/[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 91 by nigel, Sat Feb 24 21:41:34 2007 UTC revision 1045 by ph10, Sun Sep 23 16:50:00 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) 1997-2006 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_" or "_pcre16_" depending on
44    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    #ifndef COMPILE_PCRE16
57    #define COMPILE_PCRE8
58    #endif
59    
60    /* If SUPPORT_UCP is defined, SUPPORT_UTF must also be defined. The
61    "configure" script ensures this, but not everybody uses "configure". */
62    
63    #if defined SUPPORT_UCP && !(defined SUPPORT_UTF)
64    #define SUPPORT_UTF 1
65    #endif
66    
67    /* We define SUPPORT_UTF if SUPPORT_UTF8 is enabled for compatibility
68    reasons with existing code. */
69    
70    #if defined SUPPORT_UTF8 && !(defined SUPPORT_UTF)
71    #define SUPPORT_UTF 1
72    #endif
73    
74    /* Fixme: SUPPORT_UTF8 should be eventually disappear from the code.
75    Until then we define it if SUPPORT_UTF is defined. */
76    
77    #if defined SUPPORT_UTF && !(defined SUPPORT_UTF8)
78    #define SUPPORT_UTF8 1
79    #endif
80    
81    /* We do not support both EBCDIC and UTF-8/16 at the same time. The "configure"
82    script prevents both being selected, but not everybody uses "configure". */
83    
84    #if defined EBCDIC && defined SUPPORT_UTF
85    #error The use of both EBCDIC and SUPPORT_UTF8/16 is not supported.
86  #endif  #endif
87    
88  /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef  /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
89  inline, and there are *still* stupid compilers about that don't like indented  inline, and there are *still* stupid compilers about that don't like indented
90  pre-processor statements, or at least there were when I first wrote this. After  pre-processor statements, or at least there were when I first wrote this. After
91  all, it had only been about 10 years then... */  all, it had only been about 10 years then...
92    
93  #ifdef DEBUG  It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so
94    be absolutely sure we get our version. */
95    
96    #undef DPRINTF
97    #ifdef PCRE_DEBUG
98  #define DPRINTF(p) printf p  #define DPRINTF(p) printf p
99  #else  #else
100  #define DPRINTF(p) /*nothing*/  #define DPRINTF(p) /* Nothing */
101  #endif  #endif
102    
103    
 /* Get the definitions provided by running "configure" */  
   
 #include "config.h"  
   
104  /* Standard C headers plus the external interface definition. The only time  /* Standard C headers plus the external interface definition. The only time
105  setjmp and stdarg are used is when NO_RECURSE is set. */  setjmp and stdarg are used is when NO_RECURSE is set. */
106    
107  #include <ctype.h>  #include <ctype.h>
108  #include <limits.h>  #include <limits.h>
 #include <setjmp.h>  
 #include <stdarg.h>  
109  #include <stddef.h>  #include <stddef.h>
110  #include <stdio.h>  #include <stdio.h>
111  #include <stdlib.h>  #include <stdlib.h>
112  #include <string.h>  #include <string.h>
113    
114  #ifndef PCRE_SPY  /* When compiling a DLL for Windows, the exported symbols have to be declared
115  #define PCRE_DEFINITION       /* Win32 __declspec(export) trigger for .dll */  using some MS magic. I found some useful information on this web page:
116    http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
117    information there, using __declspec(dllexport) without "extern" we have a
118    definition; with "extern" we have a declaration. The settings here override the
119    setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL,
120    which is all that is needed for applications (they just import the symbols). We
121    use:
122    
123      PCRE_EXP_DECL       for declarations
124      PCRE_EXP_DEFN       for definitions of exported functions
125      PCRE_EXP_DATA_DEFN  for definitions of exported variables
126    
127    The reason for the two DEFN macros is that in non-Windows environments, one
128    does not want to have "extern" before variable definitions because it leads to
129    compiler warnings. So we distinguish between functions and variables. In
130    Windows, the two should always be the same.
131    
132    The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest,
133    which is an application, but needs to import this file in order to "peek" at
134    internals, can #include pcre.h first to get an application's-eye view.
135    
136    In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
137    special-purpose environments) might want to stick other stuff in front of
138    exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and
139    PCRE_EXP_DATA_DEFN only if they are not already set. */
140    
141    #ifndef PCRE_EXP_DECL
142    #  ifdef _WIN32
143    #    ifndef PCRE_STATIC
144    #      define PCRE_EXP_DECL       extern __declspec(dllexport)
145    #      define PCRE_EXP_DEFN       __declspec(dllexport)
146    #      define PCRE_EXP_DATA_DEFN  __declspec(dllexport)
147    #    else
148    #      define PCRE_EXP_DECL       extern
149    #      define PCRE_EXP_DEFN
150    #      define PCRE_EXP_DATA_DEFN
151    #    endif
152    #  else
153    #    ifdef __cplusplus
154    #      define PCRE_EXP_DECL       extern "C"
155    #    else
156    #      define PCRE_EXP_DECL       extern
157    #    endif
158    #    ifndef PCRE_EXP_DEFN
159    #      define PCRE_EXP_DEFN       PCRE_EXP_DECL
160    #    endif
161    #    ifndef PCRE_EXP_DATA_DEFN
162    #      define PCRE_EXP_DATA_DEFN
163    #    endif
164    #  endif
165    #endif
166    
167    /* When compiling with the MSVC compiler, it is sometimes necessary to include
168    a "calling convention" before exported function names. (This is secondhand
169    information; I know nothing about MSVC myself). For example, something like
170    
171      void __cdecl function(....)
172    
173    might be needed. In order so make this easy, all the exported functions have
174    PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
175    set, we ensure here that it has no effect. */
176    
177    #ifndef PCRE_CALL_CONVENTION
178    #define PCRE_CALL_CONVENTION
179  #endif  #endif
180    
181  /* We need to have types that specify unsigned 16-bit and 32-bit integers. We  /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
182  cannot determine these outside the compilation (e.g. by running a program as  cannot determine these outside the compilation (e.g. by running a program as
183  part of "configure") because PCRE is often cross-compiled for use on other  part of "configure") because PCRE is often cross-compiled for use on other
184  systems. Instead we make use of the maximum sizes that are available at  systems. Instead we make use of the maximum sizes that are available at
185  preprocessor time in standard C environments. */  preprocessor time in standard C environments. */
186    
187    typedef unsigned char pcre_uint8;
188    
189  #if USHRT_MAX == 65535  #if USHRT_MAX == 65535
190    typedef unsigned short pcre_uint16;    typedef unsigned short pcre_uint16;
191      typedef short pcre_int16;
192  #elif UINT_MAX == 65535  #elif UINT_MAX == 65535
193    typedef unsigned int pcre_uint16;    typedef unsigned int pcre_uint16;
194      typedef int pcre_int16;
195  #else  #else
196    #error Cannot determine a type for 16-bit unsigned integers    #error Cannot determine a type for 16-bit unsigned integers
197  #endif  #endif
198    
199  #if UINT_MAX == 4294967295  #if UINT_MAX == 4294967295
200    typedef unsigned int pcre_uint32;    typedef unsigned int pcre_uint32;
201      typedef int pcre_int32;
202  #elif ULONG_MAX == 4294967295  #elif ULONG_MAX == 4294967295
203    typedef unsigned long int pcre_uint32;    typedef unsigned long int pcre_uint32;
204      typedef long int pcre_int32;
205  #else  #else
206    #error Cannot determine a type for 32-bit unsigned integers    #error Cannot determine a type for 32-bit unsigned integers
207  #endif  #endif
208    
209    /* When checking for integer overflow in pcre_compile(), we need to handle
210    large integers. If a 64-bit integer type is available, we can use that.
211    Otherwise we have to cast to double, which of course requires floating point
212    arithmetic. Handle this by defining a macro for the appropriate type. If
213    stdint.h is available, include it; it may define INT64_MAX. Systems that do not
214    have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
215    by "configure". */
216    
217    #if HAVE_STDINT_H
218    #include <stdint.h>
219    #elif HAVE_INTTYPES_H
220    #include <inttypes.h>
221    #endif
222    
223    #if defined INT64_MAX || defined int64_t
224    #define INT64_OR_DOUBLE int64_t
225    #else
226    #define INT64_OR_DOUBLE double
227    #endif
228    
229  /* All character handling must be done as unsigned characters. Otherwise there  /* All character handling must be done as unsigned characters. Otherwise there
230  are problems with top-bit-set characters and functions such as isspace().  are problems with top-bit-set characters and functions such as isspace().
231  However, we leave the interface to the outside world as char *, because that  However, we leave the interface to the outside world as char * or short *,
232  should make things easier for callers. We define a short type for unsigned char  because that should make things easier for callers. This character type is
233  to save lots of typing. I tried "uchar", but it causes problems on Digital  called pcre_uchar.
234  Unix, where it is defined in sys/types, so use "uschar" instead. */  
235    The IN_UCHARS macro multiply its argument with the byte size of the current
236  typedef unsigned char uschar;  pcre_uchar type. Useful for memcpy and such operations, whose require the
237    byte size of their input/output buffers.
238  /* PCRE is able to support 3 different kinds of newline (CR, LF, CRLF). The  
239  following macro is used to package up testing for newlines. NLBLOCK is defined  The MAX_255 macro checks whether its pcre_uchar input is less than 256.
240  in the various modules to indicate in which datablock the parameters exist. */  
241    The TABLE_GET macro is designed for accessing elements of tables whose contain
242    exactly 256 items. When the character is able to contain more than 256
243    items, some check is needed before accessing these tables.
244    */
245    
246    #ifdef COMPILE_PCRE8
247    
248    typedef unsigned char pcre_uchar;
249    #define IN_UCHARS(x) (x)
250    #define MAX_255(c) 1
251    #define TABLE_GET(c, table, default) ((table)[c])
252    
253    #else
254    
255    #ifdef COMPILE_PCRE16
256    #if USHRT_MAX != 65535
257    /* This is a warning message. Change PCRE_UCHAR16 to a 16 bit data type in
258    pcre.h(.in) and disable (comment out) this message. */
259    #error Warning: PCRE_UCHAR16 is not a 16 bit data type.
260    #endif
261    
262    typedef pcre_uint16 pcre_uchar;
263    #define IN_UCHARS(x) ((x) << 1)
264    #define MAX_255(c) ((c) <= 255u)
265    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
266    
267    #else
268    #error Unsupported compiling mode
269    #endif /* COMPILE_PCRE16 */
270    
271    #endif /* COMPILE_PCRE8 */
272    
273    /* This is an unsigned int value that no character can ever have. UTF-8
274    characters only go up to 0x7fffffff (though Unicode doesn't go beyond
275    0x0010ffff). */
276    
277    #define NOTACHAR 0xffffffff
278    
279    /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
280    "any" and "anycrlf" at present). The following macros are used to package up
281    testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
282    modules to indicate in which datablock the parameters exist, and what the
283    start/end of string field names are. */
284    
285    #define NLTYPE_FIXED    0     /* Newline is a fixed length string */
286    #define NLTYPE_ANY      1     /* Newline is any Unicode line ending */
287    #define NLTYPE_ANYCRLF  2     /* Newline is CR, LF, or CRLF */
288    
289    /* This macro checks for a newline at the given position */
290    
291  #define IS_NEWLINE(p) \  #define IS_NEWLINE(p) \
292    ((p)[0] == NLBLOCK->nl[0] && \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
293    (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]))      ((p) < NLBLOCK->PSEND && \
294         PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
295           &(NLBLOCK->nllen), utf)) \
296        : \
297        ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
298         (p)[0] == NLBLOCK->nl[0] && \
299         (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \
300        ) \
301      )
302    
303    /* This macro checks for a newline immediately preceding the given position */
304    
305    #define WAS_NEWLINE(p) \
306      ((NLBLOCK->nltype != NLTYPE_FIXED)? \
307        ((p) > NLBLOCK->PSSTART && \
308         PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
309           &(NLBLOCK->nllen), utf)) \
310        : \
311        ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
312         (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
313         (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \
314        ) \
315      )
316    
317  /* When PCRE is compiled as a C++ library, the subject pointer can be replaced  /* When PCRE is compiled as a C++ library, the subject pointer can be replaced
318  with a custom type. This makes it possible, for example, to allow pcre_exec()  with a custom type. This makes it possible, for example, to allow pcre_exec()
# Line 132  used for the external interface and appe Line 324  used for the external interface and appe
324  must begin with PCRE_. */  must begin with PCRE_. */
325    
326  #ifdef CUSTOM_SUBJECT_PTR  #ifdef CUSTOM_SUBJECT_PTR
327  #define PCRE_SPTR CUSTOM_SUBJECT_PTR  #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
 #define USPTR CUSTOM_SUBJECT_PTR  
328  #else  #else
329  #define PCRE_SPTR const char *  #define PCRE_PUCHAR const pcre_uchar *
 #define USPTR const unsigned char *  
330  #endif  #endif
331    
332  /* Include the public PCRE header and the definitions of UCP character property  /* Include the public PCRE header and the definitions of UCP character property
# Line 150  need to have their names changed. PCRE m Line 340  need to have their names changed. PCRE m
340  option on the command line. */  option on the command line. */
341    
342  #ifdef VPCOMPAT  #ifdef VPCOMPAT
343    #define strlen(s)        _strlen(s)
344  #define strncmp(s1,s2,m) _strncmp(s1,s2,m)  #define strncmp(s1,s2,m) _strncmp(s1,s2,m)
345    #define memcmp(s,c,n)    _memcmp(s,c,n)
346  #define memcpy(d,s,n)    _memcpy(d,s,n)  #define memcpy(d,s,n)    _memcpy(d,s,n)
347  #define memmove(d,s,n)   _memmove(d,s,n)  #define memmove(d,s,n)   _memmove(d,s,n)
348  #define memset(s,c,n)    _memset(s,c,n)  #define memset(s,c,n)    _memset(s,c,n)
# Line 159  option on the command line. */ Line 351  option on the command line. */
351  /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),  /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
352  define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY  define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
353  is set. Otherwise, include an emulating function for those systems that have  is set. Otherwise, include an emulating function for those systems that have
354  neither (there some non-Unix environments where this is the case). This assumes  neither (there some non-Unix environments where this is the case). */
 that all calls to memmove are moving strings upwards in store, which is the  
 case in PCRE. */  
355    
356  #if ! HAVE_MEMMOVE  #ifndef HAVE_MEMMOVE
357  #undef  memmove        /* some systems may have a macro */  #undef  memmove        /* some systems may have a macro */
358  #if HAVE_BCOPY  #ifdef HAVE_BCOPY
359  #define memmove(a, b, c) bcopy(b, a, c)  #define memmove(a, b, c) bcopy(b, a, c)
360  #else  /* HAVE_BCOPY */  #else  /* HAVE_BCOPY */
361  static void *  static void *
362  pcre_memmove(unsigned char *dest, const unsigned char *src, size_t n)  pcre_memmove(void *d, const void *s, size_t n)
363  {  {
364  size_t i;  size_t i;
365  dest += n;  unsigned char *dest = (unsigned char *)d;
366  src += n;  const unsigned char *src = (const unsigned char *)s;
367  for (i = 0; i < n; ++i) *(--dest) =  *(--src);  if (dest > src)
368  return dest;    {
369      dest += n;
370      src += n;
371      for (i = 0; i < n; ++i) *(--dest) = *(--src);
372      return (void *)dest;
373      }
374    else
375      {
376      for (i = 0; i < n; ++i) *dest++ = *src++;
377      return (void *)(dest - n);
378      }
379  }  }
380  #define memmove(a, b, c) pcre_memmove(a, b, c)  #define memmove(a, b, c) pcre_memmove(a, b, c)
381  #endif   /* not HAVE_BCOPY */  #endif   /* not HAVE_BCOPY */
# Line 196  The macros are controlled by the value o Line 396  The macros are controlled by the value o
396  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
397  is automated on Unix systems via the "configure" command. */  is automated on Unix systems via the "configure" command. */
398    
