internal.h

这个是正则表达式的delphi类,如有需要,就下载看看已经是最新版了

/*************************************************
*      Perl-Compatible Regular Expressions       *
*************************************************/


/* This is a library of functions to support regular expressions whose syntax
and semantics are as close as possible to those of the Perl 5 language. See
the file Tech.Notes for some information on the internals.

Written by: Philip Hazel <[e-mail]>

           Copyright (c) 1997-2003 University of Cambridge

-----------------------------------------------------------------------------
Permission is granted to anyone to use this software for any purpose on any
computer system, and to redistribute it freely, subject to the following
restrictions:

1. This software is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

2. The origin of this software must not be misrepresented, either by
   explicit claim or by omission.

3. Altered versions must be plainly marked as such, and must not be
   misrepresented as being the original software.

4. If PCRE is embedded in any software that is released under the GNU
   General Purpose Licence (GPL), then the terms of that licence shall
   supersede any condition above with which it is incompatible.
-----------------------------------------------------------------------------
*/

/* This header contains definitions that are shared between the different
modules, but which are not relevant to the outside. */

/* Get the definitions provided by running "configure" */

#include "config.h"

/* Standard C headers plus the external interface definition. The only time
setjmp and stdarg are used is when NO_RECURSE is set. */

#include <ctype.h>
#include <limits.h>
#include <setjmp.h>
#include <stdarg.h>
#include <stddef.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#ifndef PCRE_SPY
#define PCRE_DEFINITION       /* Win32 __declspec(export) trigger for .dll */
#endif

#include "pcre.h"

/* When compiling for use with the Virtual Pascal compiler, these functions
need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
option on the command line. */

#ifdef VPCOMPAT
#define strncmp(s1,s2,m) _strncmp(s1,s2,m)
#define memcpy(d,s,n)    _memcpy(d,s,n)
#define memmove(d,s,n)   _memmove(d,s,n)
#define memset(s,c,n)    _memset(s,c,n)
#else  /* VPCOMPAT */

/* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
is set. Otherwise, include an emulating function for those systems that have
neither (there some non-Unix environments where this is the case). This assumes
that all calls to memmove are moving strings upwards in store, which is the
case in PCRE. */

#if ! HAVE_MEMMOVE
#undef  memmove        /* some systems may have a macro */
#if HAVE_BCOPY
#define memmove(a, b, c) bcopy(b, a, c)
#else  /* HAVE_BCOPY */
void *
pcre_memmove(unsigned char *dest, const unsigned char *src, size_t n)
{
int i;
dest += n;
src += n;
for (i = 0; i < n; ++i) *(--dest) =  *(--src);
}
#define memmove(a, b, c) pcre_memmove(a, b, c)
#endif   /* not HAVE_BCOPY */
#endif   /* not HAVE_MEMMOVE */
#endif   /* not VPCOMPAT */


/* PCRE keeps offsets in its compiled code as 2-byte quantities by default.
These are used, for example, to link from the start of a subpattern to its
alternatives and its end. The use of 2 bytes per offset limits the size of the
compiled regex to around 64K, which is big enough for almost everybody.
However, I received a request for an even bigger limit. For this reason, and
also to make the code easier to maintain, the storing and loading of offsets
from the byte string is now handled by the macros that are defined here.

The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
the config.h file, but can be overridden by using -D on the command line. This
is automated on Unix systems via the "configure" command. */

#if LINK_SIZE == 2

#define PUT(a,n,d)   \
  (a[n] = (d) >> 8), \
  (a[(n)+1] = (d) & 255)

#define GET(a,n) \
  (((a)[n] << 8) | (a)[(n)+1])

#define MAX_PATTERN_SIZE (1 << 16)


#elif LINK_SIZE == 3

#define PUT(a,n,d)       \
  (a[n] = (d) >> 16),    \
  (a[(n)+1] = (d) >> 8), \
  (a[(n)+2] = (d) & 255)

#define GET(a,n) \
  (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])

