2007-07-31 Matthias Clasen <mclasen@redhat.com>
+ * glib/pcre/*: Update the internal PCRE to 7.2
+
+2007-07-31 Matthias Clasen <mclasen@redhat.com>
+
* glib/pltcheck.sh: Fix some glitches
* glib/gregex.c: Remove debug spew
PCRE LICENCE
-------------
-PCRE 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.
-
-Release 7 of PCRE is distributed under the terms of the "BSD" licence, as
-specified below. The documentation for PCRE, supplied in the "doc"
-directory, is distributed under the same terms as the software itself.
-
-The basic library functions are written in C and are freestanding. Also
-included in the distribution is a set of C++ wrapper functions.
-
-
-THE BASIC LIBRARY FUNCTIONS
----------------------------
-
-Written by: Philip Hazel
-Email local part: ph10
-Email domain: cam.ac.uk
-
-University of Cambridge Computing Service,
-Cambridge, England. Phone: +44 1223 334714.
-
-Copyright (c) 1997-2006 University of Cambridge
-All rights reserved.
-
-
-THE C++ WRAPPER FUNCTIONS
--------------------------
-
-Contributed by: Google Inc.
-
-Copyright (c) 2006, Google Inc.
-All rights reserved.
-
-
-THE "BSD" LICENCE
------------------
-
-Redistribution and use in source and binary forms, with or without
-modification, are permitted provided that the following conditions are met:
-
- * Redistributions of source code must retain the above copyright notice,
- this list of conditions and the following disclaimer.
-
- * Redistributions in binary form must reproduce the above copyright
- notice, this list of conditions and the following disclaimer in the
- documentation and/or other materials provided with the distribution.
-
- * Neither the name of the University of Cambridge nor the name of Google
- Inc. nor the names of their contributors may be used to endorse or
- promote products derived from this software without specific prior
- written permission.
-
-THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
-AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
-ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
-LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
-CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
-SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
-INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
-CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
-ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-POSSIBILITY OF SUCH DAMAGE.
+Please see the file LICENCE in the PCRE distribution for licensing details.
End
-DMAX_NAME_COUNT=10000 \
-DMAX_DUPLENGTH=30000 \
-DLINK_SIZE=2 \
- -DEBCDIC=0 \
+ -UEBCDIC \
-DPOSIX_MALLOC_THRESHOLD=10 \
-I$(top_srcdir) \
-I$(srcdir) \
/* This is the public header file for the PCRE library, to be #included by
applications that call the PCRE functions.
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
/* The current PCRE version information. */
-/* NOTES FOR FUTURE MAINTAINERS: Do not use numbers with leading zeros, because
-they may be treated as octal constants. The PCRE_PRERELEASE feature is for
-identifying release candidates. It might be defined as -RC2, for example. In
-real releases, it should be defined empty. Do not change the alignment of these
-statments. The code in ./configure greps out the version numbers by using "cut"
-to get values from column 29 onwards. These are substituted into pcre-config
-and libpcre.pc. The values are not put into configure.ac and substituted here
-(which would simplify this issue) because that makes life harder for those who
-cannot run ./configure. As it now stands, this file need not be edited in that
-circumstance. */
-
#define PCRE_MAJOR 7
-#define PCRE_MINOR 0
-#define PCRE_PRERELEASE
-#define PCRE_DATE 18-Dec-2006
-
-/* Win32 uses DLL by default; it needs special stuff for exported functions
-when building PCRE. */
-
-/* But don't do that when building as part of GLib */
-#if 0
-#ifdef _WIN32
-# ifdef PCRE_DEFINITION
-# ifdef DLL_EXPORT
-# define PCRE_DATA_SCOPE __declspec(dllexport)
-# endif
-# else
+#define PCRE_MINOR 2
+#define PCRE_PRERELEASE
+#define PCRE_DATE 2007-06-19
+
+/* When an application links to a PCRE DLL in Windows, the symbols that are
+imported have to be identified as such. When building PCRE, the appropriate
+export setting is defined in pcre_internal.h, which includes this file. So we
+don't change an existing definition of PCRE_EXP_DECL. */
+
+#ifndef PCRE_EXP_DECL
+# ifdef _WIN32
# ifndef PCRE_STATIC
-# define PCRE_DATA_SCOPE extern __declspec(dllimport)
+# define PCRE_EXP_DECL extern __declspec(dllimport)
# endif
# endif
#endif
-#endif
-/* Otherwise, we use the standard "extern". */
+/* By default, we use the standard "extern" declarations. */
-#ifndef PCRE_DATA_SCOPE
+#ifndef PCRE_EXP_DECL
# ifdef __cplusplus
-# define PCRE_DATA_SCOPE extern "C"
+# define PCRE_EXP_DECL extern "C"
# else
-# define PCRE_DATA_SCOPE extern
+# define PCRE_EXP_DECL extern
# endif
#endif
#define PCRE_NEWLINE_LF 0x00200000
#define PCRE_NEWLINE_CRLF 0x00300000
#define PCRE_NEWLINE_ANY 0x00400000
+#define PCRE_NEWLINE_ANYCRLF 0x00500000
/* Exec-time and get/set-time error codes */
#define PCRE_INFO_NAMETABLE 9
#define PCRE_INFO_STUDYSIZE 10
#define PCRE_INFO_DEFAULT_TABLES 11
+#define PCRE_INFO_OKPARTIAL 12
+#define PCRE_INFO_JCHANGED 13
/* Request types for pcre_config(). Do not re-arrange, in order to remain
compatible. */
#define pcre_free g_free
#define pcre_stack_malloc g_try_malloc
-PCRE_DATA_SCOPE int (*pcre_callout)(pcre_callout_block *);
+PCRE_EXP_DECL int (*pcre_callout)(pcre_callout_block *);
/* Exported PCRE functions */
-PCRE_DATA_SCOPE pcre *pcre_compile(const char *, int, const char **, int *,
+PCRE_EXP_DECL pcre *pcre_compile(const char *, int, const char **, int *,
const unsigned char *);
-PCRE_DATA_SCOPE pcre *pcre_compile2(const char *, int, int *, const char **,
+PCRE_EXP_DECL pcre *pcre_compile2(const char *, int, int *, const char **,
int *, const unsigned char *);
-PCRE_DATA_SCOPE int pcre_config(int, void *);
-PCRE_DATA_SCOPE int pcre_copy_named_substring(const pcre *, const char *,
+PCRE_EXP_DECL int pcre_config(int, void *);
+PCRE_EXP_DECL int pcre_copy_named_substring(const pcre *, const char *,
int *, int, const char *, char *, int);
-PCRE_DATA_SCOPE int pcre_copy_substring(const char *, int *, int, int, char *,
+PCRE_EXP_DECL int pcre_copy_substring(const char *, int *, int, int, char *,
int);
-PCRE_DATA_SCOPE int pcre_dfa_exec(const pcre *, const pcre_extra *,
+PCRE_EXP_DECL int pcre_dfa_exec(const pcre *, const pcre_extra *,
const char *, int, int, int, int *, int , int *, int);
-PCRE_DATA_SCOPE int pcre_exec(const pcre *, const pcre_extra *, PCRE_SPTR,
+PCRE_EXP_DECL int pcre_exec(const pcre *, const pcre_extra *, PCRE_SPTR,
int, int, int, int *, int);
-PCRE_DATA_SCOPE void pcre_free_substring(const char *);
-PCRE_DATA_SCOPE void pcre_free_substring_list(const char **);
-PCRE_DATA_SCOPE int pcre_fullinfo(const pcre *, const pcre_extra *, int,
+PCRE_EXP_DECL void pcre_free_substring(const char *);
+PCRE_EXP_DECL void pcre_free_substring_list(const char **);
+PCRE_EXP_DECL int pcre_fullinfo(const pcre *, const pcre_extra *, int,
void *);
-PCRE_DATA_SCOPE int pcre_get_named_substring(const pcre *, const char *,
+PCRE_EXP_DECL int pcre_get_named_substring(const pcre *, const char *,
int *, int, const char *, const char **);
-PCRE_DATA_SCOPE int pcre_get_stringnumber(const pcre *, const char *);
-PCRE_DATA_SCOPE int pcre_get_stringtable_entries(const pcre *, const char *,
+PCRE_EXP_DECL int pcre_get_stringnumber(const pcre *, const char *);
+PCRE_EXP_DECL int pcre_get_stringtable_entries(const pcre *, const char *,
char **, char **);
-PCRE_DATA_SCOPE int pcre_get_substring(const char *, int *, int, int,
+PCRE_EXP_DECL int pcre_get_substring(const char *, int *, int, int,
const char **);
-PCRE_DATA_SCOPE int pcre_get_substring_list(const char *, int *, int,
+PCRE_EXP_DECL int pcre_get_substring_list(const char *, int *, int,
const char ***);
-PCRE_DATA_SCOPE int pcre_info(const pcre *, int *, int *);
-PCRE_DATA_SCOPE const unsigned char *pcre_maketables(void);
-PCRE_DATA_SCOPE int pcre_refcount(pcre *, int);
-PCRE_DATA_SCOPE pcre_extra *pcre_study(const pcre *, int, const char **);
-PCRE_DATA_SCOPE const char *pcre_version(void);
+PCRE_EXP_DECL int pcre_info(const pcre *, int *, int *);
+PCRE_EXP_DECL const unsigned char *pcre_maketables(void);
+PCRE_EXP_DECL int pcre_refcount(pcre *, int);
+PCRE_EXP_DECL pcre_extra *pcre_study(const pcre *, int, const char **);
+PCRE_EXP_DECL const char *pcre_version(void);
#ifdef __cplusplus
} /* extern "C" */
-/* This file is autogenerated by ../update-pcre/update.sh during
- * the update of the local copy of PCRE.
- */
/*************************************************
* Perl-Compatible Regular Expressions *
*************************************************/
-/* This file is automatically written by the dftables auxiliary
-program. If you edit it by hand, you might like to edit the Makefile to
-prevent its ever being regenerated.
+/* This file contains character tables that are used when no external tables
+are passed to PCRE by the application that calls it. The tables are used only
+for characters whose code values are less than 256.
-This file contains the default tables for characters with codes less than
-128 (ASCII characters). These tables are used when no external tables are
-passed to PCRE.
+This is a default version of the tables that assumes ASCII encoding. A program
+called dftables (which is distributed with PCRE) can be used to build
+alternative versions of this file. This is necessary if you are running in an
+EBCDIC environment, or if you want to default to a different encoding, for
+example ISO-8859-1. When dftables is run, it creates these tables in the
+current locale. If PCRE is configured with --enable-rebuild-chartables, this
+happens automatically.
-The following #include is present because without it gcc 4.x may remove
-the array definition from the final binary if PCRE is built into a static
-library and dead code stripping is activated. This leads to link errors.
-Pulling in the header ensures that the array gets flagged as "someone
-outside this compilation unit might reference this" and so it will always
-be supplied to the linker. */
+The following #include is present because without it gcc 4.x may remove the
+array definition from the final binary if PCRE is built into a static library
+and dead code stripping is activated. This leads to link errors. Pulling in the
+header ensures that the array gets flagged as "someone outside this compilation
+unit might reference this" and so it will always be supplied to the linker. */
#include "pcre_internal.h"
240,241,242,243,244,245,246,247,
248,249,250,251,252,253,254,255,
-/* This table contains bit maps for various character classes.
-Each map is 32 bytes long and the bits run from the least
-significant end of each byte. The classes that have their own
-maps are: space, xdigit, digit, upper, lower, word, graph
-print, punct, and cntrl. Other classes are built from combinations. */
+/* This table contains bit maps for various character classes. Each map is 32
+bytes long and the bits run from the least significant end of each byte. The
+classes that have their own maps are: space, xdigit, digit, upper, lower, word,
+graph, print, punct, and cntrl. Other classes are built from combinations. */
0x00,0x3e,0x00,0x00,0x01,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */
-/* End of chartables.c */
+/* End of pcre_chartables.c */
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
#endif
+/* Macro for setting individual bits in class bitmaps. */
+
+#define SETBIT(a,b) a[b/8] |= (1 << (b%8))
+
+
/*************************************************
* Code parameters and static tables *
*************************************************/
on. Zero means further processing is needed (for things like \x), or the escape
is invalid. */
-#if !EBCDIC /* This is the "normal" table for ASCII systems */
+#ifndef EBCDIC /* This is the "normal" table for ASCII systems */
static const short int escapes[] = {
0, 0, 0, 0, 0, 0, 0, 0, /* 0 - 7 */
0, 0, ':', ';', '<', '=', '>', '?', /* 8 - ? */
'@', -ESC_A, -ESC_B, -ESC_C, -ESC_D, -ESC_E, 0, -ESC_G, /* @ - G */
- 0, 0, 0, 0, 0, 0, 0, 0, /* H - O */
--ESC_P, -ESC_Q, -ESC_R, -ESC_S, 0, 0, 0, -ESC_W, /* P - W */
+-ESC_H, 0, 0, -ESC_K, 0, 0, 0, 0, /* H - O */
+-ESC_P, -ESC_Q, -ESC_R, -ESC_S, 0, 0, -ESC_V, -ESC_W, /* P - W */
-ESC_X, 0, -ESC_Z, '[', '\\', ']', '^', '_', /* X - _ */
'`', 7, -ESC_b, 0, -ESC_d, ESC_e, ESC_f, 0, /* ` - g */
- 0, 0, 0, -ESC_k, 0, 0, ESC_n, 0, /* h - o */
--ESC_p, 0, ESC_r, -ESC_s, ESC_tee, 0, 0, -ESC_w, /* p - w */
+-ESC_h, 0, 0, -ESC_k, 0, 0, ESC_n, 0, /* h - o */
+-ESC_p, 0, ESC_r, -ESC_s, ESC_tee, 0, -ESC_v, -ESC_w, /* p - w */
0, 0, -ESC_z /* x - z */
};
-#else /* This is the "abnormal" table for EBCDIC systems */
+#else /* This is the "abnormal" table for EBCDIC systems */
static const short int escapes[] = {
/* 48 */ 0, 0, 0, '.', '<', '(', '+', '|',
/* 50 */ '&', 0, 0, 0, 0, 0, 0, 0,
/* 70 */ 0, 0, 0, 0, 0, 0, 0, 0,
/* 78 */ 0, '`', ':', '#', '@', '\'', '=', '"',
/* 80 */ 0, 7, -ESC_b, 0, -ESC_d, ESC_e, ESC_f, 0,
-/* 88 */ 0, 0, 0, '{', 0, 0, 0, 0,
+/* 88 */-ESC_h, 0, 0, '{', 0, 0, 0, 0,
/* 90 */ 0, 0, -ESC_k, 'l', 0, ESC_n, 0, -ESC_p,
/* 98 */ 0, ESC_r, 0, '}', 0, 0, 0, 0,
-/* A0 */ 0, '~', -ESC_s, ESC_tee, 0, 0, -ESC_w, 0,
+/* A0 */ 0, '~', -ESC_s, ESC_tee, 0,-ESC_v, -ESC_w, 0,
/* A8 */ 0,-ESC_z, 0, 0, 0, '[', 0, 0,
/* B0 */ 0, 0, 0, 0, 0, 0, 0, 0,
/* B8 */ 0, 0, 0, 0, 0, ']', '=', '-',
/* C0 */ '{',-ESC_A, -ESC_B, -ESC_C, -ESC_D,-ESC_E, 0, -ESC_G,
-/* C8 */ 0, 0, 0, 0, 0, 0, 0, 0,
+/* C8 */-ESC_H, 0, 0, 0, 0, 0, 0, 0,
/* D0 */ '}', 0, 0, 0, 0, 0, 0, -ESC_P,
/* D8 */-ESC_Q,-ESC_R, 0, 0, 0, 0, 0, 0,
-/* E0 */ '\\', 0, -ESC_S, 0, 0, 0, -ESC_W, -ESC_X,
+/* E0 */ '\\', 0, -ESC_S, 0, 0,-ESC_V, -ESC_W, -ESC_X,
/* E8 */ 0,-ESC_Z, 0, 0, 0, 0, 0, 0,
/* F0 */ 0, 0, 0, 0, 0, 0, 0, 0,
/* F8 */ 0, 0, 0, 0, 0, 0, 0, 0
#define DEAD(s) "\0"
-static const char error_texts[] =
+static const char error_texts[] =
"no error\0"
"\\ at end of pattern\0"
"\\c at end of pattern\0"
"malformed number or name after (?(\0"
"conditional group contains more than two branches\0"
"assertion expected after (?(\0"
- "(?R or (?digits must be followed by )\0"
+ "(?R or (?[+-]digits must be followed by )\0"
/* 30 */
"unknown POSIX class name\0"
"POSIX collating elements are not supported\0"
/* 55 */
"repeating a DEFINE group is not allowed\0"
"inconsistent NEWLINE options\0"
- "\\g is not followed by an (optionally braced) non-zero number";
+ "\\g is not followed by a braced name or an optionally braced non-zero number\0"
+ "(?+ or (?- or (?(+ or (?(- must be followed by a non-zero number";
static const int error_texts_offsets[] = {
0,
1796,
1839,
1879,
- 1908
+ 1908,
+ 1984
};
-
-
/* Definition to allow mutual recursion */
static BOOL
- compile_regex(int, int, uschar **, const uschar **, int *, BOOL, int, int *,
- int *, branch_chain *, compile_data *, int *);
+ compile_regex(int, int, uschar **, const uschar **, int *, BOOL, BOOL, int,
+ int *, int *, branch_chain *, compile_data *, int *);
a table. A non-zero result is something that can be returned immediately.
