTEXACT,       /* 67 Exactly n matches */
1998    
1999    OP_XCLASS,         /* 74 Extended class for handling UTF-8 chars within the    OP_NOTPOSSTAR,     /* 68 Possessified versions, caseful */
2000                             class. This does both positive and negative. */    OP_NOTPOSPLUS,     /* 69 */
2001      OP_NOTPOSQUERY,    /* 70 */
2002    OP_REF,            /* 75 Match a back reference */    OP_NOTPOSUPTO,     /* 71 */
2003    OP_RECURSE,        /* 76 Match a numbered subpattern (possibly recursive) */  
2004    OP_CALLOUT,        /* 77 Call out to external function if provided */    /**** Negated single character, caseless; must follow the caseful ones ****/
2005    
2006    OP_ALT,            /* 78 Start of alternation */    OP_NOTSTARI,       /* 72 */
2007    OP_KET,            /* 79 End of group that doesn't have an unbounded repeat */    OP_NOTMINSTARI,    /* 73 */
2008    OP_KETRMAX,        /* 80 These two must remain together and in this */    OP_NOTPLUSI,       /* 74 */
2009    OP_KETRMIN,        /* 81 order. They are for groups the repeat for ever. */    OP_NOTMINPLUSI,    /* 75 */
2010      OP_NOTQUERYI,      /* 76 */
2011    /* The assertions must come before BRA, CBRA, ONCE, and COND.*/    OP_NOTMINQUERYI,   /* 77 */
2012    
2013    OP_ASSERT,         /* 82 Positive lookahead */    OP_NOTUPTOI,       /* 78 From 0 to n matches, caseless */
2014    OP_ASSERT_NOT,     /* 83 Negative lookahead */    OP_NOTMINUPTOI,    /* 79 */
2015    OP_ASSERTBACK,     /* 84 Positive lookbehind */    OP_NOTEXACTI,      /* 80 Exactly n matches */
2016    OP_ASSERTBACK_NOT, /* 85 Negative lookbehind */  
2017    OP_REVERSE,        /* 86 Move pointer back - used in lookbehind assertions */    OP_NOTPOSSTARI,    /* 81 Possessified versions, caseless */
2018      OP_NOTPOSPLUSI,    /* 82 */
2019    /* ONCE, BRA, CBRA, and COND must come after the assertions, with ONCE first,    OP_NOTPOSQUERYI,   /* 83 */
2020    as there's a test for >= ONCE for a subpattern that isn't an assertion. */    OP_NOTPOSUPTOI,    /* 84 */
2021    
2022    OP_ONCE,           /* 87 Atomic group */    /**** Character types ****/
2023    OP_BRA,            /* 88 Start of non-capturing bracket */  
2024    OP_CBRA,           /* 89 Start of capturing bracket */    OP_TYPESTAR,       /* 85 The maximizing and minimizing versions of */
2025    OP_COND,           /* 90 Conditional group */    OP_TYPEMINSTAR,    /* 86 these six opcodes must come in pairs, with */
2026      OP_TYPEPLUS,       /* 87 the minimizing one second. These codes must */
2027      OP_TYPEMINPLUS,    /* 88 be in exactly the same order as those above. */
2028      OP_TYPEQUERY,      /* 89 */
2029      OP_TYPEMINQUERY,   /* 90 */
2030    
2031      OP_TYPEUPTO,       /* 91 From 0 to n matches */
2032      OP_TYPEMINUPTO,    /* 92 */
2033      OP_TYPEEXACT,      /* 93 Exactly n matches */
2034    
2035      OP_TYPEPOSSTAR,    /* 94 Possessified versions */
2036      OP_TYPEPOSPLUS,    /* 95 */
2037      OP_TYPEPOSQUERY,   /* 96 */
2038      OP_TYPEPOSUPTO,    /* 97 */
2039    
2040      /* These are used for character classes and back references; only the
2041      first six are the same as the sets above. */
2042    
2043      OP_CRSTAR,         /* 98 The maximizing and minimizing versions of */
2044      OP_CRMINSTAR,      /* 99 all these opcodes must come in pairs, with */
2045      OP_CRPLUS,         /* 100 the minimizing one second. These codes must */
2046      OP_CRMINPLUS,      /* 101 be in exactly the same order as those above. */
2047      OP_CRQUERY,        /* 102 */
2048      OP_CRMINQUERY,     /* 103 */
2049    
2050      OP_CRRANGE,        /* 104 These are different to the three sets above. */
2051      OP_CRMINRANGE,     /* 105 */
2052    
2053      /* End of quantifier opcodes */
2054    
2055      OP_CLASS,          /* 106 Match a character class, chars < 256 only */
2056      OP_NCLASS,         /* 107 Same, but the bitmap was created from a negative
2057                                  class - the difference is relevant only when a
2058                                  character > 255 is encountered. */
2059      OP_XCLASS,         /* 108 Extended class for handling > 255 chars within the
2060                                  class. This does both positive and negative. */
2061      OP_REF,            /* 109 Match a back reference, casefully */
2062      OP_REFI,           /* 110 Match a back reference, caselessly */
2063      OP_RECURSE,        /* 111 Match a numbered subpattern (possibly recursive) */
2064      OP_CALLOUT,        /* 112 Call out to external function if provided */
2065    
2066      OP_ALT,            /* 113 Start of alternation */
2067      OP_KET,            /* 114 End of group that doesn't have an unbounded repeat */
2068      OP_KETRMAX,        /* 115 These two must remain together and in this */
2069      OP_KETRMIN,        /* 116 order. They are for groups the repeat for ever. */
2070      OP_KETRPOS,        /* 117 Possessive unlimited repeat. */
2071    
2072      /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
2073      asserts must remain in order. */
2074    
2075      OP_REVERSE,        /* 118 Move pointer back - used in lookbehind assertions */
2076      OP_ASSERT,         /* 119 Positive lookahead */
2077      OP_ASSERT_NOT,     /* 120 Negative lookahead */
2078      OP_ASSERTBACK,     /* 121 Positive lookbehind */
2079      OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
2080    
2081      /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
2082      after the assertions, with ONCE first, as there's a test for >= ONCE for a
2083      subpattern that isn't an assertion. The POS versions must immediately follow
2084      the non-POS versions in each case. */
2085    
2086      OP_ONCE,           /* 123 Atomic group, contains captures */
2087      OP_ONCE_NC,        /* 124 Atomic group containing no captures */
2088      OP_BRA,            /* 125 Start of non-capturing bracket */
2089      OP_BRAPOS,         /* 126 Ditto, with unlimited, possessive repeat */
2090      OP_CBRA,           /* 127 Start of capturing bracket */
2091      OP_CBRAPOS,        /* 128 Ditto, with unlimited, possessive repeat */
2092      OP_COND,           /* 129 Conditional group */
2093    
2094    /* These three must follow the previous three, in the same order. There's a    /* These five must follow the previous five, in the same order. There's a
2095    check for >= SBRA to distinguish the two sets. */    check for >= SBRA to distinguish the two sets. */
2096    
2097    OP_SBRA,           /* 91 Start of non-capturing bracket, check empty  */    OP_SBRA,           /* 130 Start of non-capturing bracket, check empty  */
2098    OP_SCBRA,          /* 92 Start of capturing bracket, check empty */    OP_SBRAPOS,        /* 131 Ditto, with unlimited, possessive repeat */
2099    OP_SCOND,          /* 93 Conditional group, check empty */    OP_SCBRA,          /* 132 Start of capturing bracket, check empty */
