-/* $Cambridge: exim/src/src/pcre/pcre_internal.h,v 1.2 2005/08/08 10:22:14 ph10 Exp $ */
+/* $Cambridge: exim/src/src/pcre/pcre_internal.h,v 1.3 2006/11/07 16:50:36 ph10 Exp $ */
/*************************************************
* Perl-Compatible Regular Expressions *
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2005 University of Cambridge
+ Copyright (c) 1997-2006 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
modules, but which are not relevant to the exported API. This includes some
functions whose names all begin with "_pcre_". */
+#ifndef PCRE_INTERNAL_H
+#define PCRE_INTERNAL_H
/* Define DEBUG to get debugging output on stdout. */
-/****
+#if 0
#define DEBUG
-****/
+#endif
/* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
inline, and there are *still* stupid compilers about that don't like indented
typedef unsigned char uschar;
-/* Include the public PCRE header */
-
-#include "pcre.h"
+/* PCRE is able to support 3 different kinds of newline (CR, LF, CRLF). The
+following macro is used to package up testing for newlines. NLBLOCK is defined
+in the various modules to indicate in which datablock the parameters exist. */
+
+#define IS_NEWLINE(p) \
+ ((p)[0] == NLBLOCK->nl[0] && \
+ (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]))
+
+/* When PCRE is compiled as a C++ library, the subject pointer can be replaced
+with a custom type. This makes it possible, for example, to allow pcre_exec()
+to process subject strings that are discontinuous by using a smart pointer
+class. It must always be possible to inspect all of the subject string in
+pcre_exec() because of the way it backtracks. Two macros are required in the
+normal case, for sign-unspecified and unsigned char pointers. The former is
+used for the external interface and appears in pcre.h, which is why its name
+must begin with PCRE_. */
+
+#ifdef CUSTOM_SUBJECT_PTR
+#define PCRE_SPTR CUSTOM_SUBJECT_PTR
+#define USPTR CUSTOM_SUBJECT_PTR
+#else
+#define PCRE_SPTR const char *
+#define USPTR const unsigned char *
+#endif
-/* Include the (copy of) the public ucp header, changing the external name into
-a private one. This does no harm, even if we aren't compiling UCP support. */
+/* Include the public PCRE header and the definitions of UCP character property
+values. */
-#define ucp_findchar _pcre_ucp_findchar
+#include "pcre.h"
#include "ucp.h"
/* When compiling for use with the Virtual Pascal compiler, these functions
#if HAVE_BCOPY
#define memmove(a, b, c) bcopy(b, a, c)
#else /* HAVE_BCOPY */
-void *
+static void *
pcre_memmove(unsigned char *dest, const unsigned char *src, size_t n)
{
-int i;
+size_t i;
dest += n;
src += n;
for (i = 0; i < n; ++i) *(--dest) = *(--src);
+return dest;
}
#define memmove(a, b, c) pcre_memmove(a, b, c)
#endif /* not HAVE_BCOPY */
#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! */
+/* Private options flags start at the most significant end of the four bytes.
+The public options defined in pcre.h start at the least significant end. Make
+sure they don't overlap! The bits are getting a bit scarce now -- when we run
+out, there is a dummy word in the structure that could be used for the private
+bits. */
+#define PCRE_NOPARTIAL 0x80000000 /* can't use partial with this regex */
#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 */
-#define PCRE_NOPARTIAL 0x04000000 /* can't use partial with this regex */
+#define PCRE_JCHANGED 0x08000000 /* j option changes within regex */
/* Options for the "extra" block produced by pcre_study(). */
#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|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE)
+ PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
+ PCRE_DUPNAMES|PCRE_NEWLINE_CR|PCRE_NEWLINE_LF)
#define PUBLIC_EXEC_OPTIONS \
(PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \
- PCRE_PARTIAL)
+ PCRE_PARTIAL|PCRE_NEWLINE_CR|PCRE_NEWLINE_LF)
#define PUBLIC_DFA_EXEC_OPTIONS \
(PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \
- PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART)
+ PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_CR| \
+ PCRE_NEWLINE_LF)
#define PUBLIC_STUDY_OPTIONS 0 /* None defined */
#define ESC_tee '\t'
#endif
+/* Codes for different types of Unicode property */
+
+#define PT_ANY 0 /* Any property - matches all chars */
+#define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
+#define PT_GC 2 /* General characteristic (e.g. L) */
+#define PT_PC 3 /* Particular characteristic (e.g. Lu) */
+#define PT_SC 4 /* Script (e.g. Han) */
+
+/* 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 */
+#define XCL_PROP 3 /* Unicode property (2-byte property code follows) */
+#define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
+
/* 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
ESC_w, ESC_dum1, ESC_C, ESC_P, ESC_p, ESC_X, 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 */
-#define XCL_PROP 3 /* Unicode property (one property code) follows */
-#define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
-
-
/* 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.