Otherwise further processing may be required. */
-#if !EBCDIC /* ASCII coding */
+#ifndef EBCDIC /* ASCII coding */
else if (c < '0' || c > 'z') {} /* Not alphameric */
else if ((i = escapes[c - '0']) != 0) c = i;
-#else /* EBCDIC coding */
+#else /* EBCDIC coding */
else if (c < 'a' || (ebcdic_chartab[c] & 0x0E) == 0) {} /* Not alphameric */
else if ((i = escapes[c - 0x48]) != 0) c = i;
#endif
/* \g must be followed by a number, either plain or braced. If positive, it
is an absolute backreference. If negative, it is a relative backreference.
- This is a Perl 5.10 feature. */
+ This is a Perl 5.10 feature. Perl 5.10 also supports \g{name} as a
+ reference to a named group. This is part of Perl's movement towards a
+ unified syntax for back references. As this is synonymous with \k{name}, we
+ fudge it up by pretending it really was \k. */
case 'g':
if (ptr[1] == '{')
{
+ const uschar *p;
+ for (p = ptr+2; *p != 0 && *p != '}'; p++)
+ if (*p != '-' && g_ascii_isdigit(*p) == 0) break;
+ if (*p != 0 && *p != '}')
+ {
+ c = -ESC_k;
+ break;
+ }
braced = TRUE;
ptr++;
}
if (c == 0 && cc == '0') continue; /* Leading zeroes */
count++;
-#if !EBCDIC /* ASCII coding */
+#ifndef EBCDIC /* ASCII coding */
if (cc >= 'a') cc -= 32; /* Convert to upper case */
c = (c << 4) + cc - ((cc < 'A')? '0' : ('A' - 10));
-#else /* EBCDIC coding */
+#else /* EBCDIC coding */
if (cc >= 'a' && cc <= 'z') cc += 64; /* Convert to upper case */
c = (c << 4) + cc - ((cc >= '0')? '0' : ('A' - 10));
#endif
{
int cc; /* Some compilers don't like ++ */
cc = *(++ptr); /* in initializers */
-#if !EBCDIC /* ASCII coding */
+#ifndef EBCDIC /* ASCII coding */
if (cc >= 'a') cc -= 32; /* Convert to upper case */
c = c * 16 + cc - ((cc < 'A')? '0' : ('A' - 10));
-#else /* EBCDIC coding */
+#else /* EBCDIC coding */
if (cc <= 'z') cc += 64; /* Convert to upper case */
c = c * 16 + cc - ((cc >= '0')? '0' : ('A' - 10));
#endif
return 0;
}
-#if !EBCDIC /* ASCII coding */
+#ifndef EBCDIC /* ASCII coding */
if (c >= 'a' && c <= 'z') c -= 32;
c ^= 0x40;
-#else /* EBCDIC coding */
+#else /* EBCDIC coding */
if (c >= 'a' && c <= 'z') c += 64;
c ^= 0xC0;
#endif
else
{
code += _pcre_OP_lengths[c];
+#ifdef SUPPORT_UTF8
if (utf8) switch(c)
{
case OP_CHAR:
if (code[-1] >= 0xc0) code += _pcre_utf8_table4[code[-1] & 0x3f];
break;
}
+#endif
}
}
}
else
{
code += _pcre_OP_lengths[c];
+#ifdef SUPPORT_UTF8
if (utf8) switch(c)
{
case OP_CHAR:
if (code[-1] >= 0xc0) code += _pcre_utf8_table4[code[-1] & 0x3f];
break;
}
+#endif
}
}
}
c = *code;
+ /* Groups with zero repeats can of course be empty; skip them. */
+
+ if (c == OP_BRAZERO || c == OP_BRAMINZERO)
+ {
+ code += _pcre_OP_lengths[c];
+ do code += GET(code, 1); while (*code == OP_ALT);
+ c = *code;
+ continue;
+ }
+
+ /* For other groups, scan the branches. */
+
if (c == OP_BRA || c == OP_CBRA || c == OP_ONCE)
{
BOOL empty_branch;
}
while (*code == OP_ALT);
if (!empty_branch) return FALSE; /* All branches are non-empty */
-
- /* Move past the KET and fudge things so that the increment in the "for"
- above has no effect. */
-
- c = OP_END;
- code += 1 + LINK_SIZE - _pcre_OP_lengths[c];
+ c = *code;
continue;
}
int yield = 0;
while (posix_name_lengths[yield] != 0)
{
- if (len == posix_name_lengths[yield] &&
- strcmp((const char *)ptr, posix_names + offset) == 0) return yield;
+ if (len == posix_name_lengths[yield] &&
+ strcmp((const char *)ptr, posix_names + offset) == 0) return yield;
offset += posix_name_lengths[yield] + 1;
yield++;
}
case OP_NOT_WORDCHAR:
return next <= 127 && (cd->ctypes[next] & ctype_word) != 0;
+ case OP_HSPACE:
+ case OP_NOT_HSPACE:
+ switch(next)
+ {
+ case 0x09:
+ case 0x20:
+ case 0xa0:
+ case 0x1680:
+ case 0x180e:
+ case 0x2000:
+ case 0x2001:
+ case 0x2002:
+ case 0x2003:
+ case 0x2004:
+ case 0x2005:
+ case 0x2006:
+ case 0x2007:
+ case 0x2008:
+ case 0x2009:
+ case 0x200A:
+ case 0x202f:
+ case 0x205f:
+ case 0x3000:
+ return op_code != OP_HSPACE;
+ default:
+ return op_code == OP_HSPACE;
+ }
+
+ case OP_VSPACE:
+ case OP_NOT_VSPACE:
+ switch(next)
+ {
+ case 0x0a:
+ case 0x0b:
+ case 0x0c:
+ case 0x0d:
+ case 0x85:
+ case 0x2028:
+ case 0x2029:
+ return op_code != OP_VSPACE;
+ default:
+ return op_code == OP_VSPACE;
+ }
+
default:
return FALSE;
}
case ESC_W:
return item <= 127 && (cd->ctypes[item] & ctype_word) != 0;
+ case ESC_h:
+ case ESC_H:
+ switch(item)
+ {
+ case 0x09:
+ case 0x20:
+ case 0xa0:
+ case 0x1680:
+ case 0x180e:
+ case 0x2000:
+ case 0x2001:
+ case 0x2002:
+ case 0x2003:
+ case 0x2004:
+ case 0x2005:
+ case 0x2006:
+ case 0x2007:
+ case 0x2008:
+ case 0x2009:
+ case 0x200A:
+ case 0x202f:
+ case 0x205f:
+ case 0x3000:
+ return -next != ESC_h;
+ default:
+ return -next == ESC_h;
+ }
+
+ case ESC_v:
+ case ESC_V:
+ switch(item)
+ {
+ case 0x0a:
+ case 0x0b:
+ case 0x0c:
+ case 0x0d:
+ case 0x85:
+ case 0x2028:
+ case 0x2029:
+ return -next != ESC_v;
+ default:
+ return -next == ESC_v;
+ }
+
default:
return FALSE;
}
case OP_DIGIT:
- return next == -ESC_D || next == -ESC_s || next == -ESC_W;
+ return next == -ESC_D || next == -ESC_s || next == -ESC_W ||
+ next == -ESC_h || next == -ESC_v;
case OP_NOT_DIGIT:
return next == -ESC_d;
return next == -ESC_S || next == -ESC_d || next == -ESC_w;
case OP_NOT_WHITESPACE:
- return next == -ESC_s;
+ return next == -ESC_s || next == -ESC_h || next == -ESC_v;
+
+ case OP_HSPACE:
+ return next == -ESC_S || next == -ESC_H || next == -ESC_d || next == -ESC_w;
+
+ case OP_NOT_HSPACE:
+ return next == -ESC_h;
+
+ /* Can't have \S in here because VT matches \S (Perl anomaly) */
+ case OP_VSPACE:
+ return next == -ESC_V || next == -ESC_d || next == -ESC_w;
+
+ case OP_NOT_VSPACE:
+ return next == -ESC_v;
case OP_WORDCHAR:
- return next == -ESC_W || next == -ESC_s;
+ return next == -ESC_W || next == -ESC_s || next == -ESC_h || next == -ESC_v;
case OP_NOT_WORDCHAR:
return next == -ESC_w || next == -ESC_d;
BOOL possessive_quantifier;
BOOL is_quantifier;
BOOL is_recurse;
+ BOOL reset_bracount;
int class_charcount;
int class_lastchar;
int newoptions;
int recno;
+ int refsign;
int skipbytes;
int subreqbyte;
int subfirstbyte;
else if (c == -ESC_d || c == -ESC_D || c == -ESC_w ||
c == -ESC_W || c == -ESC_s || c == -ESC_S) continue;
+ /* We need to deal with \H, \h, \V, and \v in both phases because
+ they use extra memory. */
+
+ if (-c == ESC_h)
+ {
+ SETBIT(classbits, 0x09); /* VT */
+ SETBIT(classbits, 0x20); /* SPACE */
+ SETBIT(classbits, 0xa0); /* NSBP */
+#ifdef SUPPORT_UTF8
+ if (utf8)
+ {
+ class_utf8 = TRUE;
+ *class_utf8data++ = XCL_SINGLE;
+ class_utf8data += _pcre_ord2utf8(0x1680, class_utf8data);
+ *class_utf8data++ = XCL_SINGLE;
+ class_utf8data += _pcre_ord2utf8(0x180e, class_utf8data);
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x2000, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x200A, class_utf8data);
+ *class_utf8data++ = XCL_SINGLE;
+ class_utf8data += _pcre_ord2utf8(0x202f, class_utf8data);
+ *class_utf8data++ = XCL_SINGLE;
+ class_utf8data += _pcre_ord2utf8(0x205f, class_utf8data);
+ *class_utf8data++ = XCL_SINGLE;
+ class_utf8data += _pcre_ord2utf8(0x3000, class_utf8data);
+ }
+#endif
+ continue;
+ }
+
+ if (-c == ESC_H)
+ {
+ for (c = 0; c < 32; c++)
+ {
+ int x = 0xff;
+ switch (c)
+ {
+ case 0x09/8: x ^= 1 << (0x09%8); break;
+ case 0x20/8: x ^= 1 << (0x20%8); break;
+ case 0xa0/8: x ^= 1 << (0xa0%8); break;
+ default: break;
+ }
+ classbits[c] |= x;
+ }
+
+#ifdef SUPPORT_UTF8
+ if (utf8)
+ {
+ class_utf8 = TRUE;
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x0100, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x167f, class_utf8data);
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x1681, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x180d, class_utf8data);
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x180f, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x1fff, class_utf8data);
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x200B, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x202e, class_utf8data);
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x2030, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x205e, class_utf8data);
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x2060, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x2fff, class_utf8data);
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x3001, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x7fffffff, class_utf8data);
+ }
+#endif
+ continue;
+ }
+
+ if (-c == ESC_v)
+ {
+ SETBIT(classbits, 0x0a); /* LF */
+ SETBIT(classbits, 0x0b); /* VT */
+ SETBIT(classbits, 0x0c); /* FF */
+ SETBIT(classbits, 0x0d); /* CR */
+ SETBIT(classbits, 0x85); /* NEL */
+#ifdef SUPPORT_UTF8
+ if (utf8)
+ {
+ class_utf8 = TRUE;
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x2028, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x2029, class_utf8data);
+ }
+#endif
+ continue;
+ }
+
+ if (-c == ESC_V)
+ {
+ for (c = 0; c < 32; c++)
+ {
+ int x = 0xff;
+ switch (c)
+ {
+ case 0x0a/8: x ^= 1 << (0x0a%8);
+ x ^= 1 << (0x0b%8);
+ x ^= 1 << (0x0c%8);
+ x ^= 1 << (0x0d%8);
+ break;
+ case 0x85/8: x ^= 1 << (0x85%8); break;
+ default: break;
+ }
+ classbits[c] |= x;
+ }
+
+#ifdef SUPPORT_UTF8
+ if (utf8)
+ {
+ class_utf8 = TRUE;
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x0100, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x2027, class_utf8data);
+ *class_utf8data++ = XCL_RANGE;
+ class_utf8data += _pcre_ord2utf8(0x2029, class_utf8data);
+ class_utf8data += _pcre_ord2utf8(0x7fffffff, class_utf8data);
+ }
+#endif
+ continue;
+ }
+
/* We need to deal with \P and \p in both phases. */
#ifdef SUPPORT_UCP
unsigned int origd = d;
while (get_othercase_range(&cc, origd, &occ, &ocd))
{
- if (occ >= c && ocd <= d) continue; /* Skip embedded ranges */
+ if (occ >= (unsigned int)c &&
+ ocd <= (unsigned int)d)
+ continue; /* Skip embedded ranges */
- if (occ < c && ocd >= c - 1) /* Extend the basic range */
+ if (occ < (unsigned int)c &&
+ ocd >= (unsigned int)c - 1) /* Extend the basic range */
{ /* if there is overlap, */
c = occ; /* noting that if occ < c */
continue; /* we can't have ocd > d */
} /* because a subrange is */
- if (ocd > d && occ <= d + 1) /* always shorter than */
+ if (ocd > (unsigned int)d &&
+ occ <= (unsigned int)d + 1) /* always shorter than */
{ /* the basic range. */
d = ocd;
continue;
skipbytes = 0;
bravalue = OP_CBRA;
save_hwm = cd->hwm;
+ reset_bracount = FALSE;
if (*(++ptr) == '?')