2100      OP_SCBRAPOS,       /* 133 Ditto, with unlimited, possessive repeat */
2101    OP_CREF,           /* 94 Used to hold a capture number as condition */    OP_SCOND,          /* 134 Conditional group, check empty */
2102    OP_RREF,           /* 95 Used to hold a recursion number as condition */  
2103    OP_DEF,            /* 96 The DEFINE condition */    /* The next two pairs must (respectively) be kept together. */
2104    
2105      OP_CREF,           /* 135 Used to hold a capture number as condition */
2106      OP_NCREF,          /* 136 Same, but generated by a name reference*/
2107      OP_RREF,           /* 137 Used to hold a recursion number as condition */
2108      OP_NRREF,          /* 138 Same, but generated by a name reference*/
2109      OP_DEF,            /* 139 The DEFINE condition */
2110    
2111      OP_BRAZERO,        /* 140 These two must remain together and in this */
2112      OP_BRAMINZERO,     /* 141 order. */
2113      OP_BRAPOSZERO,     /* 142 */
2114    
2115      /* These are backtracking control verbs */
2116    
2117      OP_MARK,           /* 143 always has an argument */
2118      OP_PRUNE,          /* 144 */
2119      OP_PRUNE_ARG,      /* 145 same, but with argument */
2120      OP_SKIP,           /* 146 */
2121      OP_SKIP_ARG,       /* 147 same, but with argument */
2122      OP_THEN,           /* 148 */
2123      OP_THEN_ARG,       /* 149 same, but with argument */
2124      OP_COMMIT,         /* 150 */
2125    
2126      /* These are forced failure and success verbs */
2127    
2128      OP_FAIL,           /* 151 */
2129      OP_ACCEPT,         /* 152 */
2130      OP_ASSERT_ACCEPT,  /* 153 Used inside assertions */
2131      OP_CLOSE,          /* 154 Used before OP_ACCEPT to close open captures */
2132    
2133      /* This is used to skip a subpattern with a {0} quantifier */
2134    
2135      OP_SKIPZERO,       /* 155 */
2136    
2137      /* This is not an opcode, but is used to check that tables indexed by opcode
2138      are the correct length, in order to catch updating errors - there have been
2139      some in the past. */
2140    
2141    OP_BRAZERO,        /* 97 These two must remain together and in this */    OP_TABLE_LENGTH
   OP_BRAMINZERO      /* 98 order. */  
2142  };  };
2143    
2144    /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
2145    definitions that follow must also be updated to match. There are also tables
2146    called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
2147    
2148    
2149  /* 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
2150  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
2151    only in pcre_printint.c, which fills out the full names in many cases (and in
2152    some cases doesn't actually use these names at all). */
2153    
2154  #define OP_NAME_LIST \  #define OP_NAME_LIST \
2155    "End", "\\A", "\\G", "\\B", "\\b", "\\D", "\\d",                \    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \
2156    "\\S", "\\s", "\\W", "\\w", "Any", "Anybyte",                   \    "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte",         \
2157    "notprop", "prop", "anynl", "extuni",                           \    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \
2158    "\\Z", "\\z",                                                   \    "extuni",  "\\Z", "\\z",                                        \
2159    "Opt", "^", "$", "char", "charnc", "not",                       \    "^", "^", "$", "$", "char", "chari", "not", "noti",             \
2160    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2161      "{", "{", "{",                                                  \
2162    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2163    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2164      "{", "{", "{",                                                  \
2165      "*+","++", "?+", "{",                                           \
2166      "*", "*?", "+", "+?", "?", "??",                                \
2167      "{", "{", "{",                                                  \
2168      "*+","++", "?+", "{",                                           \
2169      "*", "*?", "+", "+?", "?", "??",                                \
2170      "{", "{", "{",                                                  \
2171    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2172    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \
2173    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2174    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \
2175    "class", "nclass", "xclass", "Ref", "Recurse", "Callout",       \    "class", "nclass", "xclass", "Ref", "Refi",                     \
2176    "Alt", "Ket", "KetRmax", "KetRmin", "Assert", "Assert not",     \    "Recurse", "Callout",                                           \
2177    "AssertB", "AssertB not", "Reverse",                            \    "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos",                  \
2178    "Once", "Bra 0", "Bra", "Cond", "SBra 0", "SBra", "SCond",      \    "Reverse", "Assert", "Assert not", "AssertB", "AssertB not",    \
2179    "Cond ref", "Cond rec", "Cond def", "Brazero", "Braminzero"    "Once", "Once_NC",                                              \
2180      "Bra", "BraPos", "CBra", "CBraPos",                             \
2181      "Cond",                                                         \
2182      "SBra", "SBraPos", "SCBra", "SCBraPos",                         \
2183      "SCond",                                                        \
2184      "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def",   \
2185      "Brazero", "Braminzero", "Braposzero",                          \
2186      "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP",                  \
2187      "*THEN", "*THEN", "*COMMIT", "*FAIL",                           \
2188      "*ACCEPT", "*ASSERT_ACCEPT",                                    \
2189      "Close", "Skip zero"
2190    
2191    
2192  /* 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 739  in UTF-8 mode. The code that uses this t Line 2200  in UTF-8 mode. The code that uses this t
2200    
2201  #define OP_LENGTHS \  #define OP_LENGTHS \
2202    1,                             /* End                                    */ \    1,                             /* End                                    */ \
2203    1, 1, 1, 1, 1, 1, 1, 1, 1, 1,  /* \A, \G, \B, \B, \D, \d, \S, \s, \W, \w */ \    1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \
2204    1, 1,                          /* Any, Anybyte                           */ \    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \
2205    3, 3, 1, 1,                    /* NOTPROP, PROP, EXTUNI, ANYNL           */ \    1, 1, 1,                       /* Any, AllAny, Anybyte                   */ \