399    #ifdef COMPILE_PCRE8
400    
401  #if LINK_SIZE == 2  #if LINK_SIZE == 2
402    
403  #define PUT(a,n,d)   \  #define PUT(a,n,d)   \
# Line 232  is automated on Unix systems via the "co Line 434  is automated on Unix systems via the "co
434  #define GET(a,n) \  #define GET(a,n) \
435    (((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])
436    
437  #define MAX_PATTERN_SIZE (1 << 30)   /* Keep it positive */  /* Keep it positive */
438    #define MAX_PATTERN_SIZE (1 << 30)
439    
440    #else
441    #error LINK_SIZE must be either 2, 3, or 4
442    #endif
443    
444    #else /* COMPILE_PCRE8 */
445    
446    #ifdef COMPILE_PCRE16
447    
448    #if LINK_SIZE == 2
449    
450    #undef LINK_SIZE
451    #define LINK_SIZE 1
452    
453    #define PUT(a,n,d)   \
454      (a[n] = (d))
455    
456    #define GET(a,n) \
457      (a[n])
458    
459    #define MAX_PATTERN_SIZE (1 << 16)
460    
461    #elif LINK_SIZE == 3 || LINK_SIZE == 4
462    
463    #undef LINK_SIZE
464    #define LINK_SIZE 2
465    
466    #define PUT(a,n,d)   \
467      (a[n] = (d) >> 16), \
468      (a[(n)+1] = (d) & 65535)
469    
470    #define GET(a,n) \
471      (((a)[n] << 16) | (a)[(n)+1])
472    
473    /* Keep it positive */
474    #define MAX_PATTERN_SIZE (1 << 30)
475    
476  #else  #else
477  #error LINK_SIZE must be either 2, 3, or 4  #error LINK_SIZE must be either 2, 3, or 4
478  #endif  #endif
479    
480    #else
481    #error Unsupported compiling mode
482    #endif /* COMPILE_PCRE16 */
483    
484    #endif /* COMPILE_PCRE8 */
485    
486  /* Convenience macro defined in terms of the others */  /* Convenience macro defined in terms of the others */
487    
# Line 249  is automated on Unix systems via the "co Line 492  is automated on Unix systems via the "co
492  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
493  capturing parenthesis numbers in back references. */  capturing parenthesis numbers in back references. */
494    
495    #ifdef COMPILE_PCRE8
496    
497    #define IMM2_SIZE 2
498    
499  #define PUT2(a,n,d)   \  #define PUT2(a,n,d)   \
500    a[n] = (d) >> 8; \    a[n] = (d) >> 8; \
501    a[(n)+1] = (d) & 255    a[(n)+1] = (d) & 255
# Line 256  capturing parenthesis numbers in back re Line 503  capturing parenthesis numbers in back re
503  #define GET2(a,n) \  #define GET2(a,n) \
504    (((a)[n] << 8) | (a)[(n)+1])    (((a)[n] << 8) | (a)[(n)+1])
505    
506  #define PUT2INC(a,n,d)  PUT2(a,n,d), a += 2  #else /* COMPILE_PCRE8 */
507    
508    #ifdef COMPILE_PCRE16
509    
510  /* When UTF-8 encoding is being used, a character is no longer just a single  #define IMM2_SIZE 1
511  byte. The macros for character handling generate simple sequences when used in  
512  byte-mode, and more complicated ones for UTF-8 characters. */  #define PUT2(a,n,d)   \
513       a[n] = d
514    
515    #define GET2(a,n) \
516       a[n]
517    
518    #else
519    #error Unsupported compiling mode
520    #endif /* COMPILE_PCRE16 */
521    
522    #endif /* COMPILE_PCRE8 */
523    
524    #define PUT2INC(a,n,d)  PUT2(a,n,d), a += IMM2_SIZE
525    
526    /* The maximum length of a MARK name is currently one data unit; it may be
527    changed in future to be a fixed number of bytes or to depend on LINK_SIZE. */
528    
529  #ifndef SUPPORT_UTF8  #define MAX_MARK ((1 << (sizeof(pcre_uchar)*8)) - 1)
530    
531    /* When UTF encoding is being used, a character is no longer just a single
532    byte. The macros for character handling generate simple sequences when used in
533    character-mode, and more complicated ones for UTF characters. GETCHARLENTEST
534    and other macros are not used when UTF is not supported, so they are not
535    defined. To make sure they can never even appear when UTF support is omitted,
536    we don't even define them. */
537    
538    #ifndef SUPPORT_UTF
539    
540    /* #define MAX_VALUE_FOR_SINGLE_CHAR */
541    /* #define HAS_EXTRALEN(c) */
542    /* #define GET_EXTRALEN(c) */
543    /* #define NOT_FIRSTCHAR(c) */
544  #define GETCHAR(c, eptr) c = *eptr;  #define GETCHAR(c, eptr) c = *eptr;
545  #define GETCHARTEST(c, eptr) c = *eptr;  #define GETCHARTEST(c, eptr) c = *eptr;
546  #define GETCHARINC(c, eptr) c = *eptr++;  #define GETCHARINC(c, eptr) c = *eptr++;
547  #define GETCHARINCTEST(c, eptr) c = *eptr++;  #define GETCHARINCTEST(c, eptr) c = *eptr++;
548  #define GETCHARLEN(c, eptr, len) c = *eptr;  #define GETCHARLEN(c, eptr, len) c = *eptr;
549  #define BACKCHAR(eptr)  /* #define GETCHARLENTEST(c, eptr, len) */
550    /* #define BACKCHAR(eptr) */
551    /* #define FORWARDCHAR(eptr) */
552    /* #define ACROSSCHAR(condition, eptr, action) */
553    
554    #else   /* SUPPORT_UTF */
555    
556    #ifdef COMPILE_PCRE8
557    
558    /* These macros were originally written in the form of loops that used data
559    from the tables whose names start with PRIV(utf8_table). They were rewritten by
560    a user so as not to use loops, because in some environments this gives a
561    significant performance advantage, and it seems never to do any harm. */
562    
563    /* Tells the biggest code point which can be encoded as a single character. */
564    
565    #define MAX_VALUE_FOR_SINGLE_CHAR 127
566    
567    /* Tests whether the code point needs extra characters to decode. */
568    
569  #else   /* SUPPORT_UTF8 */  #define HAS_EXTRALEN(c) ((c) >= 0xc0)
570    
571    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
572    Otherwise it has an undefined behaviour. */
573    
574    #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
575    
576    /* Returns TRUE, if the given character is not the first character
577    of a UTF sequence. */
578    
579    #define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80)
580    
581    /* Base macro to pick up the remaining bytes of a UTF-8 character, not
582    advancing the pointer. */
583    
584    #define GETUTF8(c, eptr) \
585        { \
586        if ((c & 0x20) == 0) \
587          c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
588        else if ((c & 0x10) == 0) \
589          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
590        else if ((c & 0x08) == 0) \
591          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
592          ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
593        else if ((c & 0x04) == 0) \
594          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
595              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
596              (eptr[4] & 0x3f); \
597        else \
598          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
599              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
600              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
601        }
602    
603  /* Get the next UTF-8 character, not advancing the pointer. This is called when  /* Get the next UTF-8 character, not advancing the pointer. This is called when
604  we know we are in UTF-8 mode. */  we know we are in UTF-8 mode. */
605    
606  #define GETCHAR(c, eptr) \  #define GETCHAR(c, eptr) \
607    c = *eptr; \    c = *eptr; \
608    if ((c & 0xc0) == 0xc0) \    if (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; \  
       } \  
     }  
609    
610  /* 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
611  pointer. */  pointer. */
612    
613  #define GETCHARTEST(c, eptr) \  #define GETCHARTEST(c, eptr) \
614    c = *eptr; \    c = *eptr; \
615    if (utf8 && (c & 0xc0) == 0xc0) \    if (utf && c >= 0xc0) GETUTF8(c, eptr);
616    
617    /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
618    the pointer. */
619    
620    #define GETUTF8INC(c, eptr) \
621      { \      { \
622      int gcii; \      if ((c & 0x20) == 0) \
623      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \        c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
624      int gcss = 6*gcaa; \      else if ((c & 0x10) == 0) \
625      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        { \
626      for (gcii = 1; gcii <= gcaa; gcii++) \        c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
627          eptr += 2; \
628          } \
629        else if ((c & 0x08) == 0) \
630          { \
631          c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
632              ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
633          eptr += 3; \
634          } \
635        else if ((c & 0x04) == 0) \
636          { \
637          c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
638              ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
639              (eptr[3] & 0x3f); \
640          eptr += 4; \
641          } \
642        else \
643        { \        { \
644        gcss -= 6; \        c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
645        c |= (eptr[gcii] & 0x3f) << gcss; \            ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
646              ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
647          eptr += 5; \
648        } \        } \
649      }      }
650    
# Line 314  know we are in UTF-8 mode. */ Line 653  know we are in UTF-8 mode. */
653    
654  #define GETCHARINC(c, eptr) \  #define GETCHARINC(c, eptr) \
655    c = *eptr++; \    c = *eptr++; \
656    if ((c & 0xc0) == 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; \  
       } \  
     }  
657    
658  /* 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.
659    This is called when we don't know if we are in UTF-8 mode. */
660    
661  #define GETCHARINCTEST(c, eptr) \  #define GETCHARINCTEST(c, eptr) \
662    c = *eptr++; \    c = *eptr++; \
663    if (utf8 && (c & 0xc0) == 0xc0) \    if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
664    
665    /* Base macro to pick up the remaining bytes of a UTF-8 character, not
666    advancing the pointer, incrementing the length. */
667    
668    #define GETUTF8LEN(c, eptr, len) \
669      { \      { \
670      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \      if ((c & 0x20) == 0) \
671      int gcss = 6*gcaa; \        { \
672      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
673      while (gcaa-- > 0) \        len++; \
674          } \
675        else if ((c & 0x10)  == 0) \
676          { \
677          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
678          len += 2; \
679          } \
680        else if ((c & 0x08)  == 0) \
681          {\
682          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
683              ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
684          len += 3; \
685          } \
686        else if ((c & 0x04)  == 0) \
687        { \        { \
688        gcss -= 6; \        c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
689        c |= (*eptr++ & 0x3f) << gcss; \            ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
690              (eptr[4] & 0x3f); \
691          len += 4; \
692          } \
693        else \
694          {\
695          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
696              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
697              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
698          len += 5; \
699        } \        } \
700      }      }
701    
# Line 347  if there are extra bytes. This is called Line 704  if there are extra bytes. This is called
704    
705  #define GETCHARLEN(c, eptr, len) \  #define GETCHARLEN(c, eptr, len) \
706    c = *eptr; \    c = *eptr; \
707    if ((c & 0xc0) == 0xc0) \    if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
708      { \  
709      int gcii; \  /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
710      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  pointer, incrementing length if there are extra bytes. This is called when we
711      int gcss = 6*gcaa; \  do not know if we are in UTF-8 mode. */
712      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
713      for (gcii = 1; gcii <= gcaa; gcii++) \  #define GETCHARLENTEST(c, eptr, len) \
714        { \    c = *eptr; \
715        gcss -= 6; \    if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
       c |= (eptr[gcii] & 0x3f) << gcss; \  
       } \  
     len += gcaa; \  
     }  
716    
717  /* 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
718  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
719    because almost all calls are already within a block of UTF-8 only code. */
720    
721    #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
722    
723    /* Same as above, just in the other direction. */
724    #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
725    
726    /* Same as above, but it allows a fully customizable form. */
727    #define ACROSSCHAR(condition, eptr, action) \
728      while((condition) && ((eptr) & 0xc0) == 0x80) action
729    
730    #else /* COMPILE_PCRE8 */
731    
732    #ifdef COMPILE_PCRE16
733    
734    /* Tells the biggest code point which can be encoded as a single character. */
735    
736    #define MAX_VALUE_FOR_SINGLE_CHAR 65535
737    
738    /* Tests whether the code point needs extra characters to decode. */
739    
740    #define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
741    
742    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
743    Otherwise it has an undefined behaviour. */
744    
745    #define GET_EXTRALEN(c) 1
746    
747    /* Returns TRUE, if the given character is not the first character
748    of a UTF sequence. */
749    
750    #define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00)
751    
752    /* Base macro to pick up the low surrogate of a UTF-16 character, not
753    advancing the pointer. */
754    
755    #define GETUTF16(c, eptr) \
756       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
757    
758    /* Get the next UTF-16 character, not advancing the pointer. This is called when
759    we know we are in UTF-16 mode. */
760    
761    #define GETCHAR(c, eptr) \
762      c = *eptr; \
763      if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
764    
765    /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
766    pointer. */
767    
768    #define GETCHARTEST(c, eptr) \
769      c = *eptr; \
770      if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
771    
772    /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
773    the pointer. */
774    
775    #define GETUTF16INC(c, eptr) \
776       { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
777    
778    /* Get the next UTF-16 character, advancing the pointer. This is called when we
779    know we are in UTF-16 mode. */
780    
781    #define GETCHARINC(c, eptr) \
782      c = *eptr++; \
783      if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
784    
785  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--;  /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
786    This is called when we don't know if we are in UTF-16 mode. */
787    
788    #define GETCHARINCTEST(c, eptr) \
789      c = *eptr++; \
790      if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
791    
792    /* Base macro to pick up the low surrogate of a UTF-16 character, not
793    advancing the pointer, incrementing the length. */
794    
795    #define GETUTF16LEN(c, eptr, len) \
796       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
797    
798    /* Get the next UTF-16 character, not advancing the pointer, incrementing
799    length if there is a low surrogate. This is called when we know we are in
800    UTF-16 mode. */
801    
802    #define GETCHARLEN(c, eptr, len) \
803      c = *eptr; \
804      if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
805    
806    /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
807    pointer, incrementing length if there is a low surrogate. This is called when
808    we do not know if we are in UTF-16 mode. */
809    
810    #define GETCHARLENTEST(c, eptr, len) \
811      c = *eptr; \
812      if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
813    
814    /* If the pointer is not at the start of a character, move it back until
815    it is. This is called only in UTF-16 mode - we don't put a test within the
816    macro because almost all calls are already within a block of UTF-16 only
817    code. */
818    
819    #define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
820    
821    /* Same as above, just in the other direction. */
822    #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
823    
824    /* Same as above, but it allows a fully customizable form. */
825    #define ACROSSCHAR(condition, eptr, action) \
826      if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
827    
828  #endif  #endif
829    
830    #endif /* COMPILE_PCRE8 */
831    
832    #endif  /* SUPPORT_UTF */
833    
834    
835    /* Tests for Unicode horizontal and vertical whitespace characters must check a
836    number of different values. Using a switch statement for this generates the
837    fastest code (no loop, no memory access), and there are several places in the
838    interpreter code where this happens. In order to ensure that all the case lists
839    remain in step, we use macros so that there is only one place where the lists
840    are defined.