#define MAX_PATTERN_SIZE (1 << 24)


#elif LINK_SIZE == 4

#define PUT(a,n,d)        \
  (a[n] = (d) >> 24),     \
  (a[(n)+1] = (d) >> 16), \
  (a[(n)+2] = (d) >> 8),  \
  (a[(n)+3] = (d) & 255)

#define GET(a,n) \
  (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])

#define MAX_PATTERN_SIZE (1 << 30)   /* Keep it positive */


#else
#error LINK_SIZE must be either 2, 3, or 4
#endif


/* Convenience macro defined in terms of the others */

#define PUTINC(a,n,d)   PUT(a,n,d), a += LINK_SIZE


/* PCRE uses some other 2-byte quantities that do not change when the size of
offsets changes. There are used for repeat counts and for other things such as
capturing parenthesis numbers in back references. */

#define PUT2(a,n,d)   \
  a[n] = (d) >> 8; \
  a[(n)+1] = (d) & 255

#define GET2(a,n) \
  (((a)[n] << 8) | (a)[(n)+1])

#define PUT2INC(a,n,d)  PUT2(a,n,d), a += 2


/* In case there is no definition of offsetof() provided - though any proper
Standard C system should have one. */

#ifndef offsetof
#define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
#endif

/* These are the public options that can change during matching. */

#define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)

/* Private options flags start at the most significant end of the four bytes,
but skip the top bit so we can use ints for convenience without getting tangled
with negative values. The public options defined in pcre.h start at the least
significant end. Make sure they don't overlap, though now that we have expanded
to four bytes there is plenty of space. */

#define PCRE_FIRSTSET      0x40000000  /* first_byte is set */
#define PCRE_REQCHSET      0x20000000  /* req_byte is set */
#define PCRE_STARTLINE     0x10000000  /* start after \n for multiline */
#define PCRE_ICHANGED      0x08000000  /* i option changes within regex */

/* Options for the "extra" block produced by pcre_study(). */

#define PCRE_STUDY_MAPPED   0x01     /* a map of starting chars exists */

/* Masks for identifying the public options which are permitted at compile
time, run time or study time, respectively. */

#define PUBLIC_OPTIONS \
  (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
   PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
   PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK)

#define PUBLIC_EXEC_OPTIONS \
  (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK)

#define PUBLIC_STUDY_OPTIONS 0   /* None defined */

/* Magic number to provide a small check against being handed junk. */

#define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */

/* Negative values for the firstchar and reqchar variables */

#define REQ_UNSET (-2)
#define REQ_NONE  (-1)

/* Flags added to firstbyte or reqbyte; a "non-literal" item is either a
variable-length repeat, or a anything other than literal characters. */

#define REQ_CASELESS 0x0100    /* indicates caselessness */
#define REQ_VARY     0x0200    /* reqbyte followed non-literal item */

/* Miscellaneous definitions */

typedef int BOOL;

#define FALSE   0
#define TRUE    1

/* Escape items that are just an encoding of a particular data value. Note that
ESC_n is defined as yet another macro, which is set in config.h to either \n
(the default) or \r (which some people want). */

#ifndef ESC_e
#define ESC_e 27
#endif

#ifndef ESC_f
#define ESC_f '\f'
#endif

#ifndef ESC_n
#define ESC_n NEWLINE
#endif

#ifndef ESC_r
#define ESC_r '\r'
#endif

/* We can't officially use ESC_t because it is a POSIX reserved identifier
(presumably because of all the others like size_t). */

#ifndef ESC_tee
#define ESC_tee '\t'
#endif

/* These are escaped items that aren't just an encoding of a particular data
value such as \n. They must have non-zero values, as check_escape() returns
their negation. Also, they must appear in the same order as in the opcode
definitions below, up to ESC_z. There's a dummy for OP_ANY because it
corresponds to "." rather than an escape sequence. The final one must be
ESC_REF as subsequent values are used for \1, \2, \3, etc. There is are two
tests in the code for an escape greater than ESC_b and less than ESC_Z to
detect the types that may be repeated. These are the types that consume a
character. If any new escapes are put in between that don't consume a
character, that code will have to change. */

enum { ESC_A = 1, ESC_G, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s, ESC_W,
       ESC_w, ESC_dum1, ESC_C, ESC_Z, ESC_z, ESC_E, ESC_Q, ESC_REF };