2206    1, 1, 2, 1, 1,                 /* \Z, \z, Opt, ^, $                      */ \    3, 3,                          /* \P, \p                                 */ \
2207      1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \
2208      1,                             /* \X                                     */ \
2209      1, 1, 1, 1, 1, 1,              /* \Z, \z, ^, ^M, $, $M                   */ \
2210    2,                             /* Char  - the minimum length             */ \    2,                             /* Char  - the minimum length             */ \
2211    2,                             /* Charnc  - the minimum length           */ \    2,                             /* Chari  - the minimum length            */ \
2212    2,                             /* not                                    */ \    2,                             /* not                                    */ \
2213    /* Positive single-char repeats                            ** These are  */ \    2,                             /* noti                                   */ \
2214    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??      ** minima in  */ \    /* Positive single-char repeats                             ** These are */ \
2215    4, 4, 4,                       /* upto, minupto, exact     ** UTF-8 mode */ \    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??       ** minima in */ \
2216    2, 2, 2, 4,                    /* *+, ++, ?+, upto+                      */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto, minupto             ** mode      */ \
2217      2+IMM2_SIZE,                   /* exact                                  */ \
2218      2, 2, 2, 2+IMM2_SIZE,          /* *+, ++, ?+, upto+                      */ \
2219      2, 2, 2, 2, 2, 2,              /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8     */ \
2220      2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto I, minupto I                      */ \
2221      2+IMM2_SIZE,                   /* exact I                                */ \
2222      2, 2, 2, 2+IMM2_SIZE,          /* *+I, ++I, ?+I, upto+I                  */ \
2223    /* Negative single-char repeats - only for chars < 256                   */ \    /* Negative single-char repeats - only for chars < 256                   */ \
2224    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \
2225    4, 4, 4,                       /* NOT upto, minupto, exact               */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto, minupto                      */ \
2226    2, 2, 2, 4,                    /* Possessive *, +, ?, upto               */ \    2+IMM2_SIZE,                   /* NOT exact                              */ \
2227      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *, +, ?, upto           */ \
2228      2, 2, 2, 2, 2, 2,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */ \
2229      2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto I, minupto I                  */ \
2230      2+IMM2_SIZE,                   /* NOT exact I                            */ \
2231      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *I, +I, ?I, upto I      */ \
2232    /* Positive type repeats                                                 */ \    /* Positive type repeats                                                 */ \
2233    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \
2234    4, 4, 4,                       /* Type upto, minupto, exact              */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* Type upto, minupto                     */ \
2235    2, 2, 2, 4,                    /* Possessive *+, ++, ?+, upto+           */ \    2+IMM2_SIZE,                   /* Type exact                             */ \
2236      2, 2, 2, 2+IMM2_SIZE,          /* Possessive *+, ++, ?+, upto+           */ \
2237    /* Character class & ref repeats                                         */ \    /* Character class & ref repeats                                         */ \
2238    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \
2239    5, 5,                          /* CRRANGE, CRMINRANGE                    */ \    1+2*IMM2_SIZE, 1+2*IMM2_SIZE,  /* CRRANGE, CRMINRANGE                    */ \
2240   33,                             /* CLASS                                  */ \    1+(32/sizeof(pcre_uchar)),     /* CLASS                                  */ \
2241   33,                             /* NCLASS                                 */ \    1+(32/sizeof(pcre_uchar)),     /* NCLASS                                 */ \
2242    0,                             /* XCLASS - variable length               */ \    0,                             /* XCLASS - variable length               */ \
2243    3,                             /* REF                                    */ \    1+IMM2_SIZE,                   /* REF                                    */ \
2244      1+IMM2_SIZE,                   /* REFI                                   */ \
2245    1+LINK_SIZE,                   /* RECURSE                                */ \    1+LINK_SIZE,                   /* RECURSE                                */ \
2246    2+2*LINK_SIZE,                 /* CALLOUT                                */ \    2+2*LINK_SIZE,                 /* CALLOUT                                */ \
2247    1+LINK_SIZE,                   /* Alt                                    */ \    1+LINK_SIZE,                   /* Alt                                    */ \
2248    1+LINK_SIZE,                   /* Ket                                    */ \    1+LINK_SIZE,                   /* Ket                                    */ \
2249    1+LINK_SIZE,                   /* KetRmax                                */ \    1+LINK_SIZE,                   /* KetRmax                                */ \
2250    1+LINK_SIZE,                   /* KetRmin                                */ \    1+LINK_SIZE,                   /* KetRmin                                */ \
2251      1+LINK_SIZE,                   /* KetRpos                                */ \
2252      1+LINK_SIZE,                   /* Reverse                                */ \
2253    1+LINK_SIZE,                   /* Assert                                 */ \    1+LINK_SIZE,                   /* Assert                                 */ \
2254    1+LINK_SIZE,                   /* Assert not                             */ \    1+LINK_SIZE,                   /* Assert not                             */ \
2255    1+LINK_SIZE,                   /* Assert behind                          */ \    1+LINK_SIZE,                   /* Assert behind                          */ \
2256    1+LINK_SIZE,                   /* Assert behind not                      */ \    1+LINK_SIZE,                   /* Assert behind not                      */ \
   1+LINK_SIZE,                   /* Reverse                                */ \  