OP_DOLL, /* 20 End of line - varies with multiline switch */
OP_CHAR, /* 21 Match one character, casefully */
OP_CHARNC, /* 22 Match one character, caselessly */
- OP_NOT, /* 23 Match anything but the following char */
+ OP_NOT, /* 23 Match one character, not the following one */
OP_STAR, /* 24 The maximizing and minimizing versions of */
OP_MINSTAR, /* 25 all these opcodes must come in pairs, with */
1, /* End */ \
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* \A, \G, \B, \B, \D, \d, \S, \s, \W, \w */ \
1, 1, /* Any, Anybyte */ \
- 2, 2, 1, /* NOTPROP, PROP, EXTUNI */ \
+ 3, 3, 1, /* NOTPROP, PROP, EXTUNI */ \
1, 1, 2, 1, 1, /* \Z, \z, Opt, ^, $ */ \
2, /* Char - the minimum length */ \
2, /* Charnc - the minimum length */ \
ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
- ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47 };
+ ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
+ ERR50, ERR51 };
/* The real format of the start of the pcre block; the index of names and the
code vector run on as long as necessary after the end. We store an explicit
unsigned int backref_map; /* Bitmap of low back refs */
int req_varyopt; /* "After variable item" flag for reqbyte */
BOOL nopartial; /* Set TRUE if partial won't work */
+ int nllen; /* 1 or 2 for newline string length */
+ uschar nl[4]; /* Newline string */
} compile_data;
/* Structure for maintaining a chain of pointers to the currently incomplete
struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
int group_num; /* Number of group that was called */
const uschar *after_call; /* "Return value": points after the call in the expr */
- const uschar *save_start; /* Old value of md->start_match */
+ USPTR save_start; /* Old value of md->start_match */
int *offset_save; /* Pointer to start of saved offsets */
int saved_max; /* Number of saved offsets */
} recursion_info;
/* When compiling in a mode that doesn't use recursive calls to match(),
a structure is used to remember local variables on the heap. It is defined in
-pcre.c, close to the match() function, so that it is easy to keep it in step
-with any changes of local variable. However, the pointer to the current frame
-must be saved in some "static" place over a longjmp(). We declare the
-structure here so that we can put a pointer in the match_data structure.
-NOTE: This isn't used for a "normal" compilation of pcre. */
+pcre_exec.c, close to the match() function, so that it is easy to keep it in
+step with any changes of local variable. However, the pointer to the current
+frame must be saved in some "static" place over a longjmp(). We declare the
+structure here so that we can put a pointer in the match_data structure. NOTE:
+This isn't used for a "normal" compilation of pcre. */
struct heapframe;
doing traditional NFA matching, so that they are thread-safe. */
typedef struct match_data {
- unsigned long int match_call_count; /* As it says */
- unsigned long int match_limit;/* As it says */
+ unsigned long int match_call_count; /* As it says */
+ unsigned long int match_limit; /* As it says */
+ unsigned long int match_limit_recursion; /* As it says */
int *offset_vector; /* Offset vector */
int offset_end; /* One past the end */
int offset_max; /* The maximum usable for return data */
+ int nllen; /* 1 or 2 for newline string length */
+ uschar nl[4]; /* Newline string */
const uschar *lcc; /* Points to lower casing table */
const uschar *ctypes; /* Points to table of type maps */
BOOL offset_overflow; /* Set if too many extractions */
BOOL partial; /* PARTIAL flag */
BOOL hitend; /* Hit the end of the subject at some point */
const uschar *start_code; /* For use when recursing */
- const uschar *start_subject; /* Start of the subject string */
- const uschar *end_subject; /* End of the subject string */
- const uschar *start_match; /* Start of this match attempt */
- const uschar *end_match_ptr; /* Subject position at end match */
+ USPTR start_subject; /* Start of the subject string */
+ USPTR end_subject; /* End of the subject string */
+ USPTR start_match; /* Start of this match attempt */
+ USPTR end_match_ptr; /* Subject position at end match */
int end_offset_top; /* Highwater mark at end of match */
int capture_last; /* Most recent capture number */
int start_offset; /* The start offset value */
const uschar *tables; /* Character tables */
int moptions; /* Match options */
int poptions; /* Pattern options */
+ int nllen; /* 1 or 2 for newline string length */
+ uschar nl[4]; /* Newline string */
void *callout_data; /* To pass back to callouts */
} dfa_match_data;
#define ctypes_offset (cbits_offset + cbit_length)
#define tables_length (ctypes_offset + 256)
-/* Layout of the UCP type table that translates property names into codes for
-ucp_findchar(). */
+/* Layout of the UCP type table that translates property names into types and
+codes. */
typedef struct {
const char *name;
- int value;
+ pcre_uint16 type;
+ pcre_uint16 value;
} ucp_type_table;
sense, but are not part of the PCRE public API. */
extern int _pcre_ord2utf8(int, uschar *);
-extern void _pcre_printint(pcre *, FILE *);
extern real_pcre * _pcre_try_flipped(const real_pcre *, real_pcre *,
const pcre_study_data *, pcre_study_data *);
-extern int _pcre_ucp_findchar(const int, int *, int *);
+extern int _pcre_ucp_findprop(const unsigned int, int *, int *);
+extern int _pcre_ucp_othercase(const int);
extern int _pcre_valid_utf8(const uschar *, int);
extern BOOL _pcre_xclass(int, const uschar *);
+#endif
+
/* End of pcre_internal.h */