841    
842    These values are also required as lists in pcre_compile.c when processing \h,
843    \H, \v and \V in a character class. The lists are defined in pcre_tables.c, but
844    macros that define the values are here so that all the definitions are
845    together. The lists must be in ascending character order, terminated by
846    NOTACHAR (which is 0xffffffff).
847    
848    Any changes should ensure that the various macros are kept in step with each
849    other. NOTE: The values also appear in pcre_jit_compile.c. */
850    
851    /* ------ ASCII/Unicode environments ------ */
852    
853    #ifndef EBCDIC
854    
855    #define HSPACE_LIST \
856      CHAR_HT, CHAR_SPACE, 0xa0, \
857      0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \
858      0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \
859      NOTACHAR
860    
861    #define HSPACE_MULTIBYTE_CASES \
862      case 0x1680:  /* OGHAM SPACE MARK */ \
863      case 0x180e:  /* MONGOLIAN VOWEL SEPARATOR */ \
864      case 0x2000:  /* EN QUAD */ \
865      case 0x2001:  /* EM QUAD */ \
866      case 0x2002:  /* EN SPACE */ \
867      case 0x2003:  /* EM SPACE */ \
868      case 0x2004:  /* THREE-PER-EM SPACE */ \
869      case 0x2005:  /* FOUR-PER-EM SPACE */ \
870      case 0x2006:  /* SIX-PER-EM SPACE */ \
871      case 0x2007:  /* FIGURE SPACE */ \
872      case 0x2008:  /* PUNCTUATION SPACE */ \
873      case 0x2009:  /* THIN SPACE */ \
874      case 0x200A:  /* HAIR SPACE */ \
875      case 0x202f:  /* NARROW NO-BREAK SPACE */ \
876      case 0x205f:  /* MEDIUM MATHEMATICAL SPACE */ \
877      case 0x3000   /* IDEOGRAPHIC SPACE */
878    
879    #define HSPACE_BYTE_CASES \
880      case CHAR_HT: \
881      case CHAR_SPACE: \
882      case 0xa0     /* NBSP */
883    
884    #define HSPACE_CASES \
885      HSPACE_BYTE_CASES: \
886      HSPACE_MULTIBYTE_CASES
887    
888    #define VSPACE_LIST \
889      CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR
890    
891    #define VSPACE_MULTIBYTE_CASES \
892      case 0x2028:    /* LINE SEPARATOR */ \
893      case 0x2029     /* PARAGRAPH SEPARATOR */
894    
895    #define VSPACE_BYTE_CASES \
896      case CHAR_LF: \
897      case CHAR_VT: \
898      case CHAR_FF: \
899      case CHAR_CR: \
900      case CHAR_NEL
901    
902    #define VSPACE_CASES \
903      VSPACE_BYTE_CASES: \
904      VSPACE_MULTIBYTE_CASES
905    
906    /* ------ EBCDIC environments ------ */
907    
908    #else
909    #define HSPACE_LIST CHAR_HT, CHAR_SPACE
910    
911    #define HSPACE_BYTE_CASES \
912      case CHAR_HT: \
913      case CHAR_SPACE
914    
915    #define HSPACE_CASES HSPACE_BYTE_CASES
916    
917    #ifdef EBCDIC_NL25
918    #define VSPACE_LIST \
919      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR
920    #else
921    #define VSPACE_LIST \
922      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR
923    #endif
924    
925    #define VSPACE_BYTE_CASES \
926      case CHAR_LF: \
927      case CHAR_VT: \
928      case CHAR_FF: \
929      case CHAR_CR: \
930      case CHAR_NEL
931    
932    #define VSPACE_CASES VSPACE_BYTE_CASES
933    #endif  /* EBCDIC */
934    
935    /* ------ End of whitespace macros ------ */
936    
937    
938  /* In case there is no definition of offsetof() provided - though any proper  /* In case there is no definition of offsetof() provided - though any proper
939  Standard C system should have one. */  Standard C system should have one. */
# Line 377  Standard C system should have one. */ Line 943  Standard C system should have one. */
943  #endif  #endif
944    
945    
946  /* These are the public options that can change during matching. */  /* Private flags containing information about the compiled regex. They used to
947    live at the top end of the options word, but that got almost full, so now they
948  #define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)  are in a 16-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as
949    the restrictions on partial matching have been lifted. It remains for backwards
950    compatibility. */
951    
952  /* Private options flags start at the most significant end of the four bytes.  #ifdef COMPILE_PCRE8
953  The public options defined in pcre.h start at the least significant end. Make  #define PCRE_MODE          0x0001  /* compiled in 8 bit mode */
954  sure they don't overlap! The bits are getting a bit scarce now -- when we run  #endif
955  out, there is a dummy word in the structure that could be used for the private  #ifdef COMPILE_PCRE16
956  bits. */  #define PCRE_MODE          0x0002  /* compiled in 16 bit mode */
957    #endif
958  #define PCRE_NOPARTIAL     0x80000000  /* can't use partial with this regex */  #define PCRE_FIRSTSET      0x0010  /* first_char is set */
959  #define PCRE_FIRSTSET      0x40000000  /* first_byte is set */  #define PCRE_FCH_CASELESS  0x0020  /* caseless first char */
960  #define PCRE_REQCHSET      0x20000000  /* req_byte is set */  #define PCRE_REQCHSET      0x0040  /* req_byte is set */
961  #define PCRE_STARTLINE     0x10000000  /* start after \n for multiline */  #define PCRE_RCH_CASELESS  0x0080  /* caseless requested char */
962  #define PCRE_JCHANGED      0x08000000  /* j option changes within regex */  #define PCRE_STARTLINE     0x0100  /* start after \n for multiline */
963    #define PCRE_NOPARTIAL     0x0200  /* can't use partial with this regex */
964    #define PCRE_JCHANGED      0x0400  /* j option used in regex */
965    #define PCRE_HASCRORLF     0x0800  /* explicit \r or \n in pattern */
966    #define PCRE_HASTHEN       0x1000  /* pattern contains (*THEN) */
967    
968  /* Options for the "extra" block produced by pcre_study(). */  /* Flags for the "extra" block produced by pcre_study(). */
969    
970  #define PCRE_STUDY_MAPPED   0x01     /* a map of starting chars exists */  #define PCRE_STUDY_MAPPED  0x0001  /* a map of starting chars exists */
971    #define PCRE_STUDY_MINLEN  0x0002  /* a minimum length field exists */
972    
973  /* Masks for identifying the public options that are permitted at compile  /* Masks for identifying the public options that are permitted at compile
974  time, run time, or study time, respectively. */  time, run time, or study time, respectively. */
975    
976  #define PUBLIC_OPTIONS \  #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
977                               PCRE_NEWLINE_ANYCRLF)
978    
979    #define PUBLIC_COMPILE_OPTIONS \
980    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
981     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
982     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| \
983     PCRE_DUPNAMES|PCRE_NEWLINE_CR|PCRE_NEWLINE_LF)     PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
984       PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE)
985    
986  #define PUBLIC_EXEC_OPTIONS \  #define PUBLIC_EXEC_OPTIONS \
987    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
988     PCRE_PARTIAL|PCRE_NEWLINE_CR|PCRE_NEWLINE_LF)     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
989       PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
990    
991  #define PUBLIC_DFA_EXEC_OPTIONS \  #define PUBLIC_DFA_EXEC_OPTIONS \
992    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
993     PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_CR| \     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
994     PCRE_NEWLINE_LF)     PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
995       PCRE_NO_START_OPTIMIZE)
996  #define PUBLIC_STUDY_OPTIONS 0   /* None defined */  
997    #define PUBLIC_STUDY_OPTIONS \
998       (PCRE_STUDY_JIT_COMPILE|PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE| \
999        PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE|PCRE_STUDY_EXTRA_NEEDED)
1000    
1001  /* Magic number to provide a small check against being handed junk. Also used  /* Magic number to provide a small check against being handed junk. */
 to detect whether a pattern was compiled on a host of different endianness. */  
1002    
1003  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */
1004    
1005    /* This variable is used to detect a loaded regular expression
1006    in different endianness. */
1007    
1008    #define REVERSED_MAGIC_NUMBER  0x45524350UL   /* 'ERCP' */
1009    
1010  /* Negative values for the firstchar and reqchar variables */  /* Negative values for the firstchar and reqchar variables */
1011    
1012  #define REQ_UNSET (-2)  #define REQ_UNSET (-2)
# Line 432  req_byte match. */ Line 1017  req_byte match. */
1017    
1018  #define REQ_BYTE_MAX 1000  #define REQ_BYTE_MAX 1000
1019    
1020  /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a  /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
1021  variable-length repeat, or a anything other than literal characters. */  environments where these macros are defined elsewhere. Unfortunately, there
1022    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 */  
1023    
1024  typedef int BOOL;  typedef int BOOL;
1025    
1026    #ifndef FALSE
1027  #define FALSE   0  #define FALSE   0
1028  #define TRUE    1  #define TRUE    1
1029    #endif
1030    
1031    /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
1032    character constants like '*' because the compiler would emit their EBCDIC code,
1033    which is different from their ASCII/UTF-8 code. Instead we define macros for
1034    the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
1035    is enabled. When UTF-8 support is not enabled, the definitions use character
1036    literals. Both character and string versions of each character are needed, and
1037    there are some longer strings as well.
1038    
1039    This means that, on EBCDIC platforms, the PCRE library can handle either
1040    EBCDIC, or UTF-8, but not both. To support both in the same compiled library
1041    would need different lookups depending on whether PCRE_UTF8 was set or not.
1042    This would make it impossible to use characters in switch/case statements,
1043    which would reduce performance. For a theoretical use (which nobody has asked
1044    for) in a minority area (EBCDIC platforms), this is not sensible. Any
1045    application that did need both could compile two versions of the library, using
1046    macros to give the functions distinct names. */
1047    
1048    #ifndef SUPPORT_UTF
1049    
1050    /* UTF-8 support is not enabled; use the platform-dependent character literals
1051    so that PCRE works in both ASCII and EBCDIC environments, but only in non-UTF
1052    mode. Newline characters are problematic in EBCDIC. Though it has CR and LF
1053    characters, a common practice has been to use its NL (0x15) character as the
1054    line terminator in C-like processing environments. However, sometimes the LF
1055    (0x25) character is used instead, according to this Unicode document:
1056    
1057    http://unicode.org/standard/reports/tr13/tr13-5.html
1058    
1059    PCRE defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25
1060    instead. Whichever is *not* chosen is defined as NEL.
1061    
1062    In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the
1063    same code point. */
1064    
1065    #ifdef EBCDIC
1066    
1067    #ifndef EBCDIC_NL25
1068    #define CHAR_NL                     '\x15'
1069    #define CHAR_NEL                    '\x25'
1070    #define STR_NL                      "\x15"
1071    #define STR_NEL                     "\x25"
1072    #else
1073    #define CHAR_NL                     '\x25'
1074    #define CHAR_NEL                    '\x15'
1075    #define STR_NL                      "\x25"
1076    #define STR_NEL                     "\x15"
1077    #endif
1078    
1079    #define CHAR_LF                     CHAR_NL
1080    #define STR_LF                      STR_NL
1081    
1082    #define CHAR_ESC                    '\047'
1083    #define CHAR_DEL                    '\007'
1084    #define STR_ESC                     "\047"
1085    #define STR_DEL                     "\007"
1086    
1087    #else  /* Not EBCDIC */
1088    
1089    /* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for
1090    compatibility. NEL is the Unicode newline character; make sure it is
1091    a positive value. */
1092    
1093    #define CHAR_LF                     '\n'
1094    #define CHAR_NL                     CHAR_LF
1095    #define CHAR_NEL                    ((unsigned char)'\x85')
1096    #define CHAR_ESC                    '\033'
1097    #define CHAR_DEL                    '\177'
1098    
1099    #define STR_LF                      "\n"
1100    #define STR_NL                      STR_LF
1101    #define STR_NEL                     "\x85"
1102    #define STR_ESC                     "\033"
1103    #define STR_DEL                     "\177"
1104    
1105    #endif  /* EBCDIC */
1106    
1107    /* The remaining definitions work in both environments. */
1108    
1109    #define CHAR_HT                     '\t'
1110    #define CHAR_VT                     '\v'
1111    #define CHAR_FF                     '\f'
1112    #define CHAR_CR                     '\r'
1113    #define CHAR_BS                     '\b'
1114    #define CHAR_BEL                    '\a'
1115    
1116    #define CHAR_SPACE                  ' '
1117    #define CHAR_EXCLAMATION_MARK       '!'
1118    #define CHAR_QUOTATION_MARK         '"'
1119    #define CHAR_NUMBER_SIGN            '#'
1120    #define CHAR_DOLLAR_SIGN            '$'
1121    #define CHAR_PERCENT_SIGN           '%'
1122    #define CHAR_AMPERSAND              '&'
1123    #define CHAR_APOSTROPHE             '\''
1124    #define CHAR_LEFT_PARENTHESIS       '('
1125    #define CHAR_RIGHT_PARENTHESIS      ')'
1126    #define CHAR_ASTERISK               '*'
1127    #define CHAR_PLUS                   '+'
1128    #define CHAR_COMMA                  ','
1129    #define CHAR_MINUS                  '-'
1130    #define CHAR_DOT                    '.'
1131    #define CHAR_SLASH                  '/'
1132    #define CHAR_0                      '0'
1133    #define CHAR_1                      '1'
1134    #define CHAR_2                      '2'
1135    #define CHAR_3                      '3'
1136    #define CHAR_4                      '4'
1137    #define CHAR_5                      '5'
1138    #define CHAR_6                      '6'
1139    #define CHAR_7                      '7'
1140    #define CHAR_8                      '8'
1141    #define CHAR_9                      '9'
1142    #define CHAR_COLON                  ':'
1143    #define CHAR_SEMICOLON              ';'
1144    #define CHAR_LESS_THAN_SIGN         '<'
1145    #define CHAR_EQUALS_SIGN            '='
1146    #define CHAR_GREATER_THAN_SIGN      '>'
1147    #define CHAR_QUESTION_MARK          '?'
1148    #define CHAR_COMMERCIAL_AT          '@'
1149    #define CHAR_A                      'A'
1150    #define CHAR_B                      'B'
1151    #define CHAR_C                      'C'
1152    #define CHAR_D                      'D'
1153    #define CHAR_E                      'E'
1154    #define CHAR_F                      'F'
1155    #define CHAR_G                      'G'
1156    #define CHAR_H                      'H'
1157    #define CHAR_I                      'I'
1158    #define CHAR_J                      'J'
1159    #define CHAR_K                      'K'
1160    #define CHAR_L                      'L'
1161    #define CHAR_M                      'M'
1162    #define CHAR_N                      'N'
1163    #define CHAR_O                      'O'
1164    #define CHAR_P                      'P'
1165    #define CHAR_Q                      'Q'
1166    #define CHAR_R                      'R'
1167    #define CHAR_S                      'S'
1168    #define CHAR_T                      'T'
1169    #define CHAR_U                      'U'
1170    #define CHAR_V                      'V'
1171    #define CHAR_W                      'W'
1172    #define CHAR_X                      'X'
1173    #define CHAR_Y                      'Y'
1174    #define CHAR_Z                      'Z'
1175    #define CHAR_LEFT_SQUARE_BRACKET    '['
1176    #define CHAR_BACKSLASH              '\\'
1177    #define CHAR_RIGHT_SQUARE_BRACKET   ']'
1178    #define CHAR_CIRCUMFLEX_ACCENT      '^'
1179    #define CHAR_UNDERSCORE             '_'
1180    #define CHAR_GRAVE_ACCENT           '`'
1181    #define CHAR_a                      'a'
1182    #define CHAR_b                      'b'
1183    #define CHAR_c                      'c'
1184    #define CHAR_d                      'd'
1185    #define CHAR_e                      'e'
1186    #define CHAR_f                      'f'
1187    #define CHAR_g                      'g'
1188    #define CHAR_h                      'h'
1189    #define CHAR_i                      'i'
1190    #define CHAR_j                      'j'
1191    #define CHAR_k                      'k'
1192    #define CHAR_l                      'l'
1193    #define CHAR_m                      'm'
1194    #define CHAR_n                      'n'
1195    #define CHAR_o                      'o'
1196    #define CHAR_p                      'p'
1197    #define CHAR_q                      'q'
1198    #define CHAR_r                      'r'
1199    #define CHAR_s                      's'
1200    #define CHAR_t                      't'
1201    #define CHAR_u                      'u'
1202    #define CHAR_v                      'v'
1203    #define CHAR_w                      'w'
1204    #define CHAR_x                      'x'
1205    #define CHAR_y                      'y'
1206    #define CHAR_z                      'z'
1207    #define CHAR_LEFT_CURLY_BRACKET     '{'
1208    #define CHAR_VERTICAL_LINE          '|'
1209    #define CHAR_RIGHT_CURLY_BRACKET    '}'
1210    #define CHAR_TILDE                  '~'
1211    
1212    #define STR_HT                      "\t"
1213    #define STR_VT                      "\v"
1214    #define STR_FF                      "\f"
1215    #define STR_CR                      "\r"
1216    #define STR_BS                      "\b"
1217    #define STR_BEL                     "\a"
1218    
1219    #define STR_SPACE                   " "
1220    #define STR_EXCLAMATION_MARK        "!"