2257    1+LINK_SIZE,                   /* ONCE                                   */ \    1+LINK_SIZE,                   /* ONCE                                   */ \
2258      1+LINK_SIZE,                   /* ONCE_NC                                */ \
2259    1+LINK_SIZE,                   /* BRA                                    */ \    1+LINK_SIZE,                   /* BRA                                    */ \
2260    3+LINK_SIZE,                   /* CBRA                                   */ \    1+LINK_SIZE,                   /* BRAPOS                                 */ \
2261      1+LINK_SIZE+IMM2_SIZE,         /* CBRA                                   */ \
2262      1+LINK_SIZE+IMM2_SIZE,         /* CBRAPOS                                */ \
2263    1+LINK_SIZE,                   /* COND                                   */ \    1+LINK_SIZE,                   /* COND                                   */ \
2264    1+LINK_SIZE,                   /* SBRA                                   */ \    1+LINK_SIZE,                   /* SBRA                                   */ \
2265    3+LINK_SIZE,                   /* SCBRA                                  */ \    1+LINK_SIZE,                   /* SBRAPOS                                */ \
2266      1+LINK_SIZE+IMM2_SIZE,         /* SCBRA                                  */ \
2267      1+LINK_SIZE+IMM2_SIZE,         /* SCBRAPOS                               */ \
2268    1+LINK_SIZE,                   /* SCOND                                  */ \    1+LINK_SIZE,                   /* SCOND                                  */ \
2269    3,                             /* CREF                                   */ \    1+IMM2_SIZE, 1+IMM2_SIZE,      /* CREF, NCREF                            */ \
2270    3,                             /* RREF                                   */ \    1+IMM2_SIZE, 1+IMM2_SIZE,      /* RREF, NRREF                            */ \
2271    1,                             /* DEF                                    */ \    1,                             /* DEF                                    */ \
2272    1, 1,                          /* BRAZERO, BRAMINZERO                    */ \    1, 1, 1,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */ \
2273      3, 1, 3,                       /* MARK, PRUNE, PRUNE_ARG                 */ \
2274      1, 3,                          /* SKIP, SKIP_ARG                         */ \
2275      1, 3,                          /* THEN, THEN_ARG                         */ \
2276      1, 1, 1, 1,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */ \
2277      1+IMM2_SIZE, 1                 /* CLOSE, SKIPZERO                        */
2278    
2279    /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
2280  /* A magic value for OP_RREF to indicate the "any recursion" condition. */  condition. */
2281    
2282  #define RREF_ANY  0xffff  #define RREF_ANY  0xffff
2283    
2284  /* 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
2285  tracked. */  easily be tracked. When a new number is added, the table called eint in
2286    pcreposix.c must be updated. */
2287    
2288  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
2289         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
2290         ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,         ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
2291         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
2292         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
2293         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57 };         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
2294           ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
2295           ERR70, ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERR77, ERRCOUNT };
2296    
2297    /* JIT compiling modes. The function list is indexed by them. */
2298    enum { JIT_COMPILE, JIT_PARTIAL_SOFT_COMPILE, JIT_PARTIAL_HARD_COMPILE,
2299           JIT_NUMBER_OF_COMPILE_MODES };
2300    
2301  /* 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
2302  code vector run on as long as necessary after the end. We store an explicit  code vector run on as long as necessary after the end. We store an explicit
# Line 817  Because people can now save and re-use c Line 2312  Because people can now save and re-use c
2312  structure should be made at the end, and something earlier (e.g. a new  structure should be made at the end, and something earlier (e.g. a new
2313  flag in the options or one of the dummy fields) should indicate that the new  flag in the options or one of the dummy fields) should indicate that the new
2314  fields are present. Currently PCRE always sets the dummy fields to zero.  fields are present. Currently PCRE always sets the dummy fields to zero.
2315  NOTE NOTE NOTE:  NOTE NOTE NOTE
2316  */  */
2317    
2318  typedef struct real_pcre {  #if defined COMPILE_PCRE8
2319    #define REAL_PCRE real_pcre
2320    #elif defined COMPILE_PCRE16
2321    #define REAL_PCRE real_pcre16
2322    #elif defined COMPILE_PCRE32
2323    #define REAL_PCRE real_pcre32
2324    #endif
2325    
2326    /* It is necessary to fork the struct for 32 bit, since it needs to use
2327     * pcre_uchar for first_char and req_char. Can't put an ifdef inside the
2328     * typedef since pcretest needs access to  the struct of the 8-, 16-
2329     * and 32-bit variants. */
2330    
2331    typedef struct real_pcre8_or_16 {
2332    pcre_uint32 magic_number;    pcre_uint32 magic_number;
2333    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2334    pcre_uint32 options;    pcre_uint32 options;            /* Public options */
2335    pcre_uint32 dummy1;             /* For future use, maybe */    pcre_uint16 flags;              /* Private flags */
2336      pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2337    pcre_uint16 top_bracket;    pcre_uint16 top_bracket;        /* Highest numbered group */
2338    pcre_uint16 top_backref;    pcre_uint16 top_backref;        /* Highest numbered back reference */
2339    pcre_uint16 first_byte;    pcre_uint16 first_char;         /* Starting character */
2340    pcre_uint16 req_byte;    pcre_uint16 req_char;           /* This character must be seen */
2341    pcre_uint16 name_table_offset;  /* Offset to name table that follows */    pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2342    pcre_uint16 name_entry_size;    /* Size of any name items */    pcre_uint16 name_entry_size;    /* Size of any name items */
2343    pcre_uint16 name_count;         /* Number of name items */    pcre_uint16 name_count;         /* Number of name items */
2344    pcre_uint16 ref_count;          /* Reference count */    pcre_uint16 ref_count;          /* Reference count */
2345      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2346      const pcre_uint8 *nullpad;      /* NULL padding */
2347    } real_pcre8_or_16;
2348    