{
/* ------------------------------------------------------------ */
+ case '|': /* Reset capture count for each branch */
+ reset_bracount = TRUE;
+ /* Fall through */
+
+ /* ------------------------------------------------------------ */
case ':': /* Non-capturing bracket */
bravalue = OP_BRA;
ptr++;
code[1+LINK_SIZE] = OP_CREF;
skipbytes = 3;
+ refsign = -1;
/* Check for a test for recursion in a named group. */
terminator = '\'';
ptr++;
}
- else terminator = 0;
+ else
+ {
+ terminator = 0;
+ if (ptr[1] == '-' || ptr[1] == '+') refsign = *(++ptr);
+ }
/* We now expect to read a name; any thing else is an error */
if (lengthptr != NULL) break;
/* In the real compile we do the work of looking for the actual
- reference. */
+ reference. If the string started with "+" or "-" we require the rest to
+ be digits, in which case recno will be set. */
+
+ if (refsign > 0)
+ {
+ if (recno <= 0)
+ {
+ *errorcodeptr = ERR58;
+ goto FAILED;
+ }
+ if (refsign == '-')
+ {
+ recno = cd->bracount - recno + 1;
+ if (recno <= 0)
+ {
+ *errorcodeptr = ERR15;
+ goto FAILED;
+ }
+ }
+ else recno += cd->bracount;
+ PUT2(code, 2+LINK_SIZE, recno);
+ break;
+ }
+
+ /* Otherwise (did not start with "+" or "-"), start by looking for the
+ name. */
slot = cd->name_table;
for (i = 0; i < cd->names_found; i++)
/* ------------------------------------------------------------ */
+ case '-': case '+':
case '0': case '1': case '2': case '3': case '4': /* Recursion or */
case '5': case '6': case '7': case '8': case '9': /* subroutine */
{
const uschar *called;
+
+ if ((refsign = *ptr) == '+') ptr++;
+ else if (refsign == '-')
+ {
+ if (g_ascii_isdigit(ptr[1]) == 0)
+ goto OTHER_CHAR_AFTER_QUERY;
+ ptr++;
+ }
+
recno = 0;
while(g_ascii_isdigit(*ptr) != 0)
recno = recno * 10 + *ptr++ - '0';
+
if (*ptr != ')')
{
*errorcodeptr = ERR29;
goto FAILED;
}
+ if (refsign == '-')
+ {
+ if (recno == 0)
+ {
+ *errorcodeptr = ERR58;
+ goto FAILED;
+ }
+ recno = cd->bracount - recno + 1;
+ if (recno <= 0)
+ {
+ *errorcodeptr = ERR15;
+ goto FAILED;
+ }
+ }
+ else if (refsign == '+')
+ {
+ if (recno == 0)
+ {
+ *errorcodeptr = ERR58;
+ goto FAILED;
+ }
+ recno += cd->bracount;
+ }
+
/* Come here from code above that handles a named recursion */
HANDLE_RECURSION:
/* ------------------------------------------------------------ */
default: /* Other characters: check option setting */
+ OTHER_CHAR_AFTER_QUERY:
set = unset = 0;
optset = &set;
errorcodeptr, /* Where to put an error message */
(bravalue == OP_ASSERTBACK ||
bravalue == OP_ASSERTBACK_NOT), /* TRUE if back assert */
+ reset_bracount, /* True if (?| group */
skipbytes, /* Skip over bracket number */
&subfirstbyte, /* For possible first char */
&subreqbyte, /* For possible last char */
is on the bracket. */
/* If this is a conditional bracket, check that there are no more than
- two branches in the group, or just one if it's a DEFINE group. */
+ two branches in the group, or just one if it's a DEFINE group. We do this
+ in the real compile phase, not in the pre-pass, where the whole group may
+ not be available. */
- if (bravalue == OP_COND)
+ if (bravalue == OP_COND && lengthptr == NULL)
{
uschar *tc = code;
int condcount = 0;
zerofirstbyte = firstbyte;
zeroreqbyte = reqbyte;
- /* \k<name> or \k'name' is a back reference by name (Perl syntax) */
+ /* \k<name> or \k'name' is a back reference by name (Perl syntax).
+ We also support \k{name} (.NET syntax) */
- if (-c == ESC_k && (ptr[1] == '<' || ptr[1] == '\''))
+ if (-c == ESC_k && (ptr[1] == '<' || ptr[1] == '\'' || ptr[1] == '{'))
{
is_recurse = FALSE;
- terminator = (*(++ptr) == '<')? '>' : '\'';
+ terminator = (*(++ptr) == '<')? '>' : (*ptr == '\'')? '\'' : '}';
goto NAMED_REF_OR_RECURSE;
}
out the amount of memory needed, as well as during the real compile phase. The
value of lengthptr distinguishes the two phases.
-Argument:
+Arguments:
options option bits, including any changes for this subpattern
oldims previous settings of ims option bits
codeptr -> the address of the current code pointer
ptrptr -> the address of the current pattern pointer
errorcodeptr -> pointer to error code variable
lookbehind TRUE if this is a lookbehind assertion
+ reset_bracount TRUE to reset the count for each branch
skipbytes skip this many bytes at start (for brackets and OP_COND)
firstbyteptr place to put the first required character, or a negative number
reqbyteptr place to put the last required character, or a negative number
static BOOL
compile_regex(int options, int oldims, uschar **codeptr, const uschar **ptrptr,
- int *errorcodeptr, BOOL lookbehind, int skipbytes, int *firstbyteptr,
- int *reqbyteptr, branch_chain *bcptr, compile_data *cd, int *lengthptr)
+ int *errorcodeptr, BOOL lookbehind, BOOL reset_bracount, int skipbytes,
+ int *firstbyteptr, int *reqbyteptr, branch_chain *bcptr, compile_data *cd,
+ int *lengthptr)
{
const uschar *ptr = *ptrptr;
uschar *code = *codeptr;
int firstbyte, reqbyte;
int branchfirstbyte, branchreqbyte;
int length;
+int orig_bracount;
+int max_bracount;
branch_chain bc;
bc.outer = bcptr;
/* Loop for each alternative branch */
+orig_bracount = max_bracount = cd->bracount;
for (;;)
{
+ /* For a (?| group, reset the capturing bracket count so that each branch
+ uses the same numbers. */
+
+ if (reset_bracount) cd->bracount = orig_bracount;
+
/* Handle a change of ims options at the start of the branch */
if ((options & PCRE_IMS) != oldims)
return FALSE;
}
+ /* Keep the highest bracket count in case (?| was used and some branch
+ has fewer than the rest. */
+
+ if (cd->bracount > max_bracount) max_bracount = cd->bracount;
+
/* In the real compile phase, there is some post-processing to be done. */
if (lengthptr == NULL)
}
}
- /* Reached end of expression, either ')' or end of pattern. Go back through
- the alternative branches and reverse the chain of offsets, with the field in
- the BRA item now becoming an offset to the first alternative. If there are
- no alternatives, it points to the end of the group. The length in the
- terminating ket is always the length of the whole bracketed item. If any of
- the ims options were changed inside the group, compile a resetting op-code
- following, except at the very end of the pattern. Return leaving the pointer
- at the terminating char. */
+ /* Reached end of expression, either ')' or end of pattern. In the real
+ compile phase, go back through the alternative branches and reverse the chain
+ of offsets, with the field in the BRA item now becoming an offset to the
+ first alternative. If there are no alternatives, it points to the end of the
+ group. The length in the terminating ket is always the length of the whole
+ bracketed item. If any of the ims options were changed inside the group,
+ compile a resetting op-code following, except at the very end of the pattern.
+ Return leaving the pointer at the terminating char. */
if (*ptr != '|')
{
- int branch_length = code - last_branch;
- do
+ if (lengthptr == NULL)
{
- int prev_length = GET(last_branch, 1);
- PUT(last_branch, 1, branch_length);
- branch_length = prev_length;
- last_branch -= branch_length;
+ int branch_length = code - last_branch;
+ do
+ {
+ int prev_length = GET(last_branch, 1);
+ PUT(last_branch, 1, branch_length);
+ branch_length = prev_length;
+ last_branch -= branch_length;
+ }
+ while (branch_length > 0);
}
- while (branch_length > 0);
/* Fill in the ket */
length += 2;
}
+ /* Retain the highest bracket number, in case resetting was used. */
+
+ cd->bracount = max_bracount;
+
/* Set values to pass back */
*codeptr = code;
return TRUE;
}
- /* Another branch follows; insert an "or" node. Its length field points back
+ /* Another branch follows. In the pre-compile phase, we can move the code
+ pointer back to where it was for the start of the first branch. (That is,
+ pretend that each branch is the only one.)
+
+ In the real compile phase, insert an ALT node. Its length field points back
to the previous branch while the bracket remains open. At the end the chain
is reversed. It's done like this so that the start of the bracket has a
zero offset until it is closed, making it possible to detect recursion. */
- *code = OP_ALT;
- PUT(code, 1, code - last_branch);
- bc.current = last_branch = code;
- code += 1 + LINK_SIZE;
+ if (lengthptr != NULL)
+ {
+ code = *codeptr + 1 + LINK_SIZE + skipbytes;
+ length += 1 + LINK_SIZE;
+ }
+ else
+ {
+ *code = OP_ALT;
+ PUT(code, 1, code - last_branch);
+ bc.current = last_branch = code;
+ code += 1 + LINK_SIZE;
+ }
+
ptr++;
- length += 1 + LINK_SIZE;
}
/* Control never reaches here */
}
with errorptr and erroroffset set
*/
-PCRE_DATA_SCOPE pcre *
+PCRE_EXP_DEFN pcre *
pcre_compile(const char *pattern, int options, const char **errorptr,
int *erroroffset, const unsigned char *tables)
{
}
-PCRE_DATA_SCOPE pcre *
+PCRE_EXP_DEFN pcre *
pcre_compile2(const char *pattern, int options, int *errorcodeptr,
const char **errorptr, int *erroroffset, const unsigned char *tables)
{
if (erroroffset == NULL)
{
errorcode = ERR16;
- goto PCRE_EARLY_ERROR_RETURN;
+ goto PCRE_EARLY_ERROR_RETURN2;
}
*erroroffset = 0;
(*erroroffset = _pcre_valid_utf8((uschar *)pattern, -1)) >= 0)
{
errorcode = ERR44;
- goto PCRE_UTF8_ERROR_RETURN;
+ goto PCRE_EARLY_ERROR_RETURN2;
}
#else
if ((options & PCRE_UTF8) != 0)
cd->ctypes = tables + ctypes_offset;
/* Handle different types of newline. The three bits give seven cases. The
-current code allows for fixed one- or two-byte sequences, plus "any". */
+current code allows for fixed one- or two-byte sequences, plus "any" and
+"anycrlf". */
switch (options & (PCRE_NEWLINE_CRLF | PCRE_NEWLINE_ANY))
{
case PCRE_NEWLINE_CR+
PCRE_NEWLINE_LF: newline = ('\r' << 8) | '\n'; break;
case PCRE_NEWLINE_ANY: newline = -1; break;
+ case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
default: errorcode = ERR56; goto PCRE_EARLY_ERROR_RETURN;
}
-if (newline < 0)
+if (newline == -2)
+ {
+ cd->nltype = NLTYPE_ANYCRLF;
+ }
+else if (newline < 0)
{
cd->nltype = NLTYPE_ANY;
}
code = cworkspace;
*code = OP_BRA;
(void)compile_regex(cd->external_options, cd->external_options & PCRE_IMS,
- &code, &ptr, &errorcode, FALSE, 0, &firstbyte, &reqbyte, NULL, cd, &length);
+ &code, &ptr, &errorcode, FALSE, FALSE, 0, &firstbyte, &reqbyte, NULL, cd,
+ &length);
if (errorcode != 0) goto PCRE_EARLY_ERROR_RETURN;
DPRINTF(("end pre-compile: length=%d workspace=%d\n", length,
code = (uschar *)codestart;
*code = OP_BRA;
(void)compile_regex(re->options, re->options & PCRE_IMS, &code, &ptr,
- &errorcode, FALSE, 0, &firstbyte, &reqbyte, NULL, cd, NULL);
+ &errorcode, FALSE, FALSE, 0, &firstbyte, &reqbyte, NULL, cd, NULL);
re->top_bracket = cd->bracount;
re->top_backref = cd->top_backref;
(pcre_free)(re);
PCRE_EARLY_ERROR_RETURN:
*erroroffset = ptr - (const uschar *)pattern;
-#ifdef SUPPORT_UTF8
- PCRE_UTF8_ERROR_RETURN:
-#endif
+ PCRE_EARLY_ERROR_RETURN2:
*errorptr = error_texts + error_texts_offsets[errorcode];
if (errorcodeptr != NULL) *errorcodeptr = errorcode;
return NULL;
else printf("Req char = \\x%02x%s\n", ch, caseless);
}
-pcre_printint(re, stdout);
+pcre_printint(re, stdout, TRUE);
/* This check is done here in the debugging case so that the code that
was compiled can be seen. */
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
Returns: 0 if data returned, negative on error
*/
-PCRE_DATA_SCOPE int
+PCRE_EXP_DEFN int
pcre_config(int what, void *where)
{
switch (what)
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
/* These are offsets that are used to turn the OP_TYPESTAR and friends opcodes
into others, under special conditions. A gap of 20 between the blocks should be
-enough. */
+enough. The resulting opcodes don't have to be less than 256 because they are
+never stored, so we push them well clear of the normal opcodes. */
-#define OP_PROP_EXTRA 100
-#define OP_EXTUNI_EXTRA 120
-#define OP_ANYNL_EXTRA 140
+#define OP_PROP_EXTRA 300
+#define OP_EXTUNI_EXTRA 320
+#define OP_ANYNL_EXTRA 340
+#define OP_HSPACE_EXTRA 360
+#define OP_VSPACE_EXTRA 380
/* This table identifies those opcodes that are followed immediately by a
character that is to be tested in some way. This makes is possible to
centralize the loading of these characters. In the case of Type * etc, the
"character" is the opcode for \D, \d, \S, \s, \W, or \w, which will always be a
-small value. */
+small value. ***NOTE*** If the start of this table is modified, the two tables
+that follow must also be modified. */
static uschar coptable[] = {
0, /* End */
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* \A, \G, \B, \b, \D, \d, \S, \s, \W, \w */
+ 0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */
+ 0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */
0, 0, /* Any, Anybyte */
- 0, 0, 0, 0, /* NOTPROP, PROP, EXTUNI, ANYNL */
+ 0, 0, 0, /* NOTPROP, PROP, EXTUNI */
+ 0, 0, 0, 0, 0, /* \R, \H, \h, \V, \v */
0, 0, 0, 0, 0, /* \Z, \z, Opt, ^, $ */
1, /* Char */
1, /* Charnc */
and \w */
static uschar toptable1[] = {
- 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0,
ctype_digit, ctype_digit,
ctype_space, ctype_space,
ctype_word, ctype_word,
};
static uschar toptable2[] = {
- 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0,
ctype_digit, 0,
ctype_space, 0,
ctype_word, 0,
const uschar *code;
int state_offset = current_state->offset;
int count, codevalue;
+#ifdef SUPPORT_UCP
int chartype, script;
+#endif
#ifdef DEBUG
printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset);
permitted.