1221    #define STR_QUOTATION_MARK          "\""
1222    #define STR_NUMBER_SIGN             "#"
1223    #define STR_DOLLAR_SIGN             "$"
1224    #define STR_PERCENT_SIGN            "%"
1225    #define STR_AMPERSAND               "&"
1226    #define STR_APOSTROPHE              "'"
1227    #define STR_LEFT_PARENTHESIS        "("
1228    #define STR_RIGHT_PARENTHESIS       ")"
1229    #define STR_ASTERISK                "*"
1230    #define STR_PLUS                    "+"
1231    #define STR_COMMA                   ","
1232    #define STR_MINUS                   "-"
1233    #define STR_DOT                     "."
1234    #define STR_SLASH                   "/"
1235    #define STR_0                       "0"
1236    #define STR_1                       "1"
1237    #define STR_2                       "2"
1238    #define STR_3                       "3"
1239    #define STR_4                       "4"
1240    #define STR_5                       "5"
1241    #define STR_6                       "6"
1242    #define STR_7                       "7"
1243    #define STR_8                       "8"
1244    #define STR_9                       "9"
1245    #define STR_COLON                   ":"
1246    #define STR_SEMICOLON               ";"
1247    #define STR_LESS_THAN_SIGN          "<"
1248    #define STR_EQUALS_SIGN             "="
1249    #define STR_GREATER_THAN_SIGN       ">"
1250    #define STR_QUESTION_MARK           "?"
1251    #define STR_COMMERCIAL_AT           "@"
1252    #define STR_A                       "A"
1253    #define STR_B                       "B"
1254    #define STR_C                       "C"
1255    #define STR_D                       "D"
1256    #define STR_E                       "E"
1257    #define STR_F                       "F"
1258    #define STR_G                       "G"
1259    #define STR_H                       "H"
1260    #define STR_I                       "I"
1261    #define STR_J                       "J"
1262    #define STR_K                       "K"
1263    #define STR_L                       "L"
1264    #define STR_M                       "M"
1265    #define STR_N                       "N"
1266    #define STR_O                       "O"
1267    #define STR_P                       "P"
1268    #define STR_Q                       "Q"
1269    #define STR_R                       "R"
1270    #define STR_S                       "S"
1271    #define STR_T                       "T"
1272    #define STR_U                       "U"
1273    #define STR_V                       "V"
1274    #define STR_W                       "W"
1275    #define STR_X                       "X"
1276    #define STR_Y                       "Y"
1277    #define STR_Z                       "Z"
1278    #define STR_LEFT_SQUARE_BRACKET     "["
1279    #define STR_BACKSLASH               "\\"
1280    #define STR_RIGHT_SQUARE_BRACKET    "]"
1281    #define STR_CIRCUMFLEX_ACCENT       "^"
1282    #define STR_UNDERSCORE              "_"
1283    #define STR_GRAVE_ACCENT            "`"
1284    #define STR_a                       "a"
1285    #define STR_b                       "b"
1286    #define STR_c                       "c"
1287    #define STR_d                       "d"
1288    #define STR_e                       "e"
1289    #define STR_f                       "f"
1290    #define STR_g                       "g"
1291    #define STR_h                       "h"
1292    #define STR_i                       "i"
1293    #define STR_j                       "j"
1294    #define STR_k                       "k"
1295    #define STR_l                       "l"
1296    #define STR_m                       "m"
1297    #define STR_n                       "n"
1298    #define STR_o                       "o"
1299    #define STR_p                       "p"
1300    #define STR_q                       "q"
1301    #define STR_r                       "r"
1302    #define STR_s                       "s"
1303    #define STR_t                       "t"
1304    #define STR_u                       "u"
1305    #define STR_v                       "v"
1306    #define STR_w                       "w"
1307    #define STR_x                       "x"
1308    #define STR_y                       "y"
1309    #define STR_z                       "z"
1310    #define STR_LEFT_CURLY_BRACKET      "{"
1311    #define STR_VERTICAL_LINE           "|"
1312    #define STR_RIGHT_CURLY_BRACKET     "}"
1313    #define STR_TILDE                   "~"
1314    
1315    #define STRING_ACCEPT0              "ACCEPT\0"
1316    #define STRING_COMMIT0              "COMMIT\0"
1317    #define STRING_F0                   "F\0"
1318    #define STRING_FAIL0                "FAIL\0"
1319    #define STRING_MARK0                "MARK\0"
1320    #define STRING_PRUNE0               "PRUNE\0"
1321    #define STRING_SKIP0                "SKIP\0"
1322    #define STRING_THEN                 "THEN"
1323    
1324    #define STRING_alpha0               "alpha\0"
1325    #define STRING_lower0               "lower\0"
1326    #define STRING_upper0               "upper\0"
1327    #define STRING_alnum0               "alnum\0"
1328    #define STRING_ascii0               "ascii\0"
1329    #define STRING_blank0               "blank\0"
1330    #define STRING_cntrl0               "cntrl\0"
1331    #define STRING_digit0               "digit\0"
1332    #define STRING_graph0               "graph\0"
1333    #define STRING_print0               "print\0"
1334    #define STRING_punct0               "punct\0"
1335    #define STRING_space0               "space\0"
1336    #define STRING_word0                "word\0"
1337    #define STRING_xdigit               "xdigit"
1338    
1339    #define STRING_DEFINE               "DEFINE"
1340    
1341    #define STRING_CR_RIGHTPAR             "CR)"
1342    #define STRING_LF_RIGHTPAR             "LF)"
1343    #define STRING_CRLF_RIGHTPAR           "CRLF)"
1344    #define STRING_ANY_RIGHTPAR            "ANY)"
1345    #define STRING_ANYCRLF_RIGHTPAR        "ANYCRLF)"
1346    #define STRING_BSR_ANYCRLF_RIGHTPAR    "BSR_ANYCRLF)"
1347    #define STRING_BSR_UNICODE_RIGHTPAR    "BSR_UNICODE)"
1348    #ifdef COMPILE_PCRE8
1349    #define STRING_UTF_RIGHTPAR            "UTF8)"
1350    #endif
1351    #ifdef COMPILE_PCRE16
1352    #define STRING_UTF_RIGHTPAR            "UTF16)"
1353    #endif
1354    #define STRING_UCP_RIGHTPAR            "UCP)"
1355    #define STRING_NO_START_OPT_RIGHTPAR   "NO_START_OPT)"
1356    
1357  /* Escape items that are just an encoding of a particular data value. Note that  #else  /* SUPPORT_UTF */
1358  ESC_n is defined as yet another macro, which is set in config.h to either \n  
1359  (the default) or \r (which some people want). */  /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1360    works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1361    only. */
1362    
1363    #define CHAR_HT                     '\011'
1364    #define CHAR_VT                     '\013'
1365    #define CHAR_FF                     '\014'
1366    #define CHAR_CR                     '\015'
1367    #define CHAR_LF                     '\012'
1368    #define CHAR_NL                     CHAR_LF
1369    #define CHAR_NEL                    ((unsigned char)'\x85')
1370    #define CHAR_BS                     '\010'
1371    #define CHAR_BEL                    '\007'
1372    #define CHAR_ESC                    '\033'
1373    #define CHAR_DEL                    '\177'
1374    
1375    #define CHAR_SPACE                  '\040'
1376    #define CHAR_EXCLAMATION_MARK       '\041'
1377    #define CHAR_QUOTATION_MARK         '\042'
1378    #define CHAR_NUMBER_SIGN            '\043'
1379    #define CHAR_DOLLAR_SIGN            '\044'
1380    #define CHAR_PERCENT_SIGN           '\045'
1381    #define CHAR_AMPERSAND              '\046'
1382    #define CHAR_APOSTROPHE             '\047'
1383    #define CHAR_LEFT_PARENTHESIS       '\050'
1384    #define CHAR_RIGHT_PARENTHESIS      '\051'
1385    #define CHAR_ASTERISK               '\052'
1386    #define CHAR_PLUS                   '\053'
1387    #define CHAR_COMMA                  '\054'
1388    #define CHAR_MINUS                  '\055'
1389    #define CHAR_DOT                    '\056'
1390    #define CHAR_SLASH                  '\057'
1391    #define CHAR_0                      '\060'
1392    #define CHAR_1                      '\061'
1393    #define CHAR_2                      '\062'
1394    #define CHAR_3                      '\063'
1395    #define CHAR_4                      '\064'
1396    #define CHAR_5                      '\065'
1397    #define CHAR_6                      '\066'
1398    #define CHAR_7                      '\067'
1399    #define CHAR_8                      '\070'
1400    #define CHAR_9                      '\071'
1401    #define CHAR_COLON                  '\072'
1402    #define CHAR_SEMICOLON              '\073'
1403    #define CHAR_LESS_THAN_SIGN         '\074'
1404    #define CHAR_EQUALS_SIGN            '\075'
1405    #define CHAR_GREATER_THAN_SIGN      '\076'
1406    #define CHAR_QUESTION_MARK          '\077'
1407    #define CHAR_COMMERCIAL_AT          '\100'
1408    #define CHAR_A                      '\101'
1409    #define CHAR_B                      '\102'
1410    #define CHAR_C                      '\103'
1411    #define CHAR_D                      '\104'
1412    #define CHAR_E                      '\105'
1413    #define CHAR_F                      '\106'
1414    #define CHAR_G                      '\107'
1415    #define CHAR_H                      '\110'
1416    #define CHAR_I                      '\111'
1417    #define CHAR_J                      '\112'
1418    #define CHAR_K                      '\113'
1419    #define CHAR_L                      '\114'
1420    #define CHAR_M                      '\115'
1421    #define CHAR_N                      '\116'
1422    #define CHAR_O                      '\117'
1423    #define CHAR_P                      '\120'
1424    #define CHAR_Q                      '\121'
1425    #define CHAR_R                      '\122'
1426    #define CHAR_S                      '\123'
1427    #define CHAR_T                      '\124'
1428    #define CHAR_U                      '\125'
1429    #define CHAR_V                      '\126'
1430    #define CHAR_W                      '\127'
1431    #define CHAR_X                      '\130'
1432    #define CHAR_Y                      '\131'
1433    #define CHAR_Z                      '\132'
1434    #define CHAR_LEFT_SQUARE_BRACKET    '\133'
1435    #define CHAR_BACKSLASH              '\134'
1436    #define CHAR_RIGHT_SQUARE_BRACKET   '\135'
1437    #define CHAR_CIRCUMFLEX_ACCENT      '\136'
1438    #define CHAR_UNDERSCORE             '\137'
1439    #define CHAR_GRAVE_ACCENT           '\140'
1440    #define CHAR_a                      '\141'
1441    #define CHAR_b                      '\142'
1442    #define CHAR_c                      '\143'
1443    #define CHAR_d                      '\144'
1444    #define CHAR_e                      '\145'
1445    #define CHAR_f                      '\146'
1446    #define CHAR_g                      '\147'
1447    #define CHAR_h                      '\150'
1448    #define CHAR_i                      '\151'
1449    #define CHAR_j                      '\152'
1450    #define CHAR_k                      '\153'
1451    #define CHAR_l                      '\154'
1452    #define CHAR_m                      '\155'
1453    #define CHAR_n                      '\156'
1454    #define CHAR_o                      '\157'
1455    #define CHAR_p                      '\160'
1456    #define CHAR_q                      '\161'
1457    #define CHAR_r                      '\162'
1458    #define CHAR_s                      '\163'
1459    #define CHAR_t                      '\164'
1460    #define CHAR_u                      '\165'
1461    #define CHAR_v                      '\166'
1462    #define CHAR_w                      '\167'
1463    #define CHAR_x                      '\170'
1464    #define CHAR_y                      '\171'
1465    #define CHAR_z                      '\172'
1466    #define CHAR_LEFT_CURLY_BRACKET     '\173'
1467    #define CHAR_VERTICAL_LINE          '\174'
1468    #define CHAR_RIGHT_CURLY_BRACKET    '\175'
1469    #define CHAR_TILDE                  '\176'
1470    
1471    #define STR_HT                      "\011"
1472    #define STR_VT                      "\013"
1473    #define STR_FF                      "\014"
1474    #define STR_CR                      "\015"
1475    #define STR_NL                      "\012"
1476    #define STR_BS                      "\010"
1477    #define STR_BEL                     "\007"
1478    #define STR_ESC                     "\033"
1479    #define STR_DEL                     "\177"
1480    
1481    #define STR_SPACE                   "\040"
1482    #define STR_EXCLAMATION_MARK        "\041"
1483    #define STR_QUOTATION_MARK          "\042"
1484    #define STR_NUMBER_SIGN             "\043"
1485    #define STR_DOLLAR_SIGN             "\044"
1486    #define STR_PERCENT_SIGN            "\045"
1487    #define STR_AMPERSAND               "\046"
1488    #define STR_APOSTROPHE              "\047"
1489    #define STR_LEFT_PARENTHESIS        "\050"
1490    #define STR_RIGHT_PARENTHESIS       "\051"
1491    #define STR_ASTERISK                "\052"
1492    #define STR_PLUS                    "\053"
1493    #define STR_COMMA                   "\054"
1494    #define STR_MINUS                   "\055"
1495    #define STR_DOT                     "\056"
1496    #define STR_SLASH                   "\057"
1497    #define STR_0                       "\060"
1498    #define STR_1                       "\061"
1499    #define STR_2                       "\062"
1500    #define STR_3                       "\063"
1501    #define STR_4                       "\064"
1502    #define STR_5                       "\065"
1503    #define STR_6                       "\066"
1504    #define STR_7                       "\067"
1505    #define STR_8                       "\070"
1506    #define STR_9                       "\071"
1507    #define STR_COLON                   "\072"
1508    #define STR_SEMICOLON               "\073"
1509    #define STR_LESS_THAN_SIGN          "\074"
1510    #define STR_EQUALS_SIGN             "\075"
1511    #define STR_GREATER_THAN_SIGN       "\076"
1512    #define STR_QUESTION_MARK           "\077"
1513    #define STR_COMMERCIAL_AT           "\100"
1514    #define STR_A                       "\101"
1515    #define STR_B                       "\102"
1516    #define STR_C                       "\103"
1517    #define STR_D                       "\104"
1518    #define STR_E                       "\105"
1519    #define STR_F                       "\106"
1520    #define STR_G                       "\107"
1521    #define STR_H                       "\110"
1522    #define STR_I                       "\111"
1523    #define STR_J                       "\112"
1524    #define STR_K                       "\113"
1525    #define STR_L                       "\114"
1526    #define STR_M                       "\115"
1527    #define STR_N                       "\116"
1528    #define STR_O                       "\117"
1529    #define STR_P                       "\120"
1530    #define STR_Q                       "\121"
1531    #define STR_R                       "\122"
1532    #define STR_S                       "\123"
1533    #define STR_T                       "\124"
1534    #define STR_U                       "\125"
1535    #define STR_V                       "\126"
1536    #define STR_W                       "\127"
1537    #define STR_X                       "\130"
1538    #define STR_Y                       "\131"
1539    #define STR_Z                       "\132"
1540    #define STR_LEFT_SQUARE_BRACKET     "\133"
1541    #define STR_BACKSLASH               "\134"
1542    #define STR_RIGHT_SQUARE_BRACKET    "\135"
1543    #define STR_CIRCUMFLEX_ACCENT       "\136"
1544    #define STR_UNDERSCORE              "\137"
1545    #define STR_GRAVE_ACCENT            "\140"
1546    #define STR_a                       "\141"
1547    #define STR_b                       "\142"
1548    #define STR_c                       "\143"
1549    #define STR_d                       "\144"
1550    #define STR_e                       "\145"
1551    #define STR_f                       "\146"
1552    #define STR_g                       "\147"
1553    #define STR_h                       "\150"
1554    #define STR_i                       "\151"
1555    #define STR_j                       "\152"
1556    #define STR_k                       "\153"
1557    #define STR_l                       "\154"
1558    #define STR_m                       "\155"
1559    #define STR_n                       "\156"
1560    #define STR_o                       "\157"