2349    typedef struct real_pcre8_or_16 real_pcre;
2350    typedef struct real_pcre8_or_16 real_pcre16;
2351    
2352    const unsigned char *tables;    /* Pointer to tables or NULL for std */  typedef struct real_pcre32 {
2353    const unsigned char *nullpad;   /* NULL padding */    pcre_uint32 magic_number;
2354  } real_pcre;    pcre_uint32 size;               /* Total that was malloced */
2355      pcre_uint32 options;            /* Public options */
2356      pcre_uint16 flags;              /* Private flags */
2357      pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2358      pcre_uint16 top_bracket;        /* Highest numbered group */
2359      pcre_uint16 top_backref;        /* Highest numbered back reference */
2360      pcre_uint32 first_char;         /* Starting character */
2361      pcre_uint32 req_char;           /* This character must be seen */
2362      pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2363      pcre_uint16 name_entry_size;    /* Size of any name items */
2364      pcre_uint16 name_count;         /* Number of name items */
2365      pcre_uint16 ref_count;          /* Reference count */
2366      pcre_uint16 dummy1;             /* for later expansion */
2367      pcre_uint16 dummy2;             /* for later expansion */
2368      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2369      void *nullpad;                  /* for later expansion */
2370    } real_pcre32;
2371    
2372    /* Assert that the size of REAL_PCRE is divisible by 8 */
2373    typedef int __assert_real_pcre_size_divisible_8[(sizeof(REAL_PCRE) % 8) == 0 ? 1 : -1];
2374    
2375    /* Needed in pcretest to access some fields in the real_pcre* structures
2376     * directly. They're unified for 8/16/32 bits since the structs only differ
2377     * after these fields; if that ever changes, need to fork those defines into
2378     * 8/16 and 32 bit versions. */
2379    #define REAL_PCRE_MAGIC(re)     (((REAL_PCRE*)re)->magic_number)
2380    #define REAL_PCRE_SIZE(re)      (((REAL_PCRE*)re)->size)
2381    #define REAL_PCRE_OPTIONS(re)   (((REAL_PCRE*)re)->options)
2382    #define REAL_PCRE_FLAGS(re)     (((REAL_PCRE*)re)->flags)
2383    
2384  /* 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
2385  remark (see NOTE above) about extending this structure applies. */  remark (see NOTE above) about extending this structure applies. */
2386    
2387  typedef struct pcre_study_data {  typedef struct pcre_study_data {
2388    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2389    pcre_uint32 options;    pcre_uint32 flags;              /* Private flags */
2390    uschar start_bits[32];    pcre_uint8 start_bits[32];      /* Starting char bits */
2391      pcre_uint32 minlength;          /* Minimum subject length */
2392  } pcre_study_data;  } pcre_study_data;
2393    
2394    /* Structure for building a chain of open capturing subpatterns during
2395    compiling, so that instructions to close them can be compiled when (*ACCEPT) is
2396    encountered. This is also used to identify subpatterns that contain recursive
2397    back references to themselves, so that they can be made atomic. */
2398    
2399    typedef struct open_capitem {
2400      struct open_capitem *next;    /* Chain link */
2401      pcre_uint16 number;           /* Capture number */
2402      pcre_uint16 flag;             /* Set TRUE if recursive back ref */
2403    } open_capitem;
2404    
2405  /* Structure for passing "static" information around between the functions  /* Structure for passing "static" information around between the functions
2406  doing the compiling, so that they are thread-safe. */  doing the compiling, so that they are thread-safe. */
2407    
2408  typedef struct compile_data {  typedef struct compile_data {
2409    const uschar *lcc;            /* Points to lower casing table */    const pcre_uint8 *lcc;            /* Points to lower casing table */
2410    const uschar *fcc;            /* Points to case-flipping table */    const pcre_uint8 *fcc;            /* Points to case-flipping table */
2411    const uschar *cbits;          /* Points to character type table */    const pcre_uint8 *cbits;          /* Points to character type table */
2412    const uschar *ctypes;         /* Points to table of type maps */    const pcre_uint8 *ctypes;         /* Points to table of type maps */
2413    const uschar *start_workspace;/* The start of working space */    const pcre_uchar *start_workspace;/* The start of working space */
2414    const uschar *start_code;     /* The start of the compiled code */    const pcre_uchar *start_code;     /* The start of the compiled code */
2415    const uschar *start_pattern;  /* The start of the pattern */    const pcre_uchar *start_pattern;  /* The start of the pattern */
2416    const uschar *end_pattern;    /* The end of the pattern */    const pcre_uchar *end_pattern;    /* The end of the pattern */
2417    uschar *hwm;                  /* High watermark of workspace */    open_capitem *open_caps;          /* Chain of open capture items */
2418    uschar *name_table;           /* The name/number table */    pcre_uchar *hwm;                  /* High watermark of workspace */
2419    int  names_found;             /* Number of entries so far */    pcre_uchar *name_table;           /* The name/number table */
2420    int  name_entry_size;         /* Size of each entry */    int  names_found;                 /* Number of entries so far */
2421    int  bracount;                /* Count of capturing parens */    int  name_entry_size;             /* Size of each entry */
2422    int  top_backref;             /* Maximum back reference */    int  workspace_size;              /* Size of workspace */
2423    unsigned int backref_map;     /* Bitmap of low back refs */    unsigned int  bracount;           /* Count of capturing parens as we compile */
2424    int  external_options;        /* External (initial) options */    int  final_bracount;              /* Saved value after first pass */
2425    int  req_varyopt;             /* "After variable item" flag for reqbyte */    int  max_lookbehind;              /* Maximum lookbehind (characters) */
2426    BOOL nopartial;               /* Set TRUE if partial won't work */    int  top_backref;                 /* Maximum back reference */
2427    int  nltype;                  /* Newline type */    unsigned int backref_map;         /* Bitmap of low back refs */
2428    int  nllen;                   /* Newline string length */    int  assert_depth;                /* Depth of nested assertions */
2429    uschar nl[4];                 /* Newline string when fixed length */    int  external_options;            /* External (initial) options */
2430      int  external_flags;              /* External flag bits to be set */
2431      int  req_varyopt;                 /* "After variable item" flag for reqbyte */
2432      BOOL had_accept;                  /* (*ACCEPT) encountered */