We also use this mechanism for opcodes such as OP_TYPEPLUS that take an
- argument that is not a data character - but is always one byte long.
- Unfortunately, we have to take special action to deal with \P, \p, and
- \X in this case. To keep the other cases fast, convert these ones to new
- opcodes. */
+ argument that is not a data character - but is always one byte long. We
+ have to take special action to deal with \P, \p, \H, \h, \V, \v and \X in
+ this case. To keep the other cases fast, convert these ones to new opcodes.
+ */
if (coptable[codevalue] > 0)
{
case OP_PROP: codevalue += OP_PROP_EXTRA; break;
case OP_ANYNL: codevalue += OP_ANYNL_EXTRA; break;
case OP_EXTUNI: codevalue += OP_EXTUNI_EXTRA; break;
+ case OP_NOT_HSPACE:
+ case OP_HSPACE: codevalue += OP_HSPACE_EXTRA; break;
+ case OP_NOT_VSPACE:
+ case OP_VSPACE: codevalue += OP_VSPACE_EXTRA; break;
default: break;
}
}
break;
-#ifdef SUPPORT_UCP
-
/*-----------------------------------------------------------------*/
/* Check the next character by Unicode property. We will get here only
if the support is in the binary; otherwise a compile-time error occurs.
*/
+#ifdef SUPPORT_UCP
case OP_PROP:
case OP_NOTPROP:
if (clen > 0)
argument. It keeps the code above fast for the other cases. The argument
is in the d variable. */
+#ifdef SUPPORT_UCP
case OP_PROP_EXTRA + OP_TYPEPLUS:
case OP_PROP_EXTRA + OP_TYPEMINPLUS:
case OP_PROP_EXTRA + OP_TYPEPOSPLUS:
ADD_NEW_DATA(-state_offset, count, ncount);
}
break;
+#endif
/*-----------------------------------------------------------------*/
case OP_ANYNL_EXTRA + OP_TYPEPLUS:
break;
/*-----------------------------------------------------------------*/
+ case OP_VSPACE_EXTRA + OP_TYPEPLUS:
+ case OP_VSPACE_EXTRA + OP_TYPEMINPLUS:
+ case OP_VSPACE_EXTRA + OP_TYPEPOSPLUS:
+ count = current_state->count; /* Already matched */
+ if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }
+ if (clen > 0)
+ {
+ BOOL OK;
+ switch (c)
+ {
+ case 0x000a:
+ case 0x000b:
+ case 0x000c:
+ case 0x000d:
+ case 0x0085:
+ case 0x2028:
+ case 0x2029:
+ OK = TRUE;
+ break;
+
+ default:
+ OK = FALSE;
+ break;
+ }
+
+ if (OK == (d == OP_VSPACE))
+ {
+ if (count > 0 && codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSPLUS)
+ {
+ active_count--; /* Remove non-match possibility */
+ next_active_state--;
+ }
+ count++;
+ ADD_NEW_DATA(-state_offset, count, 0);
+ }
+ }
+ break;
+
+ /*-----------------------------------------------------------------*/
+ case OP_HSPACE_EXTRA + OP_TYPEPLUS:
+ case OP_HSPACE_EXTRA + OP_TYPEMINPLUS:
+ case OP_HSPACE_EXTRA + OP_TYPEPOSPLUS:
+ count = current_state->count; /* Already matched */
+ if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }
+ if (clen > 0)
+ {
+ BOOL OK;
+ switch (c)
+ {
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ OK = TRUE;
+ break;
+
+ default:
+ OK = FALSE;
+ break;
+ }
+
+ if (OK == (d == OP_HSPACE))
+ {
+ if (count > 0 && codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSPLUS)
+ {
+ active_count--; /* Remove non-match possibility */
+ next_active_state--;
+ }
+ count++;
+ ADD_NEW_DATA(-state_offset, count, 0);
+ }
+ }
+ break;
+
+ /*-----------------------------------------------------------------*/
+#ifdef SUPPORT_UCP
case OP_PROP_EXTRA + OP_TYPEQUERY:
case OP_PROP_EXTRA + OP_TYPEMINQUERY:
case OP_PROP_EXTRA + OP_TYPEPOSQUERY:
ADD_NEW_DATA(-(state_offset + count), 0, ncount);
}
break;
+#endif
/*-----------------------------------------------------------------*/
case OP_ANYNL_EXTRA + OP_TYPEQUERY:
break;
/*-----------------------------------------------------------------*/
+ case OP_VSPACE_EXTRA + OP_TYPEQUERY:
+ case OP_VSPACE_EXTRA + OP_TYPEMINQUERY:
+ case OP_VSPACE_EXTRA + OP_TYPEPOSQUERY:
+ count = 2;
+ goto QS4;
+
+ case OP_VSPACE_EXTRA + OP_TYPESTAR:
+ case OP_VSPACE_EXTRA + OP_TYPEMINSTAR:
+ case OP_VSPACE_EXTRA + OP_TYPEPOSSTAR:
+ count = 0;
+
+ QS4:
+ ADD_ACTIVE(state_offset + 2, 0);
+ if (clen > 0)
+ {
+ BOOL OK;
+ switch (c)
+ {
+ case 0x000a:
+ case 0x000b:
+ case 0x000c:
+ case 0x000d:
+ case 0x0085:
+ case 0x2028:
+ case 0x2029:
+ OK = TRUE;
+ break;
+
+ default:
+ OK = FALSE;
+ break;
+ }
+ if (OK == (d == OP_VSPACE))
+ {
+ if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSSTAR ||
+ codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSQUERY)
+ {
+ active_count--; /* Remove non-match possibility */
+ next_active_state--;
+ }
+ ADD_NEW_DATA(-(state_offset + count), 0, 0);
+ }
+ }
+ break;
+
+ /*-----------------------------------------------------------------*/
+ case OP_HSPACE_EXTRA + OP_TYPEQUERY:
+ case OP_HSPACE_EXTRA + OP_TYPEMINQUERY:
+ case OP_HSPACE_EXTRA + OP_TYPEPOSQUERY:
+ count = 2;
+ goto QS5;
+
+ case OP_HSPACE_EXTRA + OP_TYPESTAR:
+ case OP_HSPACE_EXTRA + OP_TYPEMINSTAR:
+ case OP_HSPACE_EXTRA + OP_TYPEPOSSTAR:
+ count = 0;
+
+ QS5:
+ ADD_ACTIVE(state_offset + 2, 0);
+ if (clen > 0)
+ {
+ BOOL OK;
+ switch (c)
+ {
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ OK = TRUE;
+ break;
+
+ default:
+ OK = FALSE;
+ break;
+ }
+
+ if (OK == (d == OP_HSPACE))
+ {
+ if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSSTAR ||
+ codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSQUERY)
+ {
+ active_count--; /* Remove non-match possibility */
+ next_active_state--;
+ }
+ ADD_NEW_DATA(-(state_offset + count), 0, 0);
+ }
+ }
+ break;
+
+ /*-----------------------------------------------------------------*/
+#ifdef SUPPORT_UCP
case OP_PROP_EXTRA + OP_TYPEEXACT:
case OP_PROP_EXTRA + OP_TYPEUPTO:
case OP_PROP_EXTRA + OP_TYPEMINUPTO:
{ ADD_NEW_DATA(-state_offset, count, ncount); }
}
break;
+#endif
/*-----------------------------------------------------------------*/
case OP_ANYNL_EXTRA + OP_TYPEEXACT:
}
break;
+ /*-----------------------------------------------------------------*/
+ case OP_VSPACE_EXTRA + OP_TYPEEXACT:
+ case OP_VSPACE_EXTRA + OP_TYPEUPTO:
+ case OP_VSPACE_EXTRA + OP_TYPEMINUPTO:
+ case OP_VSPACE_EXTRA + OP_TYPEPOSUPTO:
+ if (codevalue != OP_VSPACE_EXTRA + OP_TYPEEXACT)
+ { ADD_ACTIVE(state_offset + 4, 0); }
+ count = current_state->count; /* Number already matched */
+ if (clen > 0)
+ {
+ BOOL OK;
+ switch (c)
+ {
+ case 0x000a:
+ case 0x000b:
+ case 0x000c:
+ case 0x000d:
+ case 0x0085:
+ case 0x2028:
+ case 0x2029:
+ OK = TRUE;
+ break;
+
+ default:
+ OK = FALSE;
+ }
+
+ if (OK == (d == OP_VSPACE))
+ {
+ if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSUPTO)
+ {
+ active_count--; /* Remove non-match possibility */
+ next_active_state--;
+ }
+ if (++count >= GET2(code, 1))
+ { ADD_NEW_DATA(-(state_offset + 4), 0, 0); }
+ else
+ { ADD_NEW_DATA(-state_offset, count, 0); }
+ }
+ }
+ break;
+
+ /*-----------------------------------------------------------------*/
+ case OP_HSPACE_EXTRA + OP_TYPEEXACT:
+ case OP_HSPACE_EXTRA + OP_TYPEUPTO:
+ case OP_HSPACE_EXTRA + OP_TYPEMINUPTO:
+ case OP_HSPACE_EXTRA + OP_TYPEPOSUPTO:
+ if (codevalue != OP_HSPACE_EXTRA + OP_TYPEEXACT)
+ { ADD_ACTIVE(state_offset + 4, 0); }
+ count = current_state->count; /* Number already matched */
+ if (clen > 0)
+ {
+ BOOL OK;
+ switch (c)
+ {
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ OK = TRUE;
+ break;
+
+ default:
+ OK = FALSE;
+ break;
+ }
+
+ if (OK == (d == OP_HSPACE))
+ {
+ if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSUPTO)
+ {
+ active_count--; /* Remove non-match possibility */
+ next_active_state--;
+ }
+ if (++count >= GET2(code, 1))
+ { ADD_NEW_DATA(-(state_offset + 4), 0, 0); }
+ else
+ { ADD_NEW_DATA(-state_offset, count, 0); }
+ }
+ }
+ break;
+
/* ========================================================================== */
/* These opcodes are followed by a character that is usually compared
to the current subject character; it is loaded into d. We still get
break;
/*-----------------------------------------------------------------*/
+ case OP_NOT_VSPACE:
+ if (clen > 0) switch(c)
+ {
+ case 0x000a:
+ case 0x000b:
+ case 0x000c:
+ case 0x000d:
+ case 0x0085:
+ case 0x2028:
+ case 0x2029:
+ break;
+
+ default:
+ ADD_NEW(state_offset + 1, 0);
+ break;
+ }
+ break;
+
+ /*-----------------------------------------------------------------*/
+ case OP_VSPACE:
+ if (clen > 0) switch(c)
+ {
+ case 0x000a:
+ case 0x000b:
+ case 0x000c:
+ case 0x000d:
+ case 0x0085:
+ case 0x2028:
+ case 0x2029:
+ ADD_NEW(state_offset + 1, 0);
+ break;
+
+ default: break;
+ }
+ break;
+
+ /*-----------------------------------------------------------------*/
+ case OP_NOT_HSPACE:
+ if (clen > 0) switch(c)
+ {
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ break;
+
+ default:
+ ADD_NEW(state_offset + 1, 0);
+ break;
+ }
+ break;
+
+ /*-----------------------------------------------------------------*/
+ case OP_HSPACE:
+ if (clen > 0) switch(c)
+ {
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ ADD_NEW(state_offset + 1, 0);
+ break;
+ }
+ break;
+
+ /*-----------------------------------------------------------------*/
/* Match a negated single character. This is only used for one-byte
characters, that is, we know that d < 256. The character we are
checking (c) can be multibyte. */
Arguments:
argument_re points to the compiled expression
- extra_data points to extra data or is NULL (not currently used)
+ extra_data points to extra data or is NULL
subject points to the subject string
length length of subject string (may contain binary zeros)
start_offset where to start in the subject string
< -1 => some kind of unexpected problem
*/
-PCRE_DATA_SCOPE int
+PCRE_EXP_DEFN int
pcre_dfa_exec(const pcre *argument_re, const pcre_extra *extra_data,
const char *subject, int length, int start_offset, int options, int *offsets,
int offsetcount, int *workspace, int wscount)
md->moptions = options;
md->poptions = re->options;
-/* Handle different types of newline. The two bits give four cases. If nothing
-is set at run time, whatever was used at compile time applies. */
+/* Handle different types of newline. The three bits give eight cases. If
+nothing is set at run time, whatever was used at compile time applies. */
-switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : options) &
+switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &
PCRE_NEWLINE_BITS)
{
case 0: newline = NEWLINE; break; /* Compile-time default */
case PCRE_NEWLINE_CR+
PCRE_NEWLINE_LF: newline = ('\r' << 8) | '\n'; break;
case PCRE_NEWLINE_ANY: newline = -1; break;
+ case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
default: return PCRE_ERROR_BADNEWLINE;
}
-if (newline < 0)
+if (newline == -2)
+ {
+ md->nltype = NLTYPE_ANYCRLF;
+ }
+else if (newline < 0)
{
md->nltype = NLTYPE_ANY;
}
{
while (current_subject <= end_subject && !WAS_NEWLINE(current_subject))
current_subject++;
+
+ /* If we have just passed a CR and the newline option is ANY or
+ ANYCRLF, and we are now at a LF, advance the match position by one more
+ character. */
+
+ if (current_subject[-1] == '\r' &&
+ (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
+ current_subject < end_subject &&
+ *current_subject == '\n')
+ current_subject++;
}
}
}
if (current_subject > end_subject) break;
- /* If we have just passed a CR and the newline option is CRLF or ANY, and we
- are now at a LF, advance the match position by one more character. */
+ /* If we have just passed a CR and the newline option is CRLF or ANY or
+ ANYCRLF, and we are now at a LF, advance the match position by one more
+ character. */
if (current_subject[-1] == '\r' &&
- (md->nltype == NLTYPE_ANY || md->nllen == 2) &&
+ (md->nltype == NLTYPE_ANY ||
+ md->nltype == NLTYPE_ANYCRLF ||
+ md->nllen == 2) &&
current_subject < end_subject &&
*current_subject == '\n')
current_subject++;
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
#include "pcre_internal.h"
+/* Undefine some potentially clashing cpp symbols */
+
+#undef min
+#undef max
+
/* The chain of eptrblocks for tail recursions uses memory in stack workspace,
obtained at top level, the size of which is defined by EPTR_WORK_SIZE. */
obtained from malloc() instead instead of on the stack. Macros are used to
achieve this so that the actual code doesn't look very different to what it
always used to.