1561    #define STR_p                       "\160"
1562    #define STR_q                       "\161"
1563    #define STR_r                       "\162"
1564    #define STR_s                       "\163"
1565    #define STR_t                       "\164"
1566    #define STR_u                       "\165"
1567    #define STR_v                       "\166"
1568    #define STR_w                       "\167"
1569    #define STR_x                       "\170"
1570    #define STR_y                       "\171"
1571    #define STR_z                       "\172"
1572    #define STR_LEFT_CURLY_BRACKET      "\173"
1573    #define STR_VERTICAL_LINE           "\174"
1574    #define STR_RIGHT_CURLY_BRACKET     "\175"
1575    #define STR_TILDE                   "\176"
1576    
1577    #define STRING_ACCEPT0              STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1578    #define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1579    #define STRING_F0                   STR_F "\0"
1580    #define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"
1581    #define STRING_MARK0                STR_M STR_A STR_R STR_K "\0"
1582    #define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"
1583    #define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"
1584    #define STRING_THEN                 STR_T STR_H STR_E STR_N
1585    
1586    #define STRING_alpha0               STR_a STR_l STR_p STR_h STR_a "\0"
1587    #define STRING_lower0               STR_l STR_o STR_w STR_e STR_r "\0"
1588    #define STRING_upper0               STR_u STR_p STR_p STR_e STR_r "\0"
1589    #define STRING_alnum0               STR_a STR_l STR_n STR_u STR_m "\0"
1590    #define STRING_ascii0               STR_a STR_s STR_c STR_i STR_i "\0"
1591    #define STRING_blank0               STR_b STR_l STR_a STR_n STR_k "\0"
1592    #define STRING_cntrl0               STR_c STR_n STR_t STR_r STR_l "\0"
1593    #define STRING_digit0               STR_d STR_i STR_g STR_i STR_t "\0"
1594    #define STRING_graph0               STR_g STR_r STR_a STR_p STR_h "\0"
1595    #define STRING_print0               STR_p STR_r STR_i STR_n STR_t "\0"
1596    #define STRING_punct0               STR_p STR_u STR_n STR_c STR_t "\0"
1597    #define STRING_space0               STR_s STR_p STR_a STR_c STR_e "\0"
1598    #define STRING_word0                STR_w STR_o STR_r STR_d       "\0"
1599    #define STRING_xdigit               STR_x STR_d STR_i STR_g STR_i STR_t
1600    
1601    #define STRING_DEFINE               STR_D STR_E STR_F STR_I STR_N STR_E
1602    
1603    #define STRING_CR_RIGHTPAR             STR_C STR_R STR_RIGHT_PARENTHESIS
1604    #define STRING_LF_RIGHTPAR             STR_L STR_F STR_RIGHT_PARENTHESIS
1605    #define STRING_CRLF_RIGHTPAR           STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1606    #define STRING_ANY_RIGHTPAR            STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1607    #define STRING_ANYCRLF_RIGHTPAR        STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1608    #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
1609    #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
1610    #ifdef COMPILE_PCRE8
1611    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1612    #endif
1613    #ifdef COMPILE_PCRE16
1614    #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
1615    #endif
1616    #define STRING_UCP_RIGHTPAR            STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1617    #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
1618    
1619    #endif  /* SUPPORT_UTF */
1620    
1621    /* Escape items that are just an encoding of a particular data value. */
1622    
1623  #ifndef ESC_e  #ifndef ESC_e
1624  #define ESC_e 27  #define ESC_e CHAR_ESC
1625  #endif  #endif
1626    
1627  #ifndef ESC_f  #ifndef ESC_f
1628  #define ESC_f '\f'  #define ESC_f CHAR_FF
1629  #endif  #endif
1630    
1631  #ifndef ESC_n  #ifndef ESC_n
1632  #define ESC_n NEWLINE  #define ESC_n CHAR_LF
1633  #endif  #endif
1634    
1635  #ifndef ESC_r  #ifndef ESC_r
1636  #define ESC_r '\r'  #define ESC_r CHAR_CR
1637  #endif  #endif
1638    
1639  /* 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
1640  (presumably because of all the others like size_t). */  (presumably because of all the others like size_t). */
1641    
1642  #ifndef ESC_tee  #ifndef ESC_tee
1643  #define ESC_tee '\t'  #define ESC_tee CHAR_HT
1644  #endif  #endif
1645    
1646  /* Codes for different types of Unicode property */  /* Codes for different types of Unicode property */
1647    
1648  #define PT_ANY        0    /* Any property - matches all chars */  #define PT_ANY        0    /* Any property - matches all chars */
1649  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */
1650  #define PT_GC         2    /* General characteristic (e.g. L) */  #define PT_GC         2    /* Specified general characteristic (e.g. L) */
1651  #define PT_PC         3    /* Particular characteristic (e.g. Lu) */  #define PT_PC         3    /* Specified particular characteristic (e.g. Lu) */
1652  #define PT_SC         4    /* Script (e.g. Han) */  #define PT_SC         4    /* Script (e.g. Han) */
1653    #define PT_ALNUM      5    /* Alphanumeric - the union of L and N */
1654    #define PT_SPACE      6    /* Perl space - Z plus 9,10,12,13 */
1655    #define PT_PXSPACE    7    /* POSIX space - Z plus 9,10,11,12,13 */
1656    #define PT_WORD       8    /* Word - L plus N plus underscore */
1657    
1658  /* 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
1659  contain UTF-8 characters with values greater than 255. */  contain characters with values greater than 255. */
1660    
1661  #define XCL_NOT    0x01    /* Flag: this is a negative class */  #define XCL_NOT    0x01    /* Flag: this is a negative class */
1662  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */
# Line 495  contain UTF-8 characters with values gre Line 1670  contain UTF-8 characters with values gre
1670  /* 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
1671  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
1672  their negation. Also, they must appear in the same order as in the opcode  their negation. Also, they must appear in the same order as in the opcode
1673  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
1674  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
1675  ESC_REF as subsequent values are used for \1, \2, \3, etc. There is are two  used for [^] in JavaScript compatibility mode, and for \C in non-utf mode. In
1676  tests in the code for an escape greater than ESC_b and less than ESC_Z to  non-DOTALL mode, "." behaves like \N.
1677  detect the types that may be repeated. These are the types that consume  
1678  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.
1679  character, that code will have to change. */  when PCRE_UCP is set, when replacement of \d etc by \p sequences is required.
1680    They must be contiguous, and remain in order so that the replacements can be
1681  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.
1682         ESC_w, ESC_dum1, ESC_C, ESC_P, ESC_p, ESC_X, ESC_Z, ESC_z, ESC_E,  
1683         ESC_Q, ESC_REF };  The final escape must be ESC_REF as subsequent values are used for
1684    backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
1685    greater than ESC_b and less than ESC_Z to detect the types that may be
1686    repeated. These are the types that consume characters. If any new escapes are
1687    put in between that don't consume a character, that code will have to change.
1688    */
1689    
1690  /* Opcode table: OP_BRA must be last, as all values >= it are used for brackets  enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1691  that extract substrings. Starting from 1 (i.e. after OP_END), the values up to         ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1692           ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
1693           ESC_E, ESC_Q, ESC_g, ESC_k,
1694           ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu,
1695           ESC_REF };
1696    
1697    /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
1698  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.
1699  Note that whenever this list is updated, the two macro definitions that follow  
1700  must also be updated to match. */  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1701    that follow must also be updated to match. There are also tables called
1702    "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1703    
1704  enum {  enum {
1705    OP_END,            /* 0 End of pattern */    OP_END,            /* 0 End of pattern */
# Line 520  enum { Line 1708  enum {
1708    
1709    OP_SOD,            /* 1 Start of data: \A */    OP_SOD,            /* 1 Start of data: \A */
1710    OP_SOM,            /* 2 Start of match (subject + offset): \G */    OP_SOM,            /* 2 Start of match (subject + offset): \G */
1711    OP_NOT_WORD_BOUNDARY,  /*  3 \B */    OP_SET_SOM,        /* 3 Set start of match (\K) */
1712    OP_WORD_BOUNDARY,      /*  4 \b */    OP_NOT_WORD_BOUNDARY,  /*  4 \B */
1713    OP_NOT_DIGIT,          /*  5 \D */    OP_WORD_BOUNDARY,      /*  5 \b */
1714    OP_DIGIT,              /*  6 \d */    OP_NOT_DIGIT,          /*  6 \D */
1715    OP_NOT_WHITESPACE,     /*  7 \S */    OP_DIGIT,              /*  7 \d */
1716    OP_WHITESPACE,         /*  8 \s */    OP_NOT_WHITESPACE,     /*  8 \S */
1717    OP_NOT_WORDCHAR,       /*  9 \W */    OP_WHITESPACE,         /*  9 \s */
1718    OP_WORDCHAR,           /* 10 \w */    OP_NOT_WORDCHAR,       /* 10 \W */
1719    OP_ANY,            /* 11 Match any character */    OP_WORDCHAR,           /* 11 \w */
1720    OP_ANYBYTE,        /* 12 Match any byte (\C); different to OP_ANY for UTF-8 */  
1721    OP_NOTPROP,        /* 13 \P (not Unicode property) */    OP_ANY,            /* 12 Match any character except newline */
1722    OP_PROP,           /* 14 \p (Unicode property) */    OP_ALLANY,         /* 13 Match any character */
1723    OP_EXTUNI,         /* 15 \X (extended Unicode sequence */    OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1724    OP_EODN,           /* 16 End of data or \n at end of data: \Z. */    OP_NOTPROP,        /* 15 \P (not Unicode property) */
1725    OP_EOD,            /* 17 End of data: \z */    OP_PROP,           /* 16 \p (Unicode property) */
1726      OP_ANYNL,          /* 17 \R (any newline sequence) */
1727    OP_OPT,            /* 18 Set runtime options */    OP_NOT_HSPACE,     /* 18 \H (not horizontal whitespace) */
1728    OP_CIRC,           /* 19 Start of line - varies with multiline switch */    OP_HSPACE,         /* 19 \h (horizontal whitespace) */
1729    OP_DOLL,           /* 20 End of line - varies with multiline switch */    OP_NOT_VSPACE,     /* 20 \V (not vertical whitespace) */
1730    OP_CHAR,           /* 21 Match one character, casefully */    OP_VSPACE,         /* 21 \v (vertical whitespace) */
1731    OP_CHARNC,         /* 22 Match one character, caselessly */    OP_EXTUNI,         /* 22 \X (extended Unicode sequence */
1732    OP_NOT,            /* 23 Match one character, not the following one */    OP_EODN,           /* 23 End of data or \n at end of data: \Z. */
1733      OP_EOD,            /* 24 End of data: \z */
1734    OP_STAR,           /* 24 The maximizing and minimizing versions of */  
1735    OP_MINSTAR,        /* 25 all these opcodes must come in pairs, with */    OP_CIRC,           /* 25 Start of line - not multiline */
1736    OP_PLUS,           /* 26 the minimizing one second. */    OP_CIRCM,          /* 26 Start of line - multiline */
1737    OP_MINPLUS,        /* 27 This first set applies to single characters */    OP_DOLL,           /* 27 End of line - not multiline */
1738    OP_QUERY,          /* 28 */    OP_DOLLM,          /* 28 End of line - multiline */
1739    OP_MINQUERY,       /* 29 */    OP_CHAR,           /* 29 Match one character, casefully */
1740    OP_UPTO,           /* 30 From 0 to n matches */    OP_CHARI,          /* 30 Match one character, caselessly */
1741    OP_MINUPTO,        /* 31 */    OP_NOT,            /* 31 Match one character, not the given one, casefully */
1742    OP_EXACT,          /* 32 Exactly n matches */    OP_NOTI,           /* 32 Match one character, not the given one, caselessly */
1743    
1744    OP_NOTSTAR,        /* 33 The maximizing and minimizing versions of */    /* The following sets of 13 opcodes must always be kept in step because
1745    OP_NOTMINSTAR,     /* 34 all these opcodes must come in pairs, with */    the offset from the first one is used to generate the others. */
1746    OP_NOTPLUS,        /* 35 the minimizing one second. */  
1747    OP_NOTMINPLUS,     /* 36 This set applies to "not" single characters */    /**** Single characters, caseful, must precede the caseless ones ****/
1748    OP_NOTQUERY,       /* 37 */  
1749    OP_NOTMINQUERY,    /* 38 */    OP_STAR,           /* 33 The maximizing and minimizing versions of */
1750    OP_NOTUPTO,        /* 39 From 0 to n matches */    OP_MINSTAR,        /* 34 these six opcodes must come in pairs, with */
1751    OP_NOTMINUPTO,     /* 40 */    OP_PLUS,           /* 35 the minimizing one second. */
1752    OP_NOTEXACT,       /* 41 Exactly n matches */    OP_MINPLUS,        /* 36 */
1753      OP_QUERY,          /* 37 */
1754    OP_TYPESTAR,       /* 42 The maximizing and minimizing versions of */    OP_MINQUERY,       /* 38 */
1755    OP_TYPEMINSTAR,    /* 43 all these opcodes must come in pairs, with */  
1756    OP_TYPEPLUS,       /* 44 the minimizing one second. These codes must */    OP_UPTO,           /* 39 From 0 to n matches of one character, caseful*/
1757    OP_TYPEMINPLUS,    /* 45 be in exactly the same order as those above. */    OP_MINUPTO,        /* 40 */
1758    OP_TYPEQUERY,      /* 46 This set applies to character types such as \d */    OP_EXACT,          /* 41 Exactly n matches */
1759    OP_TYPEMINQUERY,   /* 47 */  
1760    OP_TYPEUPTO,       /* 48 From 0 to n matches */    OP_POSSTAR,        /* 42 Possessified star, caseful */
1761    OP_TYPEMINUPTO,    /* 49 */    OP_POSPLUS,        /* 43 Possessified plus, caseful */
1762    OP_TYPEEXACT,      /* 50 Exactly n matches */    OP_POSQUERY,       /* 44 Posesssified query, caseful */
1763      OP_POSUPTO,        /* 45 Possessified upto, caseful */
1764    OP_CRSTAR,         /* 51 The maximizing and minimizing versions of */  
1765    OP_CRMINSTAR,      /* 52 all these opcodes must come in pairs, with */    /**** Single characters, caseless, must follow the caseful ones */
1766    OP_CRPLUS,         /* 53 the minimizing one second. These codes must */  
1767    OP_CRMINPLUS,      /* 54 be in exactly the same order as those above. */    OP_STARI,          /* 46 */
1768    OP_CRQUERY,        /* 55 These are for character classes and back refs */    OP_MINSTARI,       /* 47 */
1769    OP_CRMINQUERY,     /* 56 */    OP_PLUSI,          /* 48 */
1770    OP_CRRANGE,        /* 57 These are different to the three sets above. */    OP_MINPLUSI,       /* 49 */
1771    OP_CRMINRANGE,     /* 58 */    OP_QUERYI,         /* 50 */
1772      OP_MINQUERYI,      /* 51 */
1773    OP_CLASS,          /* 59 Match a character class, chars < 256 only */  
1774    OP_NCLASS,         /* 60 Same, but the bitmap was created from a negative    OP_UPTOI,          /* 52 From 0 to n matches of one character, caseless */
1775                             class - the difference is relevant only when a UTF-8    OP_MINUPTOI,       /* 53 */
1776                             character > 255 is encountered. */    OP_EXACTI,         /* 54 */
1777    
1778    OP_XCLASS,         /* 61 Extended class for handling UTF-8 chars within the    OP_POSSTARI,       /* 55 Possessified star, caseless */
1779                             class. This does both positive and negative. */    OP_POSPLUSI,       /* 56 Possessified plus, caseless */
1780      OP_POSQUERYI,      /* 57 Posesssified query, caseless */
1781    OP_REF,            /* 62 Match a back reference */    OP_POSUPTOI,       /* 58 Possessified upto, caseless */
1782    OP_RECURSE,        /* 63 Match a numbered subpattern (possibly recursive) */  
1783    OP_CALLOUT,        /* 64 Call out to external function if provided */    /**** The negated ones must follow the non-negated ones, and match them ****/
1784      /**** Negated single character, caseful; must precede the caseless ones ****/
1785    OP_ALT,            /* 65 Start of alternation */  
1786    OP_KET,            /* 66 End of group that doesn't have an unbounded repeat */    OP_NOTSTAR,        /* 59 The maximizing and minimizing versions of */
1787    OP_KETRMAX,        /* 67 These two must remain together and in this */    OP_NOTMINSTAR,     /* 60 these six opcodes must come in pairs, with */