2433      BOOL had_pruneorskip;             /* (*PRUNE) or (*SKIP) encountered */
2434      BOOL check_lookbehind;            /* Lookbehinds need later checking */
2435      int  nltype;                      /* Newline type */
2436      int  nllen;                       /* Newline string length */
2437      pcre_uchar nl[4];                 /* Newline string when fixed length */
2438  } compile_data;  } compile_data;
2439    
2440  /* Structure for maintaining a chain of pointers to the currently incomplete  /* Structure for maintaining a chain of pointers to the currently incomplete
2441  branches, for testing for left recursion. */  branches, for testing for left recursion while compiling. */
2442    
2443  typedef struct branch_chain {  typedef struct branch_chain {
2444    struct branch_chain *outer;    struct branch_chain *outer;
2445    uschar *current;    pcre_uchar *current_branch;
2446  } branch_chain;  } branch_chain;
2447    
2448  /* Structure for items in a linked list that represents an explicit recursive  /* Structure for items in a linked list that represents an explicit recursive
2449  call within the pattern. */  call within the pattern; used by pcre_exec(). */
2450    
2451  typedef struct recursion_info {  typedef struct recursion_info {
2452    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
2453    int group_num;                /* Number of group that was called */    unsigned int group_num;         /* Number of group that was called */
2454    const uschar *after_call;     /* "Return value": points after the call in the expr */    int *offset_save;               /* Pointer to start of saved offsets */
2455    USPTR save_start;             /* Old value of md->start_match */    int saved_max;                  /* Number of saved offsets */
2456    int *offset_save;             /* Pointer to start of saved offsets */    PCRE_PUCHAR subject_position;   /* Position at start of recursion */
   int saved_max;                /* Number of saved offsets */  
2457  } recursion_info;  } recursion_info;
2458    
2459  /* When compiling in a mode that doesn't use recursive calls to match(),  /* A similar structure for pcre_dfa_exec(). */
 a structure is used to remember local variables on the heap. It is defined in  
 pcre_exec.c, close to the match() function, so that it is easy to keep it in  
 step with any changes of local variable. However, the pointer to the current  
 frame must be saved in some "static" place over a longjmp(). We declare the  
 structure here so that we can put a pointer in the match_data structure. NOTE:  
 This isn't used for a "normal" compilation of pcre. */  
2460    
2461  struct heapframe;  typedef struct dfa_recursion_info {
2462      struct dfa_recursion_info *prevrec;
2463      int group_num;
2464      PCRE_PUCHAR subject_position;
2465    } dfa_recursion_info;
2466    
2467  /* 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
2468  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
2469  has been matched by a subpattern - to break infinite loops. */  has been matched by a subpattern - to break infinite loops; used by
2470    pcre_exec(). */
2471    
2472  typedef struct eptrblock {  typedef struct eptrblock {
2473    struct eptrblock *epb_prev;    struct eptrblock *epb_prev;
2474    USPTR epb_saved_eptr;    PCRE_PUCHAR epb_saved_eptr;
2475  } eptrblock;  } eptrblock;
2476    
2477    
# Line 922  typedef struct match_data { Line 2482  typedef struct match_data {
2482    unsigned long int match_call_count;      /* As it says */    unsigned long int match_call_count;      /* As it says */
2483    unsigned long int match_limit;           /* As it says */    unsigned long int match_limit;           /* As it says */
2484    unsigned long int match_limit_recursion; /* As it says */    unsigned long int match_limit_recursion; /* As it says */
2485    int   *offset_vector;         /* Offset vector */    int   *offset_vector;           /* Offset vector */
2486    int    offset_end;            /* One past the end */    int    offset_end;              /* One past the end */
2487    int    offset_max;            /* The maximum usable for return data */    int    offset_max;              /* The maximum usable for return data */
2488    int    nltype;                /* Newline type */    int    nltype;                  /* Newline type */
2489    int    nllen;                 /* Newline string length */    int    nllen;                   /* Newline string length */
2490    uschar nl[4];                 /* Newline string when fixed */    int    name_count;              /* Number of names in name table */
2491    const uschar *lcc;            /* Points to lower casing table */    int    name_entry_size;         /* Size of entry in names table */
2492    const uschar *ctypes;         /* Points to table of type maps */    pcre_uchar *name_table;         /* Table of names */
2493    BOOL   offset_overflow;       /* Set if too many extractions */    pcre_uchar nl[4];               /* Newline string when fixed */
2494    BOOL   notbol;                /* NOTBOL flag */    const  pcre_uint8 *lcc;         /* Points to lower casing table */
2495    BOOL   noteol;                /* NOTEOL flag */    const  pcre_uint8 *fcc;         /* Points to case-flipping table */
2496    BOOL   utf8;                  /* UTF8 flag */    const  pcre_uint8 *ctypes;      /* Points to table of type maps */
2497    BOOL   endonly;               /* Dollar not before final \n */    BOOL   offset_overflow;         /* Set if too many extractions */
2498    BOOL   notempty;              /* Empty string match not wanted */    BOOL   notbol;                  /* NOTBOL flag */
2499    BOOL   partial;               /* PARTIAL flag */    BOOL   noteol;                  /* NOTEOL flag */
2500    BOOL   hitend;                /* Hit the end of the subject at some point */    BOOL   utf;                     /* UTF-8 / UTF-16 flag */
2501    const uschar *start_code;     /* For use when recursing */    BOOL   jscript_compat;          /* JAVASCRIPT_COMPAT flag */
2502    USPTR  start_subject;         /* Start of the subject string */    BOOL   use_ucp;                 /* PCRE_UCP flag */
2503    USPTR  end_subject;           /* End of the subject string */    BOOL   endonly;                 /* Dollar not before final \n */
2504    USPTR  start_match;           /* Start of this match attempt */    BOOL   notempty;                /* Empty string match not wanted */
2505    USPTR  end_match_ptr;         /* Subject position at end match */    BOOL   notempty_atstart;        /* Empty string match at start not wanted */
2506    int    end_offset_top;        /* Highwater mark at end of match */    BOOL   hitend;                  /* Hit the end of the subject at some point */
2507    int    capture_last;          /* Most recent capture number */    BOOL   bsr_anycrlf;             /* \R is just any CRLF, not full Unicode */