/* Flag bits and data types for the extended class (OP_XCLASS) for classes that
contain UTF-8 characters with values greater than 255. */

#define XCL_NOT    0x01    /* Flag: this is a negative class */
#define XCL_MAP    0x02    /* Flag: a 32-byte map is present */

#define XCL_END       0    /* Marks end of individual items */
#define XCL_SINGLE    1    /* Single item (one multibyte char) follows */
#define XCL_RANGE     2    /* A range (two multibyte chars) follows */


/* Opcode table: OP_BRA must be last, as all values >= it are used for brackets
that extract substrings. Starting from 1 (i.e. after OP_END), the values up to
OP_EOD must correspond in order to the list of escapes immediately above.
Note that whenever this list is updated, the two macro definitions that follow
must also be updated to match. */

enum {
  OP_END,            /* 0 End of pattern */

  /* Values corresponding to backslashed metacharacters */

  OP_SOD,            /* 1 Start of data: \A */
  OP_SOM,            /* 2 Start of match (subject + offset): \G */
  OP_NOT_WORD_BOUNDARY,  /*  3 \B */
  OP_WORD_BOUNDARY,      /*  4 \b */
  OP_NOT_DIGIT,          /*  5 \D */
  OP_DIGIT,              /*  6 \d */
  OP_NOT_WHITESPACE,     /*  7 \S */
  OP_WHITESPACE,         /*  8 \s */
  OP_NOT_WORDCHAR,       /*  9 \W */
  OP_WORDCHAR,           /* 10 \w */
  OP_ANY,            /* 11 Match any character */
  OP_ANYBYTE,        /* 12 Match any byte (\C); different to OP_ANY for UTF-8 */
  OP_EODN,           /* 13 End of data or \n at end of data: \Z. */
  OP_EOD,            /* 14 End of data: \z */

  OP_OPT,            /* 15 Set runtime options */
  OP_CIRC,           /* 16 Start of line - varies with multiline switch */
  OP_DOLL,           /* 17 End of line - varies with multiline switch */
  OP_CHARS,          /* 18 Match string of characters */
  OP_NOT,            /* 19 Match anything but the following char */

  OP_STAR,           /* 20 The maximizing and minimizing versions of */
  OP_MINSTAR,        /* 21 all these opcodes must come in pairs, with */
  OP_PLUS,           /* 22 the minimizing one second. */
  OP_MINPLUS,        /* 23 This first set applies to single characters */
  OP_QUERY,          /* 24 */
  OP_MINQUERY,       /* 25 */
  OP_UPTO,           /* 26 From 0 to n matches */
  OP_MINUPTO,        /* 27 */
  OP_EXACT,          /* 28 Exactly n matches */

  OP_NOTSTAR,        /* 29 The maximizing and minimizing versions of */
  OP_NOTMINSTAR,     /* 30 all these opcodes must come in pairs, with */
  OP_NOTPLUS,        /* 31 the minimizing one second. */
  OP_NOTMINPLUS,     /* 32 This set applies to "not" single characters */
  OP_NOTQUERY,       /* 33 */
  OP_NOTMINQUERY,    /* 34 */
  OP_NOTUPTO,        /* 35 From 0 to n matches */
  OP_NOTMINUPTO,     /* 36 */
  OP_NOTEXACT,       /* 37 Exactly n matches */

  OP_TYPESTAR,       /* 38 The maximizing and minimizing versions of */
  OP_TYPEMINSTAR,    /* 39 all these opcodes must come in pairs, with */
  OP_TYPEPLUS,       /* 40 the minimizing one second. These codes must */