1788    OP_KETRMIN,        /* 68 order. They are for groups the repeat for ever. */    OP_NOTPLUS,        /* 61 the minimizing one second. They must be in */
1789      OP_NOTMINPLUS,     /* 62 exactly the same order as those above. */
1790    /* The assertions must come before ONCE and COND */    OP_NOTQUERY,       /* 63 */
1791      OP_NOTMINQUERY,    /* 64 */
1792    OP_ASSERT,         /* 69 Positive lookahead */  
1793    OP_ASSERT_NOT,     /* 70 Negative lookahead */    OP_NOTUPTO,        /* 65 From 0 to n matches, caseful */
1794    OP_ASSERTBACK,     /* 71 Positive lookbehind */    OP_NOTMINUPTO,     /* 66 */
1795    OP_ASSERTBACK_NOT, /* 72 Negative lookbehind */    OP_NOTEXACT,       /* 67 Exactly n matches */
1796    OP_REVERSE,        /* 73 Move pointer back - used in lookbehind assertions */  
1797      OP_NOTPOSSTAR,     /* 68 Possessified versions, caseful */
1798    /* ONCE and COND must come after the assertions, with ONCE first, as there's    OP_NOTPOSPLUS,     /* 69 */
1799    a test for >= ONCE for a subpattern that isn't an assertion. */    OP_NOTPOSQUERY,    /* 70 */
1800      OP_NOTPOSUPTO,     /* 71 */
1801    OP_ONCE,           /* 74 Once matched, don't back up into the subpattern */  
1802    OP_COND,           /* 75 Conditional group */    /**** Negated single character, caseless; must follow the caseful ones ****/
1803    OP_CREF,           /* 76 Used to hold an extraction string number (cond ref) */  
1804      OP_NOTSTARI,       /* 72 */
1805    OP_BRAZERO,        /* 77 These two must remain together and in this */    OP_NOTMINSTARI,    /* 73 */
1806    OP_BRAMINZERO,     /* 78 order. */    OP_NOTPLUSI,       /* 74 */
1807      OP_NOTMINPLUSI,    /* 75 */
1808    OP_BRANUMBER,      /* 79 Used for extracting brackets whose number is greater    OP_NOTQUERYI,      /* 76 */
1809                             than can fit into an opcode. */    OP_NOTMINQUERYI,   /* 77 */
1810    
1811    OP_BRA             /* 80 This and greater values are used for brackets that    OP_NOTUPTOI,       /* 78 From 0 to n matches, caseless */
1812                             extract substrings up to EXTRACT_BASIC_MAX. After    OP_NOTMINUPTOI,    /* 79 */
1813                             that, use is made of OP_BRANUMBER. */    OP_NOTEXACTI,      /* 80 Exactly n matches */
1814  };  
1815      OP_NOTPOSSTARI,    /* 81 Possessified versions, caseless */
1816      OP_NOTPOSPLUSI,    /* 82 */
1817      OP_NOTPOSQUERYI,   /* 83 */
1818      OP_NOTPOSUPTOI,    /* 84 */
1819    
1820      /**** Character types ****/
1821    
1822      OP_TYPESTAR,       /* 85 The maximizing and minimizing versions of */
1823      OP_TYPEMINSTAR,    /* 86 these six opcodes must come in pairs, with */
1824      OP_TYPEPLUS,       /* 87 the minimizing one second. These codes must */
1825      OP_TYPEMINPLUS,    /* 88 be in exactly the same order as those above. */
1826      OP_TYPEQUERY,      /* 89 */
1827      OP_TYPEMINQUERY,   /* 90 */
1828    
1829      OP_TYPEUPTO,       /* 91 From 0 to n matches */
1830      OP_TYPEMINUPTO,    /* 92 */
1831      OP_TYPEEXACT,      /* 93 Exactly n matches */
1832    
1833      OP_TYPEPOSSTAR,    /* 94 Possessified versions */
1834      OP_TYPEPOSPLUS,    /* 95 */
1835      OP_TYPEPOSQUERY,   /* 96 */
1836      OP_TYPEPOSUPTO,    /* 97 */
1837    
1838      /* These are used for character classes and back references; only the
1839      first six are the same as the sets above. */
1840    
1841      OP_CRSTAR,         /* 98 The maximizing and minimizing versions of */
1842      OP_CRMINSTAR,      /* 99 all these opcodes must come in pairs, with */
1843      OP_CRPLUS,         /* 100 the minimizing one second. These codes must */
1844      OP_CRMINPLUS,      /* 101 be in exactly the same order as those above. */
1845      OP_CRQUERY,        /* 102 */
1846      OP_CRMINQUERY,     /* 103 */
1847    
1848      OP_CRRANGE,        /* 104 These are different to the three sets above. */
1849      OP_CRMINRANGE,     /* 105 */
1850    
1851      /* End of quantifier opcodes */
1852    
1853      OP_CLASS,          /* 106 Match a character class, chars < 256 only */
1854      OP_NCLASS,         /* 107 Same, but the bitmap was created from a negative
1855                                  class - the difference is relevant only when a
1856                                  character > 255 is encountered. */
1857      OP_XCLASS,         /* 108 Extended class for handling > 255 chars within the
1858                                  class. This does both positive and negative. */
1859      OP_REF,            /* 109 Match a back reference, casefully */
1860      OP_REFI,           /* 110 Match a back reference, caselessly */
1861      OP_RECURSE,        /* 111 Match a numbered subpattern (possibly recursive) */
1862      OP_CALLOUT,        /* 112 Call out to external function if provided */
1863    
1864      OP_ALT,            /* 113 Start of alternation */
1865      OP_KET,            /* 114 End of group that doesn't have an unbounded repeat */
1866      OP_KETRMAX,        /* 115 These two must remain together and in this */
1867      OP_KETRMIN,        /* 116 order. They are for groups the repeat for ever. */
1868      OP_KETRPOS,        /* 117 Possessive unlimited repeat. */
1869    
1870      /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
1871      asserts must remain in order. */
1872    
1873      OP_REVERSE,        /* 118 Move pointer back - used in lookbehind assertions */
1874      OP_ASSERT,         /* 119 Positive lookahead */
1875      OP_ASSERT_NOT,     /* 120 Negative lookahead */
1876      OP_ASSERTBACK,     /* 121 Positive lookbehind */
1877      OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
1878    
1879      /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
1880      after the assertions, with ONCE first, as there's a test for >= ONCE for a
1881      subpattern that isn't an assertion. The POS versions must immediately follow
1882      the non-POS versions in each case. */
1883    
1884      OP_ONCE,           /* 123 Atomic group, contains captures */
1885      OP_ONCE_NC,        /* 124 Atomic group containing no captures */
1886      OP_BRA,            /* 125 Start of non-capturing bracket */
1887      OP_BRAPOS,         /* 126 Ditto, with unlimited, possessive repeat */
1888      OP_CBRA,           /* 127 Start of capturing bracket */
1889      OP_CBRAPOS,        /* 128 Ditto, with unlimited, possessive repeat */
1890      OP_COND,           /* 129 Conditional group */
1891    
1892      /* These five must follow the previous five, in the same order. There's a
1893      check for >= SBRA to distinguish the two sets. */
1894    
1895      OP_SBRA,           /* 130 Start of non-capturing bracket, check empty  */
1896      OP_SBRAPOS,        /* 131 Ditto, with unlimited, possessive repeat */
1897      OP_SCBRA,          /* 132 Start of capturing bracket, check empty */
1898      OP_SCBRAPOS,       /* 133 Ditto, with unlimited, possessive repeat */
1899      OP_SCOND,          /* 134 Conditional group, check empty */
1900    
1901      /* The next two pairs must (respectively) be kept together. */
1902    
1903      OP_CREF,           /* 135 Used to hold a capture number as condition */
1904      OP_NCREF,          /* 136 Same, but generated by a name reference*/
1905      OP_RREF,           /* 137 Used to hold a recursion number as condition */
1906      OP_NRREF,          /* 138 Same, but generated by a name reference*/
1907      OP_DEF,            /* 139 The DEFINE condition */
1908    
1909      OP_BRAZERO,        /* 140 These two must remain together and in this */
1910      OP_BRAMINZERO,     /* 141 order. */
1911      OP_BRAPOSZERO,     /* 142 */
1912    
1913      /* These are backtracking control verbs */
1914    
1915      OP_MARK,           /* 143 always has an argument */
1916      OP_PRUNE,          /* 144 */
1917      OP_PRUNE_ARG,      /* 145 same, but with argument */
1918      OP_SKIP,           /* 146 */
1919      OP_SKIP_ARG,       /* 147 same, but with argument */
1920      OP_THEN,           /* 148 */
1921      OP_THEN_ARG,       /* 149 same, but with argument */
1922      OP_COMMIT,         /* 150 */
1923    
1924      /* These are forced failure and success verbs */
1925    
1926      OP_FAIL,           /* 151 */
1927      OP_ACCEPT,         /* 152 */
1928      OP_ASSERT_ACCEPT,  /* 153 Used inside assertions */
1929      OP_CLOSE,          /* 154 Used before OP_ACCEPT to close open captures */
1930    
1931      /* This is used to skip a subpattern with a {0} quantifier */
1932    
1933      OP_SKIPZERO,       /* 155 */
1934    
1935      /* This is not an opcode, but is used to check that tables indexed by opcode
1936      are the correct length, in order to catch updating errors - there have been
1937      some in the past. */
1938    
1939  /* WARNING WARNING WARNING: There is an implicit assumption in pcre.c and    OP_TABLE_LENGTH
1940  study.c that all opcodes are less than 128 in value. This makes handling UTF-8  };
 character sequences easier. */  
   
 /* The highest extraction number before we have to start using additional  
 bytes. (Originally PCRE didn't have support for extraction counts highter than  
 this number.) The value is limited by the number of opcodes left after OP_BRA,  
 i.e. 255 - OP_BRA. We actually set it a bit lower to leave room for additional  
 opcodes. */  
1941    
1942  #define EXTRACT_BASIC_MAX  100  /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
1943    definitions that follow must also be updated to match. There are also tables
1944    called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1945    
1946    
1947  /* 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
1948  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
1949    only in pcre_printint.c, which fills out the full names in many cases (and in
1950    some cases doesn't actually use these names at all). */
1951    
1952  #define OP_NAME_LIST \  #define OP_NAME_LIST \
1953    "End", "\\A", "\\G", "\\B", "\\b", "\\D", "\\d",                \    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \
1954    "\\S", "\\s", "\\W", "\\w", "Any", "Anybyte",                   \    "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte",         \
1955    "notprop", "prop", "extuni",                                    \    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \
1956    "\\Z", "\\z",                                                   \    "extuni",  "\\Z", "\\z",                                        \
1957    "Opt", "^", "$", "char", "charnc", "not",                       \    "^", "^", "$", "$", "char", "chari", "not", "noti",             \
1958    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
1959    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "{", "{", "{",                                                  \
1960      "*+","++", "?+", "{",                                           \
1961      "*", "*?", "+", "+?", "?", "??",                                \
1962      "{", "{", "{",                                                  \
1963      "*+","++", "?+", "{",                                           \
1964      "*", "*?", "+", "+?", "?", "??",                                \
1965      "{", "{", "{",                                                  \
1966      "*+","++", "?+", "{",                                           \
1967      "*", "*?", "+", "+?", "?", "??",                                \
1968      "{", "{", "{",                                                  \
1969      "*+","++", "?+", "{",                                           \
1970    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \
1971      "*+","++", "?+", "{",                                           \
1972    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \
1973    "class", "nclass", "xclass", "Ref", "Recurse", "Callout",       \    "class", "nclass", "xclass", "Ref", "Refi",                     \
1974    "Alt", "Ket", "KetRmax", "KetRmin", "Assert", "Assert not",     \    "Recurse", "Callout",                                           \
1975    "AssertB", "AssertB not", "Reverse", "Once", "Cond", "Cond ref",\    "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos",                  \
1976    "Brazero", "Braminzero", "Branumber", "Bra"    "Reverse", "Assert", "Assert not", "AssertB", "AssertB not",    \
1977      "Once", "Once_NC",                                              \
1978      "Bra", "BraPos", "CBra", "CBraPos",                             \
1979      "Cond",                                                         \
1980      "SBra", "SBraPos", "SCBra", "SCBraPos",                         \
1981      "SCond",                                                        \
1982      "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def",   \
1983      "Brazero", "Braminzero", "Braposzero",                          \
1984      "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP",                  \
1985      "*THEN", "*THEN", "*COMMIT", "*FAIL",                           \
1986      "*ACCEPT", "*ASSERT_ACCEPT",                                    \
1987      "Close", "Skip zero"
1988    
1989    
1990  /* 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 668  in UTF-8 mode. The code that uses this t Line 1998  in UTF-8 mode. The code that uses this t
1998    
1999  #define OP_LENGTHS \  #define OP_LENGTHS \
2000    1,                             /* End                                    */ \    1,                             /* End                                    */ \
2001    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                     */ \
2002    1, 1,                          /* Any, Anybyte                           */ \    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \
2003    3, 3, 1,                       /* NOTPROP, PROP, EXTUNI                  */ \    1, 1, 1,                       /* Any, AllAny, Anybyte                   */ \
2004    1, 1, 2, 1, 1,                 /* \Z, \z, Opt, ^, $                      */ \    3, 3,                          /* \P, \p                                 */ \
2005      1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \
2006      1,                             /* \X                                     */ \
2007      1, 1, 1, 1, 1, 1,              /* \Z, \z, ^, ^M, $, $M                   */ \
2008    2,                             /* Char  - the minimum length             */ \    2,                             /* Char  - the minimum length             */ \
2009    2,                             /* Charnc  - the minimum length           */ \    2,                             /* Chari  - the minimum length            */ \
2010    2,                             /* not                                    */ \    2,                             /* not                                    */ \
2011    /* Positive single-char repeats                            ** These are  */ \    2,                             /* noti                                   */ \
2012    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??      ** minima in  */ \    /* Positive single-char repeats                             ** These are */ \
2013    4, 4, 4,                       /* upto, minupto, exact     ** UTF-8 mode */ \    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??       ** minima in */ \
2014      2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto, minupto             ** mode      */ \
2015      2+IMM2_SIZE,                   /* exact                                  */ \
2016      2, 2, 2, 2+IMM2_SIZE,          /* *+, ++, ?+, upto+                      */ \
2017      2, 2, 2, 2, 2, 2,              /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8     */ \
2018      2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto I, minupto I                      */ \
2019      2+IMM2_SIZE,                   /* exact I                                */ \
2020      2, 2, 2, 2+IMM2_SIZE,          /* *+I, ++I, ?+I, upto+I                  */ \
2021    /* Negative single-char repeats - only for chars < 256                   */ \    /* Negative single-char repeats - only for chars < 256                   */ \
2022    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \
2023    4, 4, 4,                       /* NOT upto, minupto, exact               */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto, minupto                      */ \
2024      2+IMM2_SIZE,                   /* NOT exact                              */ \
2025      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *, +, ?, upto           */ \
2026      2, 2, 2, 2, 2, 2,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */ \
2027      2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto I, minupto I                  */ \
2028      2+IMM2_SIZE,                   /* NOT exact I                            */ \
2029      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *I, +I, ?I, upto I      */ \
2030    /* Positive type repeats                                                 */ \    /* Positive type repeats                                                 */ \
2031    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \
2032    4, 4, 4,                       /* Type upto, minupto, exact              */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* Type upto, minupto                     */ \
2033      2+IMM2_SIZE,                   /* Type exact                             */ \
2034      2, 2, 2, 2+IMM2_SIZE,          /* Possessive *+, ++, ?+, upto+           */ \
2035    /* Character class & ref repeats                                         */ \    /* Character class & ref repeats                                         */ \
2036    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \
2037    5, 5,                          /* CRRANGE, CRMINRANGE                    */ \    1+2*IMM2_SIZE, 1+2*IMM2_SIZE,  /* CRRANGE, CRMINRANGE                    */ \
2038   33,                             /* CLASS                                  */ \    1+(32/sizeof(pcre_uchar)),     /* CLASS                                  */ \
2039   33,                             /* NCLASS                                 */ \    1+(32/sizeof(pcre_uchar)),     /* NCLASS                                 */ \
2040    0,                             /* XCLASS - variable length               */ \    0,                             /* XCLASS - variable length               */ \
2041    3,                             /* REF                                    */ \    1+IMM2_SIZE,                   /* REF                                    */ \
2042      1+IMM2_SIZE,                   /* REFI                                   */ \
2043    1+LINK_SIZE,                   /* RECURSE                                */ \    1+LINK_SIZE,                   /* RECURSE                                */ \
2044    2+2*LINK_SIZE,                 /* CALLOUT                                */ \    2+2*LINK_SIZE,                 /* CALLOUT                                */ \
2045    1+LINK_SIZE,                   /* Alt                                    */ \    1+LINK_SIZE,                   /* Alt                                    */ \
2046    1+LINK_SIZE,                   /* Ket                                    */ \    1+LINK_SIZE,                   /* Ket                                    */ \
2047    1+LINK_SIZE,                   /* KetRmax                                */ \    1+LINK_SIZE,                   /* KetRmax                                */ \
2048    1+LINK_SIZE,                   /* KetRmin                                */ \    1+LINK_SIZE,                   /* KetRmin                                */ \
2049      1+LINK_SIZE,                   /* KetRpos                                */ \
2050      1+LINK_SIZE,                   /* Reverse                                */ \
2051    1+LINK_SIZE,                   /* Assert                                 */ \    1+LINK_SIZE,                   /* Assert                                 */ \
2052    1+LINK_SIZE,                   /* Assert not                             */ \    1+LINK_SIZE,                   /* Assert not                             */ \
2053    1+LINK_SIZE,                   /* Assert behind                          */ \    1+LINK_SIZE,                   /* Assert behind                          */ \
2054    1+LINK_SIZE,                   /* Assert behind not                      */ \    1+LINK_SIZE,                   /* Assert behind not                      */ \
2055    1+LINK_SIZE,                   /* Reverse                                */ \    1+LINK_SIZE,                   /* ONCE                                   */ \
2056    1+LINK_SIZE,                   /* Once                                   */ \    1+LINK_SIZE,                   /* ONCE_NC                                */ \
2057      1+LINK_SIZE,                   /* BRA                                    */ \
2058      1+LINK_SIZE,                   /* BRAPOS                                 */ \
2059      1+LINK_SIZE+IMM2_SIZE,         /* CBRA                                   */ \
2060      1+LINK_SIZE+IMM2_SIZE,         /* CBRAPOS                                */ \
2061    1+LINK_SIZE,                   /* COND                                   */ \    1+LINK_SIZE,                   /* COND                                   */ \
2062    3,                             /* CREF                                   */ \    1+LINK_SIZE,                   /* SBRA                                   */ \
2063    1, 1,                          /* BRAZERO, BRAMINZERO                    */ \    1+LINK_SIZE,                   /* SBRAPOS                                */ \
2064    3,                             /* BRANUMBER                              */ \    1+LINK_SIZE+IMM2_SIZE,         /* SCBRA                                  */ \
2065    1+LINK_SIZE                    /* BRA                                    */ \    1+LINK_SIZE+IMM2_SIZE,         /* SCBRAPOS                               */ \
2066      1+LINK_SIZE,                   /* SCOND                                  */ \
2067      1+IMM2_SIZE, 1+IMM2_SIZE,      /* CREF, NCREF                            */ \
2068  /* A magic value for OP_CREF to indicate the "in recursion" condition. */    1+IMM2_SIZE, 1+IMM2_SIZE,      /* RREF, NRREF                            */ \
2069      1,                             /* DEF                                    */ \
2070  #define CREF_RECURSE  0xffff    1, 1, 1,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */ \
2071      3, 1, 3,                       /* MARK, PRUNE, PRUNE_ARG                 */ \
2072  /* Error code numbers. They are given names so that they can more easily be    1, 3,                          /* SKIP, SKIP_ARG                         */ \
2073  tracked. */    1, 3,                          /* THEN, THEN_ARG                         */ \
2074      1, 1, 1, 1,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */ \
2075      1+IMM2_SIZE, 1                 /* CLOSE, SKIPZERO                        */
2076    
2077    /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
2078    condition. */
2079    
2080    #define RREF_ANY  0xffff
2081    
2082    /* Compile time error code numbers. They are given names so that they can more
2083    easily be tracked. When a new number is added, the table called eint in
2084    pcreposix.c must be updated. */
2085    
2086  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
2087         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
2088         ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,         ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
2089         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
2090         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
2091         ERR50, ERR51 };         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
2092           ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
2093           ERR70, ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERRCOUNT };
2094    
2095    /* JIT compiling modes. The function list is indexed by them. */
2096    enum { JIT_COMPILE, JIT_PARTIAL_SOFT_COMPILE, JIT_PARTIAL_HARD_COMPILE,
2097           JIT_NUMBER_OF_COMPILE_MODES };
2098    
2099  /* 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
2100  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 738  Because people can now save and re-use c Line 2110  Because people can now save and re-use c
2110  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
2111  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
2112  fields are present. Currently PCRE always sets the dummy fields to zero.  fields are present. Currently PCRE always sets the dummy fields to zero.
2113  NOTE NOTE NOTE:  NOTE NOTE NOTE
2114  */  */
2115    
2116  typedef struct real_pcre {  #ifdef COMPILE_PCRE8
2117    #define REAL_PCRE real_pcre
2118    #else
2119    #define REAL_PCRE real_pcre16
2120    #endif
2121    
2122    typedef struct REAL_PCRE {
2123    pcre_uint32 magic_number;    pcre_uint32 magic_number;
2124    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2125    pcre_uint32 options;    pcre_uint32 options;            /* Public options */
2126    pcre_uint32 dummy1;             /* For future use, maybe */    pcre_uint16 flags;              /* Private flags */
2127      pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2128    pcre_uint16 top_bracket;    pcre_uint16 top_bracket;        /* Highest numbered group */
2129    pcre_uint16 top_backref;    pcre_uint16 top_backref;        /* Highest numbered back reference */
2130    pcre_uint16 first_byte;    pcre_uint16 first_char;         /* Starting character */
2131    pcre_uint16 req_byte;    pcre_uint16 req_char;           /* This character must be seen */
2132    pcre_uint16 name_table_offset;  /* Offset to name table that follows */    pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2133    pcre_uint16 name_entry_size;    /* Size of any name items */    pcre_uint16 name_entry_size;    /* Size of any name items */
2134    pcre_uint16 name_count;         /* Number of name items */    pcre_uint16 name_count;         /* Number of name items */
2135    pcre_uint16 ref_count;          /* Reference count */    pcre_uint16 ref_count;          /* Reference count */
2136      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2137    const unsigned char *tables;    /* Pointer to tables or NULL for std */    const pcre_uint8 *nullpad;      /* NULL padding */
2138    const unsigned char *nullpad;   /* NULL padding */  } REAL_PCRE;
 } real_pcre;  
2139    
2140  /* 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
2141  remark (see NOTE above) about extending this structure applies. */  remark (see NOTE above) about extending this structure applies. */
2142    
2143  typedef struct pcre_study_data {  typedef struct pcre_study_data {
2144    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2145    pcre_uint32 options;    pcre_uint32 flags;              /* Private flags */
2146    uschar start_bits[32];    pcre_uint8 start_bits[32];      /* Starting char bits */
2147      pcre_uint32 minlength;          /* Minimum subject length */
2148  } pcre_study_data;  } pcre_study_data;
2149    
2150    /* Structure for building a chain of open capturing subpatterns during
2151    compiling, so that instructions to close them can be compiled when (*ACCEPT) is
2152    encountered. This is also used to identify subpatterns that contain recursive
2153    back references to themselves, so that they can be made atomic. */
2154    
2155    typedef struct open_capitem {
2156      struct open_capitem *next;    /* Chain link */
2157      pcre_uint16 number;           /* Capture number */
2158      pcre_uint16 flag;             /* Set TRUE if recursive back ref */
2159    } open_capitem;
2160    
2161  /* Structure for passing "static" information around between the functions  /* Structure for passing "static" information around between the functions
2162  doing the compiling, so that they are thread-safe. */  doing the compiling, so that they are thread-safe. */
2163    
2164  typedef struct compile_data {  typedef struct compile_data {
2165    const uschar *lcc;            /* Points to lower casing table */    const pcre_uint8 *lcc;            /* Points to lower casing table */
2166    const uschar *fcc;            /* Points to case-flipping table */    const pcre_uint8 *fcc;            /* Points to case-flipping table */
2167    const uschar *cbits;          /* Points to character type table */    const pcre_uint8 *cbits;          /* Points to character type table */
2168    const uschar *ctypes;         /* Points to table of type maps */    const pcre_uint8 *ctypes;         /* Points to table of type maps */
2169    const uschar *start_code;     /* The start of the compiled code */    const pcre_uchar *start_workspace;/* The start of working space */
2170    const uschar *start_pattern;  /* The start of the pattern */    const pcre_uchar *start_code;     /* The start of the compiled code */
2171    uschar *name_table;           /* The name/number table */    const pcre_uchar *start_pattern;  /* The start of the pattern */
2172    int  names_found;             /* Number of entries so far */    const pcre_uchar *end_pattern;    /* The end of the pattern */
2173    int  name_entry_size;         /* Size of each entry */    open_capitem *open_caps;          /* Chain of open capture items */
2174    int  top_backref;             /* Maximum back reference */    pcre_uchar *hwm;                  /* High watermark of workspace */
2175    unsigned int backref_map;     /* Bitmap of low back refs */    pcre_uchar *name_table;           /* The name/number table */
2176    int  req_varyopt;             /* "After variable item" flag for reqbyte */    int  names_found;                 /* Number of entries so far */
2177    BOOL nopartial;               /* Set TRUE if partial won't work */    int  name_entry_size;             /* Size of each entry */
2178    int  nllen;                   /* 1 or 2 for newline string length */    int  workspace_size;              /* Size of workspace */
2179    uschar nl[4];                 /* Newline string */    int  bracount;                    /* Count of capturing parens as we compile */
2180      int  final_bracount;              /* Saved value after first pass */
2181      int  max_lookbehind;              /* Maximum lookbehind (characters) */
2182      int  top_backref;                 /* Maximum back reference */
2183      unsigned int backref_map;         /* Bitmap of low back refs */
2184      int  assert_depth;                /* Depth of nested assertions */
2185      int  external_options;            /* External (initial) options */
2186      int  external_flags;              /* External flag bits to be set */
2187      int  req_varyopt;                 /* "After variable item" flag for reqbyte */
2188      BOOL had_accept;                  /* (*ACCEPT) encountered */
2189      BOOL had_pruneorskip;             /* (*PRUNE) or (*SKIP) encountered */
2190      BOOL check_lookbehind;            /* Lookbehinds need later checking */
2191      int  nltype;                      /* Newline type */
2192      int  nllen;                       /* Newline string length */
2193      pcre_uchar nl[4];                 /* Newline string when fixed length */
2194  } compile_data;  } compile_data;
2195    
2196  /* Structure for maintaining a chain of pointers to the currently incomplete  /* Structure for maintaining a chain of pointers to the currently incomplete
2197  branches, for testing for left recursion. */  branches, for testing for left recursion while compiling. */
2198    
2199  typedef struct branch_chain {  typedef struct branch_chain {
2200    struct branch_chain *outer;    struct branch_chain *outer;
2201    uschar *current;    pcre_uchar *current_branch;
2202  } branch_chain;  } branch_chain;
2203    
2204  /* Structure for items in a linked list that represents an explicit recursive  /* Structure for items in a linked list that represents an explicit recursive
2205  call within the pattern. */  call within the pattern; used by pcre_exec(). */
2206    
2207  typedef struct recursion_info {  typedef struct recursion_info {
2208    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
2209    int group_num;                /* Number of group that was called */    int group_num;                  /* Number of group that was called */
2210    const uschar *after_call;     /* "Return value": points after the call in the expr */    int *offset_save;               /* Pointer to start of saved offsets */
2211    USPTR save_start;             /* Old value of md->start_match */    int saved_max;                  /* Number of saved offsets */
2212    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 */  
2213  } recursion_info;  } recursion_info;
2214    
2215  /* When compiling in a mode that doesn't use recursive calls to match(),  /* A similar structure for pcre_dfa_exec(). */
2216  a structure is used to remember local variables on the heap. It is defined in  
2217  pcre_exec.c, close to the match() function, so that it is easy to keep it in  typedef struct dfa_recursion_info {
2218  step with any changes of local variable. However, the pointer to the current    struct dfa_recursion_info *prevrec;
2219  frame must be saved in some "static" place over a longjmp(). We declare the    int group_num;