+
+The original heap-recursive code used longjmp(). However, it seems that this
+can be very slow on some operating systems. Following a suggestion from Stan
+Switzer, the use of longjmp() has been abolished, at the cost of having to
+provide a unique number for each call to RMATCH. There is no way of generating
+a sequence of numbers at compile time in C. I have given them names, to make
+them stand out more clearly.
+
+Crude tests on x86 Linux show a small speedup of around 5-8%. However, on
+FreeBSD, avoiding longjmp() more than halves the time taken to run the standard
+tests. Furthermore, not using longjmp() means that local dynamic variables
+don't have indeterminate values; this has meant that the frame size can be
+reduced because the result can be "passed back" by straight setting of the
+variable instead of being passed in the frame.
****************************************************************************
***************************************************************************/
+/* Numbers for RMATCH calls */
+
+enum { RM1=1, RM2, RM3, RM4, RM5, RM6, RM7, RM8, RM9, RM10,
+ RM11, RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,
+ RM21, RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,
+ RM31, RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
+ RM41, RM42, RM43, RM44, RM45, RM46, RM47 };
+
+
/* These versions of the macros use the stack, as normal. There are debugging
-versions and production versions. */
+versions and production versions. Note that the "rw" argument of RMATCH isn't
+actuall used in this definition. */
#ifndef NO_RECURSE
#define REGISTER register
+
#ifdef DEBUG
-#define RMATCH(rx,ra,rb,rc,rd,re,rf,rg) \
+#define RMATCH(ra,rb,rc,rd,re,rf,rg,rw) \
{ \
printf("match() called in line %d\n", __LINE__); \
- rx = match(ra,rb,rc,rd,re,rf,rg,rdepth+1); \
+ rrc = match(ra,rb,mstart,rc,rd,re,rf,rg,rdepth+1); \
printf("to line %d\n", __LINE__); \
}
#define RRETURN(ra) \
return ra; \
}
#else
-#define RMATCH(rx,ra,rb,rc,rd,re,rf,rg) \
- rx = match(ra,rb,rc,rd,re,rf,rg,rdepth+1)
+#define RMATCH(ra,rb,rc,rd,re,rf,rg,rw) \
+ rrc = match(ra,rb,mstart,rc,rd,re,rf,rg,rdepth+1)
#define RRETURN(ra) return ra
#endif
#else
-/* These versions of the macros manage a private stack on the heap. Note
-that the rd argument of RMATCH isn't actually used. It's the md argument of
-match(), which never changes. */
+/* These versions of the macros manage a private stack on the heap. Note that
+the "rd" argument of RMATCH isn't actually used in this definition. It's the md
+argument of match(), which never changes. */
#define REGISTER
-#define RMATCH(rx,ra,rb,rc,rd,re,rf,rg)\
+#define RMATCH(ra,rb,rc,rd,re,rf,rg,rw)\
{\
heapframe *newframe = (pcre_stack_malloc)(sizeof(heapframe));\
- if (setjmp(frame->Xwhere) == 0)\
- {\
- newframe->Xeptr = ra;\
- newframe->Xecode = rb;\
- newframe->Xoffset_top = rc;\
- newframe->Xims = re;\
- newframe->Xeptrb = rf;\
- newframe->Xflags = rg;\
- newframe->Xrdepth = frame->Xrdepth + 1;\
- newframe->Xprevframe = frame;\
- frame = newframe;\
- DPRINTF(("restarting from line %d\n", __LINE__));\
- goto HEAP_RECURSE;\
- }\
- else\
- {\
- DPRINTF(("longjumped back to line %d\n", __LINE__));\
- frame = md->thisframe;\
- rx = frame->Xresult;\
- }\
+ frame->Xwhere = rw; \
+ newframe->Xeptr = ra;\
+ newframe->Xecode = rb;\
+ newframe->Xmstart = mstart;\
+ newframe->Xoffset_top = rc;\
+ newframe->Xims = re;\
+ newframe->Xeptrb = rf;\
+ newframe->Xflags = rg;\
+ newframe->Xrdepth = frame->Xrdepth + 1;\
+ newframe->Xprevframe = frame;\
+ frame = newframe;\
+ DPRINTF(("restarting from line %d\n", __LINE__));\
+ goto HEAP_RECURSE;\
+ L_##rw:\
+ DPRINTF(("jumped back to line %d\n", __LINE__));\
}
#define RRETURN(ra)\
(pcre_stack_free)(newframe);\
if (frame != NULL)\
{\
- frame->Xresult = ra;\
- md->thisframe = frame;\
- longjmp(frame->Xwhere, 1);\
+ rrc = ra;\
+ goto HEAP_RETURN;\
}\
return ra;\
}
const uschar *Xeptr;
const uschar *Xecode;
+ const uschar *Xmstart;
int Xoffset_top;
long int Xims;
eptrblock *Xeptrb;
int Xprop_category;
int Xprop_chartype;
int Xprop_script;
+ int Xoclength;
+ uschar Xocchars[8];
#endif
int Xctype;
eptrblock Xnewptrb;
- /* Place to pass back result, and where to jump back to */
+ /* Where to jump back to */
- int Xresult;
- jmp_buf Xwhere;
+ int Xwhere;
} heapframe;
Arguments:
eptr pointer to current character in subject
ecode pointer to current position in compiled code
+ mstart pointer to the current match start position (can be modified
+ by encountering \K)
offset_top current top pointer
md pointer to "static" info for the match
ims current /i, /m, and /s options
*/
static int
-match(REGISTER USPTR eptr, REGISTER const uschar *ecode,
+match(REGISTER USPTR eptr, REGISTER const uschar *ecode, const uschar *mstart,
int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb,
int flags, unsigned int rdepth)
{
frame->Xeptr = eptr;
frame->Xecode = ecode;
+frame->Xmstart = mstart;
frame->Xoffset_top = offset_top;
frame->Xims = ims;
frame->Xeptrb = eptrb;
#define eptr frame->Xeptr
#define ecode frame->Xecode
+#define mstart frame->Xmstart
#define offset_top frame->Xoffset_top
#define ims frame->Xims
#define eptrb frame->Xeptrb
#define prop_category frame->Xprop_category
#define prop_chartype frame->Xprop_chartype
#define prop_script frame->Xprop_script
+#define oclength frame->Xoclength
+#define occhars frame->Xocchars
#endif
#define ctype frame->Xctype
int prop_category;
int prop_chartype;
int prop_script;
+int oclength;
+uschar occhars[8];
#endif
int ctype;
complicated macro. It has to be used in one particular way. This shouldn't,
however, impact performance when true recursion is being used. */
+#ifdef SUPPORT_UTF8
+utf8 = md->utf8; /* Local copy of the flag */
+#else
+utf8 = FALSE;
+#endif
+
/* First check that we haven't called match() too many times, or that we
haven't exceeded the recursive call limit. */
original_ims = ims; /* Save for resetting on ')' */
-#ifdef SUPPORT_UTF8
-utf8 = md->utf8; /* Local copy of the flag */
-#else
-utf8 = FALSE;
-#endif
-
/* At the start of a group with an unlimited repeat that may match an empty
string, the match_cbegroup flag is set. When this is the case, add the current
subject pointer to the chain of such remembered pointers, to be checked when we
if (md->partial &&
eptr >= md->end_subject &&
- eptr > md->start_match)
+ eptr > mstart)
md->hitend = TRUE;
switch(op)
flags = (op == OP_SCBRA)? match_cbegroup : 0;
do
{
- RMATCH(rrc, eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
- ims, eptrb, flags);
+ RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
+ ims, eptrb, flags, RM1);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
md->capture_last = save_capture_last;
ecode += GET(ecode, 1);
/* For non-final alternatives, continue the loop for a NOMATCH result;
otherwise return. */
- RMATCH(rrc, eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
- eptrb, flags);
+ RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
+ eptrb, flags, RM2);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
ecode += GET(ecode, 1);
}
else
{
- RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL,
- match_condassert);
+ RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL,
+ match_condassert, RM3);
if (rrc == MATCH_MATCH)
{
condition = TRUE;
md->recursive = rec->prevrec;
memmove(md->offset_vector, rec->offset_save,
rec->saved_max * sizeof(int));
- md->start_match = rec->save_start;
+ mstart = rec->save_start;
ims = original_ims;
ecode = rec->after_call;
break;
/* Otherwise, if PCRE_NOTEMPTY is set, fail if we have matched an empty
string - backtracking will then try other alternatives, if any. */
- if (md->notempty && eptr == md->start_match) RRETURN(MATCH_NOMATCH);
- md->end_match_ptr = eptr; /* Record where we ended */
- md->end_offset_top = offset_top; /* and how many extracts were taken */
+ if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);
+ md->end_match_ptr = eptr; /* Record where we ended */
+ md->end_offset_top = offset_top; /* and how many extracts were taken */
+ md->start_match_ptr = mstart; /* and the start (\K can modify) */
RRETURN(MATCH_MATCH);
/* Change option settings */
case OP_ASSERTBACK:
do
{
- RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0);
+ RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
+ RM4);
if (rrc == MATCH_MATCH) break;
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
ecode += GET(ecode, 1);
case OP_ASSERTBACK_NOT:
do
{
- RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0);
+ RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
+ RM5);
if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
ecode += GET(ecode,1);
cb.offset_vector = md->offset_vector;
cb.subject = (PCRE_SPTR)md->start_subject;
cb.subject_length = md->end_subject - md->start_subject;
- cb.start_match = md->start_match - md->start_subject;
+ cb.start_match = mstart - md->start_subject;
cb.current_position = eptr - md->start_subject;
cb.pattern_position = GET(ecode, 2);
cb.next_item_length = GET(ecode, 2 + LINK_SIZE);
memcpy(new_recursive.offset_save, md->offset_vector,
new_recursive.saved_max * sizeof(int));
- new_recursive.save_start = md->start_match;
- md->start_match = eptr;
+ new_recursive.save_start = mstart;
+ mstart = eptr;
/* OK, now we can do the recursion. For each top-level alternative we
restore the offset and recursion data. */
flags = (*callpat >= OP_SBRA)? match_cbegroup : 0;
do
{
- RMATCH(rrc, eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,
- md, ims, eptrb, flags);
+ RMATCH(eptr, callpat + _pcre_OP_lengths[*callpat], offset_top,
+ md, ims, eptrb, flags, RM6);
if (rrc == MATCH_MATCH)
{
DPRINTF(("Recursion matched\n"));
do
{
- RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims,
- eptrb, 0);
+ RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims,
+ eptrb, 0, RM7);
if (rrc == MATCH_MATCH) break;
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
ecode += GET(ecode,1);
if (*ecode == OP_KETRMIN)
{
- RMATCH(rrc, eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0,
+ RM8);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
ecode = prev;
flags = match_tail_recursed;
}
else /* OP_KETRMAX */
{
- RMATCH(rrc, eptr, prev, offset_top, md, ims, eptrb, match_cbegroup);
+ RMATCH(eptr, prev, offset_top, md, ims, eptrb, match_cbegroup, RM9);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
ecode += 1 + LINK_SIZE;
flags = match_tail_recursed;
case OP_BRAZERO:
{
next = ecode+1;
- RMATCH(rrc, eptr, next, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, next, offset_top, md, ims, eptrb, 0, RM10);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
do next += GET(next,1); while (*next == OP_ALT);
ecode = next + 1 + LINK_SIZE;
{
next = ecode+1;
do next += GET(next, 1); while (*next == OP_ALT);
- RMATCH(rrc, eptr, next + 1+LINK_SIZE, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, next + 1+LINK_SIZE, offset_top, md, ims, eptrb, 0, RM11);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
ecode++;
}
recursion_info *rec = md->recursive;
DPRINTF(("Recursion (%d) succeeded - continuing\n", number));
md->recursive = rec->prevrec;
- md->start_match = rec->save_start;
+ mstart = rec->save_start;
memcpy(md->offset_vector, rec->offset_save,
rec->saved_max * sizeof(int));
ecode = rec->after_call;
if (*ecode == OP_KETRMIN)
{
- RMATCH(rrc, eptr, ecode + 1+LINK_SIZE, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode + 1+LINK_SIZE, offset_top, md, ims, eptrb, 0,
+ RM12);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
ecode = prev;
flags |= match_tail_recursed;
}
else /* OP_KETRMAX */
{
- RMATCH(rrc, eptr, prev, offset_top, md, ims, eptrb, flags);
+ RMATCH(eptr, prev, offset_top, md, ims, eptrb, flags, RM13);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
ecode += 1 + LINK_SIZE;
flags = match_tail_recursed;
ecode++;
break;
+ /* Reset the start of match point */
+
+ case OP_SET_SOM:
+ mstart = eptr;
+ ecode++;
+ break;
+
/* Assert before internal newline if multiline, or before a terminating
newline unless endonly is set, else end of subject unless noteol is set. */
ecode++;
break;
+ case OP_NOT_HSPACE:
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ GETCHARINCTEST(c, eptr);
+ switch(c)
+ {
+ default: break;
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ RRETURN(MATCH_NOMATCH);
+ }
+ ecode++;
+ break;
+
+ case OP_HSPACE:
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ GETCHARINCTEST(c, eptr);
+ switch(c)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ break;
+ }
+ ecode++;
+ break;
+
+ case OP_NOT_VSPACE:
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ GETCHARINCTEST(c, eptr);
+ switch(c)
+ {
+ default: break;
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ case 0x2028: /* LINE SEPARATOR */
+ case 0x2029: /* PARAGRAPH SEPARATOR */
+ RRETURN(MATCH_NOMATCH);
+ }
+ ecode++;
+ break;
+
+ case OP_VSPACE:
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ GETCHARINCTEST(c, eptr);
+ switch(c)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ case 0x2028: /* LINE SEPARATOR */
+ case 0x2029: /* PARAGRAPH SEPARATOR */
+ break;
+ }
+ ecode++;
+ break;
+
#ifdef SUPPORT_UCP
/* Check the next character by Unicode property. We will get here only
if the support is in the binary; otherwise a compile-time error occurs. */
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || !match_ref(offset, eptr, length, md, ims))
RRETURN(MATCH_NOMATCH);
}
while (eptr >= pp)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM15);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
eptr -= length;
}
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
GETCHARINC(c, eptr);
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
c = *eptr++;
}
for (;;)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM18);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (eptr-- == pp) break; /* Stop if tried at original pos */
BACKCHAR(eptr);
}
while (eptr >= pp)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM19);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
eptr--;
}
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
GETCHARINC(c, eptr);
}
for(;;)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (eptr-- == pp) break; /* Stop if tried at original pos */
BACKCHAR(eptr)
if (length > 1)
{
- int oclength = 0;
- uschar occhars[8];
-
#ifdef SUPPORT_UCP
unsigned int othercase;
if ((ims & PCRE_CASELESS) != 0 &&
(othercase = _pcre_ucp_othercase(fc)) != NOTACHAR)
oclength = _pcre_ord2utf8(othercase, occhars);
+ else oclength = 0;
#endif /* SUPPORT_UCP */