2508    int    start_offset;          /* The start offset value */    BOOL   hasthen;                 /* Pattern contains (*THEN) */
2509    eptrblock *eptrchain;         /* Chain of eptrblocks for tail recursions */    BOOL   ignore_skip_arg;         /* For re-run when SKIP name not found */
2510    int    eptrn;                 /* Next free eptrblock */    const  pcre_uchar *start_code;  /* For use when recursing */
2511    recursion_info *recursive;    /* Linked list of recursion data */    PCRE_PUCHAR start_subject;      /* Start of the subject string */
2512    void  *callout_data;          /* To pass back to callouts */    PCRE_PUCHAR end_subject;        /* End of the subject string */
2513    struct heapframe *thisframe;  /* Used only when compiling for no recursion */    PCRE_PUCHAR start_match_ptr;    /* Start of matched string */
2514      PCRE_PUCHAR end_match_ptr;      /* Subject position at end match */
2515      PCRE_PUCHAR start_used_ptr;     /* Earliest consulted character */
2516      int    partial;                 /* PARTIAL options */
2517      int    end_offset_top;          /* Highwater mark at end of match */
2518      int    capture_last;            /* Most recent capture number */
2519      int    start_offset;            /* The start offset value */
2520      int    match_function_type;     /* Set for certain special calls of MATCH() */
2521      eptrblock *eptrchain;           /* Chain of eptrblocks for tail recursions */
2522      int    eptrn;                   /* Next free eptrblock */
2523      recursion_info *recursive;      /* Linked list of recursion data */
2524      void  *callout_data;            /* To pass back to callouts */
2525      const  pcre_uchar *mark;        /* Mark pointer to pass back on success */
2526      const  pcre_uchar *nomatch_mark;/* Mark pointer to pass back on failure */
2527      const  pcre_uchar *once_target; /* Where to back up to for atomic groups */
2528    #ifdef NO_RECURSE
2529      void  *match_frames_base;       /* For remembering malloc'd frames */
2530    #endif
2531  } match_data;  } match_data;
2532    
2533  /* 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
2534  functions. */  functions. */
2535    
2536  typedef struct dfa_match_data {  typedef struct dfa_match_data {
2537    const uschar *start_code;     /* Start of the compiled pattern */    const pcre_uchar *start_code;     /* Start of the compiled pattern */
2538    const uschar *start_subject;  /* Start of the subject string */    const pcre_uchar *start_subject ; /* Start of the subject string */
2539    const uschar *end_subject;    /* End of subject string */    const pcre_uchar *end_subject;    /* End of subject string */
2540    const uschar *tables;         /* Character tables */    const pcre_uchar *start_used_ptr; /* Earliest consulted character */
2541    int   moptions;               /* Match options */    const pcre_uint8 *tables;         /* Character tables */
2542    int   poptions;               /* Pattern options */    int   start_offset;               /* The start offset value */
2543    int    nltype;                /* Newline type */    int   moptions;                   /* Match options */
2544    int    nllen;                 /* Newline string length */    int   poptions;                   /* Pattern options */
2545    uschar nl[4];                 /* Newline string when fixed */    int   nltype;                     /* Newline type */
2546    void  *callout_data;          /* To pass back to callouts */    int   nllen;                      /* Newline string length */
2547      pcre_uchar nl[4];                 /* Newline string when fixed */
2548      void *callout_data;               /* To pass back to callouts */
2549      dfa_recursion_info *recursive;    /* Linked list of recursion data */
2550  } dfa_match_data;  } dfa_match_data;
2551    
2552  /* Bit definitions for entries in the pcre_ctypes table. */  /* Bit definitions for entries in the pcre_ctypes table. */
# Line 975  typedef struct dfa_match_data { Line 2555  typedef struct dfa_match_data {
2555  #define ctype_letter  0x02  #define ctype_letter  0x02
2556  #define ctype_digit   0x04  #define ctype_digit   0x04
2557  #define ctype_xdigit  0x08  #define ctype_xdigit  0x08
2558  #define ctype_word    0x10   /* alphameric or '_' */  #define ctype_word    0x10   /* alphanumeric or '_' */
2559  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */
2560    
2561  /* 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 1002  total length. */ Line 2582  total length. */
2582  #define ctypes_offset (cbits_offset + cbit_length)  #define ctypes_offset (cbits_offset + cbit_length)
2583  #define tables_length (ctypes_offset + 256)  #define tables_length (ctypes_offset + 256)
2584    
2585    /* Internal function and data prefixes. */
2586    
2587    #if defined COMPILE_PCRE8
2588    #ifndef PUBL
2589    #define PUBL(name) pcre_##name
2590    #endif
2591    #ifndef PRIV
2592    #define PRIV(name) _pcre_##name
2593    #endif
2594    #elif defined COMPILE_PCRE16
2595    #ifndef PUBL
2596    #define PUBL(name) pcre16_##name
2597    #endif
2598    #ifndef PRIV
2599    #define PRIV(name) _pcre16_##name
2600    #endif
2601    #elif defined COMPILE_PCRE32
2602    #ifndef PUBL
2603    #define PUBL(name) pcre32_##name
2604    #endif
2605    #ifndef PRIV
2606    #define PRIV(name) _pcre32_##name
2607    #endif
2608    #else
2609    #error Unsupported compiling mode
2610    #endif /* COMPILE_PCRE[8|16|32] */
2611    
2612  /* 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
2613  codes. */  codes. Each entry used to point directly to a name, but to reduce the number of
2614    relocations in shared libraries, it now has an offset into a single string
2615    instead. */
2616    
2617  typedef struct {  typedef struct {
2618    const char *name;    pcre_uint16 name_offset;
2619    pcre_uint16 type;    pcre_uint16 type;
2620    pcre_uint16 value;    pcre_uint16 value;
2621  } ucp_type_table;  } ucp_type_table;
# Line 1017  of the exported public functions. They h Line 2626  of the exported public functions. They h
2626  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
2627  pcre_tables.c module. */  pcre_tables.c module. */
2628    
2629  extern const int    _pcre_utf8_table1[];  #ifdef COMPILE_PCRE8
2630  extern const int    _pcre_utf8_table2[];  extern const int            PRIV(utf8_table1)[];
2631  extern const int    _pcre_utf8_table3[];  extern const int            PRIV(utf8_table1_size);
2632  extern const uschar _pcre_utf8_table4[];  extern const int            PRIV(utf8_table2)[];
2633    extern const int            PRIV(utf8_table3)[];
2634  extern const int    _pcre_utf8_table1_size;  extern const pcre_uint8     PRIV(utf8_table4)[];
2635    #endif /* COMPILE_PCRE8 */
2636    
2637    extern const char           PRIV(utt_names)[];
2638    extern const ucp_type_table PRIV(utt)[];
2639    extern const int            PRIV(utt_size);