2220  structure here so that we can put a pointer in the match_data structure. NOTE:    PCRE_PUCHAR subject_position;
2221  This isn't used for a "normal" compilation of pcre. */  } dfa_recursion_info;
2222    
2223    /* Structure for building a chain of data for holding the values of the subject
2224    pointer at the start of each subpattern, so as to detect when an empty string
2225    has been matched by a subpattern - to break infinite loops; used by
2226    pcre_exec(). */
2227    
2228    typedef struct eptrblock {
2229      struct eptrblock *epb_prev;
2230      PCRE_PUCHAR epb_saved_eptr;
2231    } eptrblock;
2232    
 struct heapframe;  
2233    
2234  /* Structure for passing "static" information around between the functions  /* Structure for passing "static" information around between the functions
2235  doing traditional NFA matching, so that they are thread-safe. */  doing traditional NFA matching, so that they are thread-safe. */
# Line 827  typedef struct match_data { Line 2238  typedef struct match_data {
2238    unsigned long int match_call_count;      /* As it says */    unsigned long int match_call_count;      /* As it says */
2239    unsigned long int match_limit;           /* As it says */    unsigned long int match_limit;           /* As it says */
2240    unsigned long int match_limit_recursion; /* As it says */    unsigned long int match_limit_recursion; /* As it says */
2241    int   *offset_vector;         /* Offset vector */    int   *offset_vector;           /* Offset vector */
2242    int    offset_end;            /* One past the end */    int    offset_end;              /* One past the end */
2243    int    offset_max;            /* The maximum usable for return data */    int    offset_max;              /* The maximum usable for return data */
2244    int    nllen;                 /* 1 or 2 for newline string length */    int    nltype;                  /* Newline type */
2245    uschar nl[4];                 /* Newline string */    int    nllen;                   /* Newline string length */
2246    const uschar *lcc;            /* Points to lower casing table */    int    name_count;              /* Number of names in name table */
2247    const uschar *ctypes;         /* Points to table of type maps */    int    name_entry_size;         /* Size of entry in names table */
2248    BOOL   offset_overflow;       /* Set if too many extractions */    pcre_uchar *name_table;         /* Table of names */
2249    BOOL   notbol;                /* NOTBOL flag */    pcre_uchar nl[4];               /* Newline string when fixed */
2250    BOOL   noteol;                /* NOTEOL flag */    const  pcre_uint8 *lcc;         /* Points to lower casing table */
2251    BOOL   utf8;                  /* UTF8 flag */    const  pcre_uint8 *fcc;         /* Points to case-flipping table */
2252    BOOL   endonly;               /* Dollar not before final \n */    const  pcre_uint8 *ctypes;      /* Points to table of type maps */
2253    BOOL   notempty;              /* Empty string match not wanted */    BOOL   offset_overflow;         /* Set if too many extractions */
2254    BOOL   partial;               /* PARTIAL flag */    BOOL   notbol;                  /* NOTBOL flag */
2255    BOOL   hitend;                /* Hit the end of the subject at some point */    BOOL   noteol;                  /* NOTEOL flag */
2256    const uschar *start_code;     /* For use when recursing */    BOOL   utf;                     /* UTF-8 / UTF-16 flag */
2257    USPTR  start_subject;         /* Start of the subject string */    BOOL   jscript_compat;          /* JAVASCRIPT_COMPAT flag */
2258    USPTR  end_subject;           /* End of the subject string */    BOOL   use_ucp;                 /* PCRE_UCP flag */
2259    USPTR  start_match;           /* Start of this match attempt */    BOOL   endonly;                 /* Dollar not before final \n */
2260    USPTR  end_match_ptr;         /* Subject position at end match */    BOOL   notempty;                /* Empty string match not wanted */
2261    int    end_offset_top;        /* Highwater mark at end of match */    BOOL   notempty_atstart;        /* Empty string match at start not wanted */
2262    int    capture_last;          /* Most recent capture number */    BOOL   hitend;                  /* Hit the end of the subject at some point */
2263    int    start_offset;          /* The start offset value */    BOOL   bsr_anycrlf;             /* \R is just any CRLF, not full Unicode */
2264    recursion_info *recursive;    /* Linked list of recursion data */    BOOL   hasthen;                 /* Pattern contains (*THEN) */
2265    void  *callout_data;          /* To pass back to callouts */    BOOL   ignore_skip_arg;         /* For re-run when SKIP name not found */
2266    struct heapframe *thisframe;  /* Used only when compiling for no recursion */    const  pcre_uchar *start_code;  /* For use when recursing */
2267      PCRE_PUCHAR start_subject;      /* Start of the subject string */
2268      PCRE_PUCHAR end_subject;        /* End of the subject string */
2269      PCRE_PUCHAR start_match_ptr;    /* Start of matched string */
2270      PCRE_PUCHAR end_match_ptr;      /* Subject position at end match */
2271      PCRE_PUCHAR start_used_ptr;     /* Earliest consulted character */
2272      int    partial;                 /* PARTIAL options */
2273      int    end_offset_top;          /* Highwater mark at end of match */
2274      int    capture_last;            /* Most recent capture number */
2275      int    start_offset;            /* The start offset value */
2276      int    match_function_type;     /* Set for certain special calls of MATCH() */
2277      eptrblock *eptrchain;           /* Chain of eptrblocks for tail recursions */
2278      int    eptrn;                   /* Next free eptrblock */
2279      recursion_info *recursive;      /* Linked list of recursion data */
2280      void  *callout_data;            /* To pass back to callouts */
2281      const  pcre_uchar *mark;        /* Mark pointer to pass back on success */
2282      const  pcre_uchar *nomatch_mark;/* Mark pointer to pass back on failure */
2283      const  pcre_uchar *once_target; /* Where to back up to for atomic groups */
2284    #ifdef NO_RECURSE
2285      void  *match_frames_base;       /* For remembering malloc'd frames */
2286    #endif
2287  } match_data;  } match_data;
2288    
2289  /* 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
2290  functions. */  functions. */
2291    
2292  typedef struct dfa_match_data {  typedef struct dfa_match_data {
2293    const uschar *start_code;     /* Start of the compiled pattern */    const pcre_uchar *start_code;     /* Start of the compiled pattern */
2294    const uschar *start_subject;  /* Start of the subject string */    const pcre_uchar *start_subject ; /* Start of the subject string */
2295    const uschar *end_subject;    /* End of subject string */    const pcre_uchar *end_subject;    /* End of subject string */
2296    const uschar *tables;         /* Character tables */    const pcre_uchar *start_used_ptr; /* Earliest consulted character */
2297    int   moptions;               /* Match options */    const pcre_uint8 *tables;         /* Character tables */
2298    int   poptions;               /* Pattern options */    int   start_offset;               /* The start offset value */
2299    int    nllen;                 /* 1 or 2 for newline string length */    int   moptions;                   /* Match options */
2300    uschar nl[4];                 /* Newline string */    int   poptions;                   /* Pattern options */
2301    void  *callout_data;          /* To pass back to callouts */    int   nltype;                     /* Newline type */
2302      int   nllen;                      /* Newline string length */
2303      pcre_uchar nl[4];                 /* Newline string when fixed */
2304      void *callout_data;               /* To pass back to callouts */
2305      dfa_recursion_info *recursive;    /* Linked list of recursion data */
2306  } dfa_match_data;  } dfa_match_data;
2307    
2308  /* Bit definitions for entries in the pcre_ctypes table. */  /* Bit definitions for entries in the pcre_ctypes table. */
# Line 876  typedef struct dfa_match_data { Line 2311  typedef struct dfa_match_data {
2311  #define ctype_letter  0x02  #define ctype_letter  0x02
2312  #define ctype_digit   0x04  #define ctype_digit   0x04
2313  #define ctype_xdigit  0x08  #define ctype_xdigit  0x08
2314  #define ctype_word    0x10   /* alphameric or '_' */  #define ctype_word    0x10   /* alphanumeric or '_' */
2315  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */
2316    
2317  /* 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 903  total length. */ Line 2338  total length. */
2338  #define ctypes_offset (cbits_offset + cbit_length)  #define ctypes_offset (cbits_offset + cbit_length)
2339  #define tables_length (ctypes_offset + 256)  #define tables_length (ctypes_offset + 256)
2340    
2341    /* Internal function and data prefixes. */
2342    
2343    #ifdef COMPILE_PCRE8
2344    #ifndef PUBL
2345    #define PUBL(name) pcre_##name
2346    #endif
2347    #ifndef PRIV
2348    #define PRIV(name) _pcre_##name
2349    #endif
2350    #else /* COMPILE_PCRE8 */
2351    #ifdef COMPILE_PCRE16
2352    #ifndef PUBL
2353    #define PUBL(name) pcre16_##name
2354    #endif
2355    #ifndef PRIV
2356    #define PRIV(name) _pcre16_##name
2357    #endif
2358    #else
2359    #error Unsupported compiling mode
2360    #endif /* COMPILE_PCRE16 */
2361    #endif /* COMPILE_PCRE8 */
2362    
2363  /* 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
2364  codes. */  codes. Each entry used to point directly to a name, but to reduce the number of
2365    relocations in shared libraries, it now has an offset into a single string
2366    instead. */
2367    
2368  typedef struct {  typedef struct {
2369    const char *name;    pcre_uint16 name_offset;
2370    pcre_uint16 type;    pcre_uint16 type;
2371    pcre_uint16 value;    pcre_uint16 value;
2372  } ucp_type_table;  } ucp_type_table;
# Line 918  of the exported public functions. They h Line 2377  of the exported public functions. They h
2377  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
2378  pcre_tables.c module. */  pcre_tables.c module. */
2379    
2380  extern const int    _pcre_utf8_table1[];  #ifdef COMPILE_PCRE8
2381  extern const int    _pcre_utf8_table2[];  extern const int            PRIV(utf8_table1)[];
2382  extern const int    _pcre_utf8_table3[];  extern const int            PRIV(utf8_table1_size);
2383  extern const uschar _pcre_utf8_table4[];  extern const int            PRIV(utf8_table2)[];
2384    extern const int            PRIV(utf8_table3)[];
2385  extern const int    _pcre_utf8_table1_size;  extern const pcre_uint8     PRIV(utf8_table4)[];
2386    #endif /* COMPILE_PCRE8 */
2387    
2388    extern const char           PRIV(utt_names)[];
2389    extern const ucp_type_table PRIV(utt)[];
2390    extern const int            PRIV(utt_size);
2391    
2392  extern const ucp_type_table _pcre_utt[];  extern const pcre_uint8     PRIV(OP_lengths)[];
2393  extern const int _pcre_utt_size;  extern const pcre_uint8     PRIV(default_tables)[];
2394    
2395  extern const uschar _pcre_default_tables[];  extern const pcre_uint32    PRIV(hspace_list)[];
2396    extern const pcre_uint32    PRIV(vspace_list)[];
 extern const uschar _pcre_OP_lengths[];  
2397    
2398    
2399  /* Internal shared functions. These are functions that are used by more than  /* Internal shared functions. These are functions that are used by more than
2400  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
2401  sense, but are not part of the PCRE public API. */  sense, but are not part of the PCRE public API. */
2402    
2403  extern int         _pcre_ord2utf8(int, uschar *);  /* String comparison functions. */
2404  extern real_pcre * _pcre_try_flipped(const real_pcre *, real_pcre *,  #ifdef COMPILE_PCRE8
2405                       const pcre_study_data *, pcre_study_data *);  
2406  extern int         _pcre_ucp_findprop(const unsigned int, int *, int *);  #define STRCMP_UC_UC(str1, str2) \
2407  extern int         _pcre_ucp_othercase(const int);    strcmp((char *)(str1), (char *)(str2))
2408  extern int         _pcre_valid_utf8(const uschar *, int);  #define STRCMP_UC_C8(str1, str2) \
2409  extern BOOL        _pcre_xclass(int, const uschar *);    strcmp((char *)(str1), (str2))
2410    #define STRNCMP_UC_UC(str1, str2, num) \
2411      strncmp((char *)(str1), (char *)(str2), (num))
2412    #define STRNCMP_UC_C8(str1, str2, num) \
2413      strncmp((char *)(str1), (str2), (num))
2414    #define STRLEN_UC(str) strlen((const char *)str)
2415    
2416    #else
2417    
2418    extern int               PRIV(strcmp_uc_uc)(const pcre_uchar *,
2419                               const pcre_uchar *);
2420    extern int               PRIV(strcmp_uc_c8)(const pcre_uchar *,
2421                               const char *);
2422    extern int               PRIV(strncmp_uc_uc)(const pcre_uchar *,
2423                               const pcre_uchar *, unsigned int num);
2424    extern int               PRIV(strncmp_uc_c8)(const pcre_uchar *,
2425                               const char *, unsigned int num);
2426    extern unsigned int      PRIV(strlen_uc)(const pcre_uchar *str);
2427    
2428    #define STRCMP_UC_UC(str1, str2) \
2429      PRIV(strcmp_uc_uc)((str1), (str2))
2430    #define STRCMP_UC_C8(str1, str2) \
2431      PRIV(strcmp_uc_c8)((str1), (str2))
2432    #define STRNCMP_UC_UC(str1, str2, num) \
2433      PRIV(strncmp_uc_uc)((str1), (str2), (num))
2434    #define STRNCMP_UC_C8(str1, str2, num) \
2435      PRIV(strncmp_uc_c8)((str1), (str2), (num))
2436    #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2437    
2438    #endif /* COMPILE_PCRE8 */
2439    
2440    extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2441    extern BOOL              PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2442                               int *, BOOL);
2443    extern int               PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
2444    extern int               PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
2445    extern BOOL              PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2446                               int *, BOOL);
2447    extern BOOL              PRIV(xclass)(int, const pcre_uchar *, BOOL);
2448    
2449    #ifdef SUPPORT_JIT
2450    extern void              PRIV(jit_compile)(const REAL_PCRE *,
2451                               PUBL(extra) *, int);
2452    extern int               PRIV(jit_exec)(const REAL_PCRE *, const PUBL(extra) *,
2453                               const pcre_uchar *, int, int, int, int *, int);
2454    extern void              PRIV(jit_free)(void *);
2455    extern int               PRIV(jit_get_size)(void *);
2456    extern const char*       PRIV(jit_get_target)(void);
2457    #endif
2458    
2459    /* Unicode character database (UCD) */
2460    
2461    typedef struct {
2462      pcre_uint8 script;     /* ucp_Arabic, etc. */
2463      pcre_uint8 chartype;   /* ucp_Cc, etc. (general categories) */
2464      pcre_uint8 gbprop;     /* ucp_gbControl, etc. (grapheme break property) */
2465      pcre_uint8 caseset;    /* offset to multichar other cases or zero */
2466      pcre_int32 other_case; /* offset to other case, or zero if none */
2467    } ucd_record;
2468    
2469    extern const pcre_uint32 PRIV(ucd_caseless_sets)[];
2470    extern const ucd_record  PRIV(ucd_records)[];
2471    extern const pcre_uint8  PRIV(ucd_stage1)[];
2472    extern const pcre_uint16 PRIV(ucd_stage2)[];
2473    extern const int         PRIV(ucp_gentype)[];
2474    extern const pcre_uint32 PRIV(ucp_gbtable)[];
2475    #ifdef SUPPORT_JIT
2476    extern const int         PRIV(ucp_typerange)[];
2477    #endif
2478    
2479    #ifdef SUPPORT_UCP
2480    /* UCD access macros */
2481    
2482    #define UCD_BLOCK_SIZE 128
2483    #define GET_UCD(ch) (PRIV(ucd_records) + \
2484            PRIV(ucd_stage2)[PRIV(ucd_stage1)[(ch) / UCD_BLOCK_SIZE] * \
2485            UCD_BLOCK_SIZE + (ch) % UCD_BLOCK_SIZE])
2486    
2487    #define UCD_CHARTYPE(ch)    GET_UCD(ch)->chartype
2488    #define UCD_SCRIPT(ch)      GET_UCD(ch)->script
2489    #define UCD_CATEGORY(ch)    PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2490    #define UCD_GRAPHBREAK(ch)  GET_UCD(ch)->gbprop
2491    #define UCD_CASESET(ch)     GET_UCD(ch)->caseset
2492    #define UCD_OTHERCASE(ch)   (ch + GET_UCD(ch)->other_case)
2493    
2494    #endif /* SUPPORT_UCP */
2495    
2496  #endif  #endif
2497    

Legend:
Removed from v.91  
changed lines
  Added in v.1045

  ViewVC Help
Powered by ViewVC 1.1.5