for (i = 1; i <= min; i++)
{
if (memcmp(eptr, charptr, length) == 0) eptr += length;
+#ifdef SUPPORT_UCP
/* Need braces because of following else */
else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }
else
if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);
eptr += oclength;
}
+#else /* without SUPPORT_UCP */
+ else { RRETURN(MATCH_NOMATCH); }
+#endif /* SUPPORT_UCP */
}
if (min == max) continue;
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
if (memcmp(eptr, charptr, length) == 0) eptr += length;
+#ifdef SUPPORT_UCP
/* Need braces because of following else */
else if (oclength == 0) { RRETURN(MATCH_NOMATCH); }
else
if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH);
eptr += oclength;
}
+#else /* without SUPPORT_UCP */
+ else { RRETURN (MATCH_NOMATCH); }
+#endif /* SUPPORT_UCP */
}
/* Control never gets here */
}
{
if (eptr > md->end_subject - length) break;
if (memcmp(eptr, charptr, length) == 0) eptr += length;
+#ifdef SUPPORT_UCP
else if (oclength == 0) break;
else
{
if (memcmp(eptr, occhars, oclength) != 0) break;
eptr += oclength;
}
+#else /* without SUPPORT_UCP */
+ else break;
+#endif /* SUPPORT_UCP */
}
if (possessive) continue;
- while (eptr >= pp)
+ for(;;)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+ if (eptr == pp) RRETURN(MATCH_NOMATCH);
+#ifdef SUPPORT_UCP
+ eptr--;
+ BACKCHAR(eptr);
+#else /* without SUPPORT_UCP */
eptr -= length;
+#endif /* SUPPORT_UCP */
}
- RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
}
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject ||
fc != md->lcc[*eptr++])
if (possessive) continue;
while (eptr >= pp)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25);
eptr--;
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
}
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject || fc != *eptr++)
RRETURN(MATCH_NOMATCH);
if (possessive) continue;
while (eptr >= pp)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27);
eptr--;
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
}
register unsigned int d;
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
GETCHARINC(d, eptr);
if (d < 256) d = md->lcc[d];
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject || fc == md->lcc[*eptr++])
RRETURN(MATCH_NOMATCH);
if (possessive) continue;
for(;;)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM30);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (eptr-- == pp) break; /* Stop if tried at original pos */
BACKCHAR(eptr);
if (possessive) continue;
while (eptr >= pp)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM31);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
eptr--;
}
register unsigned int d;
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
GETCHARINC(d, eptr);
if (fi >= max || eptr >= md->end_subject || fc == d)
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject || fc == *eptr++)
RRETURN(MATCH_NOMATCH);
if (possessive) continue;
for(;;)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM34);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (eptr-- == pp) break; /* Stop if tried at original pos */
BACKCHAR(eptr);
if (possessive) continue;
while (eptr >= pp)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM35);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
eptr--;
}
for (i = 1; i <= min; i++)
{
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
- GETCHARINC(c, eptr);
+ GETCHARINCTEST(c, eptr);
}
break;
for (i = 1; i <= min; i++)
{
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
- GETCHARINC(c, eptr);
+ GETCHARINCTEST(c, eptr);
prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
if ((prop_chartype == ucp_Lu ||
prop_chartype == ucp_Ll ||
for (i = 1; i <= min; i++)
{
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
- GETCHARINC(c, eptr);
+ GETCHARINCTEST(c, eptr);
prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
if ((prop_category == prop_value) == prop_fail_result)
RRETURN(MATCH_NOMATCH);
for (i = 1; i <= min; i++)
{
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
- GETCHARINC(c, eptr);
+ GETCHARINCTEST(c, eptr);
prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
if ((prop_chartype == prop_value) == prop_fail_result)
RRETURN(MATCH_NOMATCH);
for (i = 1; i <= min; i++)
{
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
- GETCHARINC(c, eptr);
+ GETCHARINCTEST(c, eptr);
prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
if ((prop_script == prop_value) == prop_fail_result)
RRETURN(MATCH_NOMATCH);
}
break;
+ case OP_NOT_HSPACE:
+ for (i = 1; i <= min; i++)
+ {
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ GETCHARINC(c, eptr);
+ switch(c)
+ {
+ default: break;
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ RRETURN(MATCH_NOMATCH);
+ }
+ }
+ break;
+
+ case OP_HSPACE:
+ for (i = 1; i <= min; i++)
+ {
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ GETCHARINC(c, eptr);
+ switch(c)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ break;
+ }
+ }
+ break;
+
+ case OP_NOT_VSPACE:
+ for (i = 1; i <= min; i++)
+ {
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ GETCHARINC(c, eptr);
+ switch(c)
+ {
+ default: break;
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ case 0x2028: /* LINE SEPARATOR */
+ case 0x2029: /* PARAGRAPH SEPARATOR */
+ RRETURN(MATCH_NOMATCH);
+ }
+ }
+ break;
+
+ case OP_VSPACE:
+ for (i = 1; i <= min; i++)
+ {
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ GETCHARINC(c, eptr);
+ switch(c)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ case 0x2028: /* LINE SEPARATOR */
+ case 0x2029: /* PARAGRAPH SEPARATOR */
+ break;
+ }
+ }
+ break;
+
case OP_NOT_DIGIT:
for (i = 1; i <= min; i++)
{
}
break;
+ case OP_NOT_HSPACE:
+ for (i = 1; i <= min; i++)
+ {
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ switch(*eptr++)
+ {
+ default: break;
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ RRETURN(MATCH_NOMATCH);
+ }
+ }
+ break;
+
+ case OP_HSPACE:
+ for (i = 1; i <= min; i++)
+ {
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ switch(*eptr++)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ break;
+ }
+ }
+ break;
+
+ case OP_NOT_VSPACE:
+ for (i = 1; i <= min; i++)
+ {
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ switch(*eptr++)
+ {
+ default: break;
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ RRETURN(MATCH_NOMATCH);
+ }
+ }
+ break;
+
+ case OP_VSPACE:
+ for (i = 1; i <= min; i++)
+ {
+ if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
+ switch(*eptr++)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ break;
+ }
+ }
+ break;
+
case OP_NOT_DIGIT:
for (i = 1; i <= min; i++)
if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);
case PT_ANY:
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
GETCHARINC(c, eptr);
case PT_LAMP:
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
GETCHARINC(c, eptr);
case PT_GC:
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
GETCHARINC(c, eptr);
case PT_PC:
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
GETCHARINC(c, eptr);
case PT_SC:
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
GETCHARINC(c, eptr);
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
GETCHARINCTEST(c, eptr);
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject ||
(ctype == OP_ANY && (ims & PCRE_DOTALL) == 0 &&
}
break;
+ case OP_NOT_HSPACE:
+ switch(c)
+ {
+ default: break;
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ RRETURN(MATCH_NOMATCH);
+ }
+ break;
+
+ case OP_HSPACE:
+ switch(c)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ break;
+ }
+ break;
+
+ case OP_NOT_VSPACE:
+ switch(c)
+ {
+ default: break;
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ case 0x2028: /* LINE SEPARATOR */
+ case 0x2029: /* PARAGRAPH SEPARATOR */
+ RRETURN(MATCH_NOMATCH);
+ }
+ break;
+
+ case OP_VSPACE:
+ switch(c)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ case 0x2028: /* LINE SEPARATOR */
+ case 0x2029: /* PARAGRAPH SEPARATOR */
+ break;
+ }
+ break;
+
case OP_NOT_DIGIT:
if (c < 256 && (md->ctypes[c] & ctype_digit) != 0)
RRETURN(MATCH_NOMATCH);
{
for (fi = min;; fi++)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (fi >= max || eptr >= md->end_subject ||
((ims & PCRE_DOTALL) == 0 && IS_NEWLINE(eptr)))
}
break;
+ case OP_NOT_HSPACE:
+ switch(c)
+ {
+ default: break;
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ RRETURN(MATCH_NOMATCH);
+ }
+ break;
+
+ case OP_HSPACE:
+ switch(c)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ break;
+ }
+ break;
+
+ case OP_NOT_VSPACE:
+ switch(c)
+ {
+ default: break;
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ RRETURN(MATCH_NOMATCH);
+ }
+ break;
+
+ case OP_VSPACE:
+ switch(c)
+ {
+ default: RRETURN(MATCH_NOMATCH);
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ break;
+ }
+ break;
+
case OP_NOT_DIGIT:
if ((md->ctypes[c] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);
break;
if (possessive) continue;
for(;;)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (eptr-- == pp) break; /* Stop if tried at original pos */
BACKCHAR(eptr);
if (possessive) continue;
for(;;)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM45);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (eptr-- == pp) break; /* Stop if tried at original pos */
for (;;) /* Move back over one extended */
}
break;
+ case OP_NOT_HSPACE:
+ case OP_HSPACE:
+ for (i = min; i < max; i++)
+ {
+ BOOL gotspace;
+ int len = 1;
+ if (eptr >= md->end_subject) break;
+ GETCHARLEN(c, eptr, len);
+ switch(c)
+ {
+ default: gotspace = FALSE; break;
+ case 0x09: /* HT */
+ case 0x20: /* SPACE */
+ case 0xa0: /* NBSP */
+ case 0x1680: /* OGHAM SPACE MARK */
+ case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */
+ case 0x2000: /* EN QUAD */
+ case 0x2001: /* EM QUAD */
+ case 0x2002: /* EN SPACE */
+ case 0x2003: /* EM SPACE */
+ case 0x2004: /* THREE-PER-EM SPACE */
+ case 0x2005: /* FOUR-PER-EM SPACE */
+ case 0x2006: /* SIX-PER-EM SPACE */
+ case 0x2007: /* FIGURE SPACE */
+ case 0x2008: /* PUNCTUATION SPACE */
+ case 0x2009: /* THIN SPACE */
+ case 0x200A: /* HAIR SPACE */
+ case 0x202f: /* NARROW NO-BREAK SPACE */
+ case 0x205f: /* MEDIUM MATHEMATICAL SPACE */
+ case 0x3000: /* IDEOGRAPHIC SPACE */
+ gotspace = TRUE;
+ break;
+ }
+ if (gotspace == (ctype == OP_NOT_HSPACE)) break;
+ eptr += len;
+ }
+ break;
+
+ case OP_NOT_VSPACE:
+ case OP_VSPACE:
+ for (i = min; i < max; i++)
+ {
+ BOOL gotspace;
+ int len = 1;
+ if (eptr >= md->end_subject) break;
+ GETCHARLEN(c, eptr, len);
+ switch(c)
+ {
+ default: gotspace = FALSE; break;
+ case 0x0a: /* LF */
+ case 0x0b: /* VT */
+ case 0x0c: /* FF */
+ case 0x0d: /* CR */
+ case 0x85: /* NEL */
+ case 0x2028: /* LINE SEPARATOR */
+ case 0x2029: /* PARAGRAPH SEPARATOR */
+ gotspace = TRUE;
+ break;
+ }
+ if (gotspace == (ctype == OP_NOT_VSPACE)) break;
+ eptr += len;
+ }
+ break;
+
case OP_NOT_DIGIT:
for (i = min; i < max; i++)
{
if (possessive) continue;
for(;;)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM46);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (eptr-- == pp) break; /* Stop if tried at original pos */
BACKCHAR(eptr);
}
break;
+ case OP_NOT_HSPACE:
+ for (i = min; i < max; i++)
+ {
+ if (eptr >= md->end_subject) break;
+ c = *eptr;
+ if (c == 0x09 || c == 0x20 || c == 0xa0) break;
+ eptr++;
+ }
+ break;
+
+ case OP_HSPACE:
+ for (i = min; i < max; i++)
+ {
+ if (eptr >= md->end_subject) break;
+ c = *eptr;
+ if (c != 0x09 && c != 0x20 && c != 0xa0) break;
+ eptr++;
+ }
+ break;
+
+ case OP_NOT_VSPACE:
+ for (i = min; i < max; i++)
+ {
+ if (eptr >= md->end_subject) break;
+ c = *eptr;
+ if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85)
+ break;
+ eptr++;
+ }
+ break;
+
+ case OP_VSPACE:
+ for (i = min; i < max; i++)
+ {
+ if (eptr >= md->end_subject) break;
+ c = *eptr;
+ if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85)
+ break;
+ eptr++;
+ }
+ break;
+
case OP_NOT_DIGIT:
for (i = min; i < max; i++)
{
if (possessive) continue;
while (eptr >= pp)
{
- RMATCH(rrc, eptr, ecode, offset_top, md, ims, eptrb, 0);
+ RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM47);
eptr--;
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
}
} /* End of main loop */
/* Control never reaches here */
+
+
+/* When compiling to use the heap rather than the stack for recursive calls to
+match(), the RRETURN() macro jumps here. The number that is saved in
+frame->Xwhere indicates which label we actually want to return to. */
+
+#ifdef NO_RECURSE
+#define LBL(val) case val: goto L_RM##val;
+HEAP_RETURN:
+switch (frame->Xwhere)
+ {
+ LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8)
+ LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(16)
+ LBL(17) LBL(18) LBL(19) LBL(20) LBL(21) LBL(22) LBL(23) LBL(24)
+ LBL(25) LBL(26) LBL(27) LBL(28) LBL(29) LBL(30) LBL(31) LBL(32)
+ LBL(33) LBL(34) LBL(35) LBL(36) LBL(37) LBL(38) LBL(39) LBL(40)
+ LBL(41) LBL(42) LBL(43) LBL(44) LBL(45) LBL(46) LBL(47)
+ default:
+ DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));
+ return PCRE_ERROR_INTERNAL;
+ }
+#undef LBL
+#endif /* NO_RECURSE */
}
#ifdef NO_RECURSE
#undef eptr
#undef ecode
+#undef mstart
#undef offset_top
#undef ims
#undef eptrb
< -1 => some kind of unexpected problem
*/
-PCRE_DATA_SCOPE int
+PCRE_EXP_DEFN int
pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,
PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,
int offsetcount)
md->lcc = tables + lcc_offset;
md->ctypes = tables + ctypes_offset;
-/* Handle different types of newline. The two bits give four cases. If nothing
-is set at run time, whatever was used at compile time applies. */
+/* Handle different types of newline. The three bits give eight cases. If
+nothing is set at run time, whatever was used at compile time applies. */
-switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : options) &
+switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &
PCRE_NEWLINE_BITS)
{
case 0: newline = NEWLINE; break; /* Compile-time default */
case PCRE_NEWLINE_CR+
PCRE_NEWLINE_LF: newline = ('\r' << 8) | '\n'; break;
case PCRE_NEWLINE_ANY: newline = -1; break;
+ case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
default: return PCRE_ERROR_BADNEWLINE;
}
-if (newline < 0)
+if (newline == -2)
+ {
+ md->nltype = NLTYPE_ANYCRLF;
+ }
+else if (newline < 0)
{
md->nltype = NLTYPE_ANY;
}
{
while (start_match <= end_subject && !WAS_NEWLINE(start_match))
start_match++;
+
+ /* If we have just passed a CR and the newline option is ANY or ANYCRLF,
+ and we are now at a LF, advance the match position by one more character.