2640    
2641  extern const ucp_type_table _pcre_utt[];  extern const pcre_uint8     PRIV(OP_lengths)[];
2642  extern const int _pcre_utt_size;  extern const pcre_uint8     PRIV(default_tables)[];
2643    
2644  extern const uschar _pcre_default_tables[];  extern const pcre_uint32    PRIV(hspace_list)[];
2645    extern const pcre_uint32    PRIV(vspace_list)[];
 extern const uschar _pcre_OP_lengths[];  
2646    
2647    
2648  /* Internal shared functions. These are functions that are used by more than  /* Internal shared functions. These are functions that are used by more than
2649  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
2650  sense, but are not part of the PCRE public API. */  sense, but are not part of the PCRE public API. */
2651    
2652  extern BOOL         _pcre_is_newline(const uschar *, const uschar *, int *,  /* String comparison functions. */
2653                        BOOL);  #if defined COMPILE_PCRE8
2654  extern int          _pcre_ord2utf8(int, uschar *);  
2655  extern real_pcre   *_pcre_try_flipped(const real_pcre *, real_pcre *,  #define STRCMP_UC_UC(str1, str2) \
2656                        const pcre_study_data *, pcre_study_data *);    strcmp((char *)(str1), (char *)(str2))
2657  extern int          _pcre_ucp_findprop(const unsigned int, int *, int *);  #define STRCMP_UC_C8(str1, str2) \
2658  extern unsigned int _pcre_ucp_othercase(const unsigned int);    strcmp((char *)(str1), (str2))
2659  extern int          _pcre_valid_utf8(const uschar *, int);  #define STRNCMP_UC_UC(str1, str2, num) \
2660  extern BOOL         _pcre_was_newline(const uschar *, const uschar *, int *,    strncmp((char *)(str1), (char *)(str2), (num))
2661                        BOOL);  #define STRNCMP_UC_C8(str1, str2, num) \
2662  extern BOOL         _pcre_xclass(int, const uschar *);    strncmp((char *)(str1), (str2), (num))
2663    #define STRLEN_UC(str) strlen((const char *)str)
2664    
2665    #elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
2666    
2667    extern int               PRIV(strcmp_uc_uc)(const pcre_uchar *,
2668                               const pcre_uchar *);
2669    extern int               PRIV(strcmp_uc_c8)(const pcre_uchar *,
2670                               const char *);
2671    extern int               PRIV(strncmp_uc_uc)(const pcre_uchar *,
2672                               const pcre_uchar *, unsigned int num);
2673    extern int               PRIV(strncmp_uc_c8)(const pcre_uchar *,
2674                               const char *, unsigned int num);
2675    extern unsigned int      PRIV(strlen_uc)(const pcre_uchar *str);
2676    
2677    #define STRCMP_UC_UC(str1, str2) \
2678      PRIV(strcmp_uc_uc)((str1), (str2))
2679    #define STRCMP_UC_C8(str1, str2) \
2680      PRIV(strcmp_uc_c8)((str1), (str2))
2681    #define STRNCMP_UC_UC(str1, str2, num) \
2682      PRIV(strncmp_uc_uc)((str1), (str2), (num))
2683    #define STRNCMP_UC_C8(str1, str2, num) \
2684      PRIV(strncmp_uc_c8)((str1), (str2), (num))
2685    #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2686    
2687    #endif /* COMPILE_PCRE[8|16|32] */
2688    
2689    #if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
2690    
2691    #define STRCMP_UC_UC_TEST(str1, str2) STRCMP_UC_UC(str1, str2)
2692    #define STRCMP_UC_C8_TEST(str1, str2) STRCMP_UC_C8(str1, str2)
2693    
2694    #elif defined COMPILE_PCRE32
2695    
2696    extern int               PRIV(strcmp_uc_uc_utf)(const pcre_uchar *,
2697                               const pcre_uchar *);
2698    extern int               PRIV(strcmp_uc_c8_utf)(const pcre_uchar *,
2699                               const char *);
2700    
2701    #define STRCMP_UC_UC_TEST(str1, str2) \
2702      (utf ? PRIV(strcmp_uc_uc_utf)((str1), (str2)) : PRIV(strcmp_uc_uc)((str1), (str2)))
2703    #define STRCMP_UC_C8_TEST(str1, str2) \
2704      (utf ? PRIV(strcmp_uc_c8_utf)((str1), (str2)) : PRIV(strcmp_uc_c8)((str1), (str2)))
2705    
2706    #endif /* COMPILE_PCRE[8|16|32] */
2707    
2708    extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2709    extern BOOL              PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2710                               int *, BOOL);
2711    extern unsigned int      PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
2712    extern int               PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
2713    extern BOOL              PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2714                               int *, BOOL);
2715    extern BOOL              PRIV(xclass)(pcre_uint32, const pcre_uchar *, BOOL);
2716    
2717    #ifdef SUPPORT_JIT
2718    extern void              PRIV(jit_compile)(const REAL_PCRE *,
2719                               PUBL(extra) *, int);
2720    extern int               PRIV(jit_exec)(const PUBL(extra) *,
2721                               const pcre_uchar *, int, int, int, int *, int);
2722    extern void              PRIV(jit_free)(void *);
2723    extern int               PRIV(jit_get_size)(void *);
2724    extern const char*       PRIV(jit_get_target)(void);
2725    #endif
2726    
2727    /* Unicode character database (UCD) */
2728    
2729    typedef struct {
2730      pcre_uint8 script;     /* ucp_Arabic, etc. */
2731      pcre_uint8 chartype;   /* ucp_Cc, etc. (general categories) */
2732      pcre_uint8 gbprop;     /* ucp_gbControl, etc. (grapheme break property) */
2733      pcre_uint8 caseset;    /* offset to multichar other cases or zero */
2734      pcre_int32 other_case; /* offset to other case, or zero if none */
2735    } ucd_record;
2736    
2737    extern const pcre_uint32 PRIV(ucd_caseless_sets)[];
2738    extern const ucd_record  PRIV(ucd_records)[];
2739    extern const pcre_uint8  PRIV(ucd_stage1)[];
2740    extern const pcre_uint16 PRIV(ucd_stage2)[];
2741    extern const pcre_uint32 PRIV(ucp_gentype)[];
2742    extern const pcre_uint32 PRIV(ucp_gbtable)[];
2743    #ifdef SUPPORT_JIT
2744    extern const int         PRIV(ucp_typerange)[];
2745    #endif
2746    
2747    #ifdef SUPPORT_UCP
2748    /* UCD access macros */
2749    
2750    #define UCD_BLOCK_SIZE 128
2751    #define GET_UCD(ch) (PRIV(ucd_records) + \
2752            PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \
2753            UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE])
2754    
2755    #define UCD_CHARTYPE(ch)    GET_UCD(ch)->chartype
2756    #define UCD_SCRIPT(ch)      GET_UCD(ch)->script
2757    #define UCD_CATEGORY(ch)    PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2758    #define UCD_GRAPHBREAK(ch)  GET_UCD(ch)->gbprop
2759    #define UCD_CASESET(ch)     GET_UCD(ch)->caseset
2760    #define UCD_OTHERCASE(ch)   ((pcre_uint32)((int)ch + (int)(GET_UCD(ch)->other_case)))
2761    
2762    #endif /* SUPPORT_UCP */
2763    
2764  #endif  #endif
2765    

Legend:
Removed from v.137  
changed lines
  Added in v.1198

  ViewVC Help
Powered by ViewVC 1.1.5