+ */
+
+ if (start_match[-1] == '\r' &&
+ (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
+ start_match < end_subject &&
+ *start_match == '\n')
+ start_match++;
}
}
/* OK, we can now run the match. */
- md->start_match = start_match;
+ md->start_match_ptr = start_match; /* Insurance */
md->match_call_count = 0;
md->eptrn = 0; /* Next free eptrchain slot */
- rc = match(start_match, md->start_code, 2, md, ims, NULL, 0, 0);
+ rc = match(start_match, md->start_code, start_match, 2, md,
+ ims, NULL, 0, 0);
/* Any return other than MATCH_NOMATCH breaks the loop. */
if (anchored || start_match > end_subject) break;
- /* If we have just passed a CR and the newline option is CRLF or ANY, and we
- are now at a LF, advance the match position by one more character. */
+ /* If we have just passed a CR and the newline option is CRLF or ANY or
+ ANYCRLF, and we are now at a LF, advance the match position by one more
+ character. */
if (start_match[-1] == '\r' &&
- (md->nltype == NLTYPE_ANY || md->nllen == 2) &&
+ (md->nltype == NLTYPE_ANY ||
+ md->nltype == NLTYPE_ANYCRLF ||
+ md->nllen == 2) &&
start_match < end_subject &&
*start_match == '\n')
start_match++;
rc = md->offset_overflow? 0 : md->end_offset_top/2;
- /* If there is space, set up the whole thing as substring 0. */
+ /* If there is space, set up the whole thing as substring 0. The value of
+ md->start_match_ptr might be modified if \K was encountered on the success
+ matching path. */
if (offsetcount < 2) rc = 0; else
{
- offsets[0] = start_match - md->start_subject;
+ offsets[0] = md->start_match_ptr - md->start_subject;
offsets[1] = md->end_match_ptr - md->start_subject;
}
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
Returns: 0 if data returned, negative on error
*/
-PCRE_DATA_SCOPE int
+PCRE_EXP_DEFN int
pcre_fullinfo(const pcre *argument_re, const pcre_extra *extra_data, int what,
void *where)
{
*((const uschar **)where) = (const uschar *)(_pcre_default_tables);
break;
+ case PCRE_INFO_OKPARTIAL:
+ *((int *)where) = (re->options & PCRE_NOPARTIAL) == 0;
+ break;
+
+ case PCRE_INFO_JCHANGED:
+ *((int *)where) = (re->options & PCRE_JCHANGED) != 0;
+ break;
+
default: return PCRE_ERROR_BADOPTION;
}
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
all threads. However, when compiling for Virtual Pascal, things are done
differently, and global variables are not used (see pcre.in). */
-
#include "pcre_internal.h"
-
-#ifdef __cplusplus
-extern "C" int (*pcre_callout)(pcre_callout_block *) = NULL;
-#else
-int (*pcre_callout)(pcre_callout_block *) = NULL;
-#endif
+PCRE_EXP_DATA_DEFN int (*pcre_callout)(pcre_callout_block *) = NULL;
/* End of pcre_globals.c */
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
or negative values on error
*/
-PCRE_DATA_SCOPE int
+PCRE_EXP_DEFN int
pcre_info(const pcre *argument_re, int *optptr, int *first_byte)
{
real_pcre internal_re;
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
#include <stdlib.h>
#include <string.h>
-#ifndef PCRE_SPY
-#define PCRE_DEFINITION /* Win32 __declspec(export) trigger for .dll */
+/* When compiling a DLL for Windows, the exported symbols have to be declared
+using some MS magic. I found some useful information on this web page:
+http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
+information there, using __declspec(dllexport) without "extern" we have a
+definition; with "extern" we have a declaration. The settings here override the
+setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL,
+which is all that is needed for applications (they just import the symbols). We
+use:
+
+ PCRE_EXP_DECL for declarations
+ PCRE_EXP_DEFN for definitions of exported functions
+ PCRE_EXP_DATA_DEFN for definitions of exported variables
+
+The reason for the two DEFN macros is that in non-Windows environments, one
+does not want to have "extern" before variable definitions because it leads to
+compiler warnings. So we distinguish between functions and variables. In
+Windows, the two should always be the same.
+
+The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest,
+which is an application, but needs to import this file in order to "peek" at
+internals, can #include pcre.h first to get an application's-eye view.
+
+In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
+special-purpose environments) might want to stick other stuff in front of
+exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and
+PCRE_EXP_DATA_DEFN only if they are not already set. */
+
+#ifndef PCRE_EXP_DECL
+# ifdef _WIN32
+# ifdef DLL_EXPORT
+# define PCRE_EXP_DECL extern __declspec(dllexport)
+# define PCRE_EXP_DEFN __declspec(dllexport)
+# define PCRE_EXP_DATA_DEFN __declspec(dllexport)
+# else
+# define PCRE_EXP_DECL extern
+# define PCRE_EXP_DEFN
+# define PCRE_EXP_DATA_DEFN
+# endif
+#
+# else
+# ifdef __cplusplus
+# define PCRE_EXP_DECL extern "C"
+# else
+# define PCRE_EXP_DECL extern
+# endif
+# ifndef PCRE_EXP_DEFN
+# define PCRE_EXP_DEFN PCRE_EXP_DECL
+# endif
+# ifndef PCRE_EXP_DATA_DEFN
+# define PCRE_EXP_DATA_DEFN
+# endif
+# endif
#endif
/* We need to have types that specify unsigned 16-bit and 32-bit integers. We
#define NOTACHAR 0xffffffff
/* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
-and "all" at present). The following macros are used to package up testing for
-newlines. NLBLOCK, PSSTART, and PSEND are defined in the various modules to
-indicate in which datablock the parameters exist, and what the start/end of
-string field names are. */
+"any" and "anycrlf" at present). The following macros are used to package up
+testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
+modules to indicate in which datablock the parameters exist, and what the
+start/end of string field names are. */
-#define NLTYPE_FIXED 0 /* Newline is a fixed length string */
-#define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
+#define NLTYPE_FIXED 0 /* Newline is a fixed length string */
+#define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
+#define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */
/* This macro checks for a newline at the given position */
#define IS_NEWLINE(p) \
((NLBLOCK->nltype != NLTYPE_FIXED)? \
((p) < NLBLOCK->PSEND && \
- _pcre_is_newline((p), NLBLOCK->PSEND, &(NLBLOCK->nllen), utf8) \
- ) \
+ _pcre_is_newline((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\
+ utf8)) \
: \
((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
(p)[0] == NLBLOCK->nl[0] && \
#define WAS_NEWLINE(p) \
((NLBLOCK->nltype != NLTYPE_FIXED)? \
((p) > NLBLOCK->PSSTART && \
- _pcre_was_newline((p), NLBLOCK->PSSTART, &(NLBLOCK->nllen), utf8) \
- ) \
+ _pcre_was_newline((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
+ &(NLBLOCK->nllen), utf8)) \
: \
((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
(p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
#define USPTR const unsigned char *
#endif
+
+
/* Include the public PCRE header and the definitions of UCP character property
values. */
-#include "pcre.h"
+#include <pcre.h>
#include "ucp.h"
/* When compiling for use with the Virtual Pascal compiler, these functions
option on the command line. */
#ifdef VPCOMPAT
+#define strlen(s) _strlen(s)
#define strncmp(s1,s2,m) _strncmp(s1,s2,m)
+#define memcmp(s,c,n) _memcmp(s,c,n)
#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)
/* 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. */
+neither (there some non-Unix environments where this is the case). */
-#if ! HAVE_MEMMOVE
+#ifndef HAVE_MEMMOVE
#undef memmove /* some systems may have a macro */
-#if HAVE_BCOPY
+#ifdef HAVE_BCOPY
#define memmove(a, b, c) bcopy(b, a, c)
#else /* HAVE_BCOPY */
static void *
-pcre_memmove(unsigned char *dest, const unsigned char *src, size_t n)
+pcre_memmove(void *d, const void *s, size_t n)
{
size_t i;
-dest += n;
-src += n;
-for (i = 0; i < n; ++i) *(--dest) = *(--src);
-return dest;
+unsigned char *dest = (unsigned char *)d;
+const unsigned char *src = (const unsigned char *)s;
+if (dest > src)
+ {
+ dest += n;
+ src += n;
+ for (i = 0; i < n; ++i) *(--dest) = *(--src);
+ return (void *)dest;
+ }
+else
+ {
+ for (i = 0; i < n; ++i) *dest++ = *src++;
+ return (void *)(dest - n);
+ }
}
#define memmove(a, b, c) pcre_memmove(a, b, c)
#endif /* not HAVE_BCOPY */
/* Masks for identifying the public options that are permitted at compile
time, run time, or study time, respectively. */
-#define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY)
+#define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
+ PCRE_NEWLINE_ANYCRLF)
#define PUBLIC_OPTIONS \
(PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
consume characters. 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_P, ESC_p, ESC_R, ESC_X, ESC_Z, ESC_z,
- ESC_E, ESC_Q, ESC_k, ESC_REF };
+enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
+ ESC_W, ESC_w, ESC_dum1, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H, ESC_h,
+ ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, ESC_E, ESC_Q, ESC_k, ESC_REF };
/* Opcode table: OP_BRA must be last, as all values >= it are used for brackets
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_NOTPROP, /* 13 \P (not Unicode property) */
- OP_PROP, /* 14 \p (Unicode property) */
- OP_ANYNL, /* 15 \R (any newline sequence) */
- OP_EXTUNI, /* 16 \X (extended Unicode sequence */
- OP_EODN, /* 17 End of data or \n at end of data: \Z. */
- OP_EOD, /* 18 End of data: \z */
-
- OP_OPT, /* 19 Set runtime options */
- OP_CIRC, /* 20 Start of line - varies with multiline switch */
- OP_DOLL, /* 21 End of line - varies with multiline switch */
- OP_CHAR, /* 22 Match one character, casefully */
- OP_CHARNC, /* 23 Match one character, caselessly */
- OP_NOT, /* 24 Match one character, not the following one */
-
- OP_STAR, /* 25 The maximizing and minimizing versions of */
- OP_MINSTAR, /* 26 these six opcodes must come in pairs, with */
- OP_PLUS, /* 27 the minimizing one second. */
- OP_MINPLUS, /* 28 This first set applies to single characters.*/
- OP_QUERY, /* 29 */
- OP_MINQUERY, /* 30 */
-
- OP_UPTO, /* 31 From 0 to n matches */
- OP_MINUPTO, /* 32 */
- OP_EXACT, /* 33 Exactly n matches */
-
- OP_POSSTAR, /* 34 Possessified star */
- OP_POSPLUS, /* 35 Possessified plus */
- OP_POSQUERY, /* 36 Posesssified query */
- OP_POSUPTO, /* 37 Possessified upto */
-
- OP_NOTSTAR, /* 38 The maximizing and minimizing versions of */
- OP_NOTMINSTAR, /* 39 these six opcodes must come in pairs, with */
- OP_NOTPLUS, /* 40 the minimizing one second. They must be in */
- OP_NOTMINPLUS, /* 41 exactly the same order as those above. */
- OP_NOTQUERY, /* 42 This set applies to "not" single characters. */
- OP_NOTMINQUERY, /* 43 */
-
- OP_NOTUPTO, /* 44 From 0 to n matches */
- OP_NOTMINUPTO, /* 45 */
- OP_NOTEXACT, /* 46 Exactly n matches */
-
- OP_NOTPOSSTAR, /* 47 Possessified versions */
- OP_NOTPOSPLUS, /* 48 */
- OP_NOTPOSQUERY, /* 49 */
- OP_NOTPOSUPTO, /* 50 */
-
- OP_TYPESTAR, /* 51 The maximizing and minimizing versions of */
- OP_TYPEMINSTAR, /* 52 these six opcodes must come in pairs, with */
- OP_TYPEPLUS, /* 53 the minimizing one second. These codes must */
- OP_TYPEMINPLUS, /* 54 be in exactly the same order as those above. */
- OP_TYPEQUERY, /* 55 This set applies to character types such as \d */
- OP_TYPEMINQUERY, /* 56 */
-
- OP_TYPEUPTO, /* 57 From 0 to n matches */
- OP_TYPEMINUPTO, /* 58 */
- OP_TYPEEXACT, /* 59 Exactly n matches */
-
- OP_TYPEPOSSTAR, /* 60 Possessified versions */
- OP_TYPEPOSPLUS, /* 61 */
- OP_TYPEPOSQUERY, /* 62 */
- OP_TYPEPOSUPTO, /* 63 */
-
- OP_CRSTAR, /* 64 The maximizing and minimizing versions of */
- OP_CRMINSTAR, /* 65 all these opcodes must come in pairs, with */
- OP_CRPLUS, /* 66 the minimizing one second. These codes must */
- OP_CRMINPLUS, /* 67 be in exactly the same order as those above. */
- OP_CRQUERY, /* 68 These are for character classes and back refs */
- OP_CRMINQUERY, /* 69 */
- OP_CRRANGE, /* 70 These are different to the three sets above. */
- OP_CRMINRANGE, /* 71 */
-
- OP_CLASS, /* 72 Match a character class, chars < 256 only */
- OP_NCLASS, /* 73 Same, but the bitmap was created from a negative
+ OP_SET_SOM, /* 3 Set start of match (\K) */
+ OP_NOT_WORD_BOUNDARY, /* 4 \B */
+ OP_WORD_BOUNDARY, /* 5 \b */
+ OP_NOT_DIGIT, /* 6 \D */
+ OP_DIGIT, /* 7 \d */
+ OP_NOT_WHITESPACE, /* 8 \S */
+ OP_WHITESPACE, /* 9 \s */
+ OP_NOT_WORDCHAR, /* 10 \W */
+ OP_WORDCHAR, /* 11 \w */
+ OP_ANY, /* 12 Match any character */
+ OP_ANYBYTE, /* 13 Match any byte (\C); different to OP_ANY for UTF-8 */
+ OP_NOTPROP, /* 14 \P (not Unicode property) */
+ OP_PROP, /* 15 \p (Unicode property) */
+ OP_ANYNL, /* 16 \R (any newline sequence) */
+ OP_NOT_HSPACE, /* 17 \H (not horizontal whitespace) */
+ OP_HSPACE, /* 18 \h (horizontal whitespace) */
+ OP_NOT_VSPACE, /* 19 \V (not vertical whitespace) */
+ OP_VSPACE, /* 20 \v (vertical whitespace) */
+ OP_EXTUNI, /* 21 \X (extended Unicode sequence */
+ OP_EODN, /* 22 End of data or \n at end of data: \Z. */
+ OP_EOD, /* 23 End of data: \z */
+
+ OP_OPT, /* 24 Set runtime options */
+ OP_CIRC, /* 25 Start of line - varies with multiline switch */
+ OP_DOLL, /* 26 End of line - varies with multiline switch */
+ OP_CHAR, /* 27 Match one character, casefully */
+ OP_CHARNC, /* 28 Match one character, caselessly */
+ OP_NOT, /* 29 Match one character, not the following one */
+
+ OP_STAR, /* 30 The maximizing and minimizing versions of */
+ OP_MINSTAR, /* 31 these six opcodes must come in pairs, with */
+ OP_PLUS, /* 32 the minimizing one second. */
+ OP_MINPLUS, /* 33 This first set applies to single characters.*/
+ OP_QUERY, /* 34 */
+ OP_MINQUERY, /* 35 */
+
+ OP_UPTO, /* 36 From 0 to n matches */
+ OP_MINUPTO, /* 37 */
+ OP_EXACT, /* 38 Exactly n matches */
+
+ OP_POSSTAR, /* 39 Possessified star */
+ OP_POSPLUS, /* 40 Possessified plus */
+ OP_POSQUERY, /* 41 Posesssified query */
+ OP_POSUPTO, /* 42 Possessified upto */
+
+ OP_NOTSTAR, /* 43 The maximizing and minimizing versions of */
+ OP_NOTMINSTAR, /* 44 these six opcodes must come in pairs, with */
+ OP_NOTPLUS, /* 45 the minimizing one second. They must be in */
+ OP_NOTMINPLUS, /* 46 exactly the same order as those above. */
+ OP_NOTQUERY, /* 47 This set applies to "not" single characters. */
+ OP_NOTMINQUERY, /* 48 */
+
+ OP_NOTUPTO, /* 49 From 0 to n matches */
+ OP_NOTMINUPTO, /* 50 */
+ OP_NOTEXACT, /* 51 Exactly n matches */
+
+ OP_NOTPOSSTAR, /* 52 Possessified versions */
+ OP_NOTPOSPLUS, /* 53 */
+ OP_NOTPOSQUERY, /* 54 */
+ OP_NOTPOSUPTO, /* 55 */
+
+ OP_TYPESTAR, /* 56 The maximizing and minimizing versions of */
+ OP_TYPEMINSTAR, /* 57 these six opcodes must come in pairs, with */
+ OP_TYPEPLUS, /* 58 the minimizing one second. These codes must */
+ OP_TYPEMINPLUS, /* 59 be in exactly the same order as those above. */
+ OP_TYPEQUERY, /* 60 This set applies to character types such as \d */
+ OP_TYPEMINQUERY, /* 61 */
+
+ OP_TYPEUPTO, /* 62 From 0 to n matches */
+ OP_TYPEMINUPTO, /* 63 */
+ OP_TYPEEXACT, /* 64 Exactly n matches */
+
+ OP_TYPEPOSSTAR, /* 65 Possessified versions */
+ OP_TYPEPOSPLUS, /* 66 */
+ OP_TYPEPOSQUERY, /* 67 */
+ OP_TYPEPOSUPTO, /* 68 */
+
+ OP_CRSTAR, /* 69 The maximizing and minimizing versions of */
+ OP_CRMINSTAR, /* 70 all these opcodes must come in pairs, with */
+ OP_CRPLUS, /* 71 the minimizing one second. These codes must */
+ OP_CRMINPLUS, /* 72 be in exactly the same order as those above. */
+ OP_CRQUERY, /* 73 These are for character classes and back refs */
+ OP_CRMINQUERY, /* 74 */
+ OP_CRRANGE, /* 75 These are different to the three sets above. */
+ OP_CRMINRANGE, /* 76 */
+
+ OP_CLASS, /* 77 Match a character class, chars < 256 only */
+ OP_NCLASS, /* 78 Same, but the bitmap was created from a negative
class - the difference is relevant only when a UTF-8
character > 255 is encountered. */
- OP_XCLASS, /* 74 Extended class for handling UTF-8 chars within the
+ OP_XCLASS, /* 79 Extended class for handling UTF-8 chars within the
class. This does both positive and negative. */
- OP_REF, /* 75 Match a back reference */
- OP_RECURSE, /* 76 Match a numbered subpattern (possibly recursive) */
- OP_CALLOUT, /* 77 Call out to external function if provided */
+ OP_REF, /* 80 Match a back reference */
+ OP_RECURSE, /* 81 Match a numbered subpattern (possibly recursive) */
+ OP_CALLOUT, /* 82 Call out to external function if provided */
- OP_ALT, /* 78 Start of alternation */
- OP_KET, /* 79 End of group that doesn't have an unbounded repeat */
- OP_KETRMAX, /* 80 These two must remain together and in this */
- OP_KETRMIN, /* 81 order. They are for groups the repeat for ever. */
+ OP_ALT, /* 83 Start of alternation */
+ OP_KET, /* 84 End of group that doesn't have an unbounded repeat */
+ OP_KETRMAX, /* 85 These two must remain together and in this */
+ OP_KETRMIN, /* 86 order. They are for groups the repeat for ever. */
/* The assertions must come before BRA, CBRA, ONCE, and COND.*/
- OP_ASSERT, /* 82 Positive lookahead */
- OP_ASSERT_NOT, /* 83 Negative lookahead */
- OP_ASSERTBACK, /* 84 Positive lookbehind */
- OP_ASSERTBACK_NOT, /* 85 Negative lookbehind */
- OP_REVERSE, /* 86 Move pointer back - used in lookbehind assertions */
+ OP_ASSERT, /* 87 Positive lookahead */
+ OP_ASSERT_NOT, /* 88 Negative lookahead */
+ OP_ASSERTBACK, /* 89 Positive lookbehind */
+ OP_ASSERTBACK_NOT, /* 90 Negative lookbehind */
+ OP_REVERSE, /* 91 Move pointer back - used in lookbehind assertions */
/* ONCE, BRA, CBRA, and COND must come after the assertions, with ONCE first,
as there's a test for >= ONCE for a subpattern that isn't an assertion. */
- OP_ONCE, /* 87 Atomic group */
- OP_BRA, /* 88 Start of non-capturing bracket */
- OP_CBRA, /* 89 Start of capturing bracket */
- OP_COND, /* 90 Conditional group */
+ OP_ONCE, /* 92 Atomic group */
+ OP_BRA, /* 83 Start of non-capturing bracket */
+ OP_CBRA, /* 94 Start of capturing bracket */
+ OP_COND, /* 95 Conditional group */
/* These three must follow the previous three, in the same order. There's a
check for >= SBRA to distinguish the two sets. */
- OP_SBRA, /* 91 Start of non-capturing bracket, check empty */
- OP_SCBRA, /* 92 Start of capturing bracket, check empty */
- OP_SCOND, /* 93 Conditional group, check empty */
+ OP_SBRA, /* 96 Start of non-capturing bracket, check empty */
+ OP_SCBRA, /* 97 Start of capturing bracket, check empty */
+ OP_SCOND, /* 98 Conditional group, check empty */
- OP_CREF, /* 94 Used to hold a capture number as condition */
- OP_RREF, /* 95 Used to hold a recursion number as condition */
- OP_DEF, /* 96 The DEFINE condition */
+ OP_CREF, /* 99 Used to hold a capture number as condition */
+ OP_RREF, /* 100 Used to hold a recursion number as condition */
+ OP_DEF, /* 101 The DEFINE condition */
- OP_BRAZERO, /* 97 These two must remain together and in this */
- OP_BRAMINZERO /* 98 order. */
+ OP_BRAZERO, /* 102 These two must remain together and in this */
+ OP_BRAMINZERO /* 103 order. */
};
for debugging. The macro is referenced only in pcre_printint.c. */
#define OP_NAME_LIST \
- "End", "\\A", "\\G", "\\B", "\\b", "\\D", "\\d", \
+ "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \
"\\S", "\\s", "\\W", "\\w", "Any", "Anybyte", \
- "notprop", "prop", "anynl", "extuni", \
- "\\Z", "\\z", \
+ "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \
+ "extuni", "\\Z", "\\z", \
"Opt", "^", "$", "char", "charnc", "not", \
"*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
"*+","++", "?+", "{", \
#define OP_LENGTHS \
1, /* End */ \
- 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 */ \
+ 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \
1, 1, /* Any, Anybyte */ \
- 3, 3, 1, 1, /* NOTPROP, PROP, EXTUNI, ANYNL */ \
+ 3, 3, 1, /* NOTPROP, PROP, EXTUNI */ \
+ 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \
1, 1, 2, 1, 1, /* \Z, \z, Opt, ^, $ */ \
2, /* Char - the minimum length */ \
2, /* Charnc - the minimum length */ \
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, ERR48, ERR49,
- ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57 };
+ ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58 };
/* 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
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 */
- USPTR save_start; /* Old value of md->start_match */
+ USPTR save_start; /* Old value of mstart */
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_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;
-
/* Structure for building a chain of data for holding the values of the subject
pointer at the start of each subpattern, so as to detect when an empty string
has been matched by a subpattern - to break infinite loops. */
const uschar *start_code; /* For use when recursing */
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 start_match_ptr; /* Start of matched string */
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 eptrn; /* Next free eptrblock */
recursion_info *recursive; /* Linked list of recursion data */
void *callout_data; /* To pass back to callouts */
- struct heapframe *thisframe; /* Used only when compiling for no recursion */
} match_data;
/* A similar structure is used for the same purpose by the DFA matching
one of the exported public functions. They have to be "external" in the C
sense, but are not part of the PCRE public API. */
-extern BOOL _pcre_is_newline(const uschar *, const uschar *, int *,
- BOOL);
+extern BOOL _pcre_is_newline(const uschar *, int, const uschar *,
+ int *, BOOL);
extern int _pcre_ord2utf8(int, uschar *);
extern real_pcre *_pcre_try_flipped(const real_pcre *, real_pcre *,
const pcre_study_data *, pcre_study_data *);
extern int _pcre_ucp_findprop(const unsigned int, int *, int *);
extern unsigned int _pcre_ucp_othercase(const unsigned int);
extern int _pcre_valid_utf8(const uschar *, int);
-extern BOOL _pcre_was_newline(const uschar *, const uschar *, int *,
- BOOL);
+extern BOOL _pcre_was_newline(const uschar *, int, const uschar *,
+ int *, BOOL);
extern BOOL _pcre_xclass(int, const uschar *);
#endif
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
one kind of newline is to be recognized. When a newline is found, its length is
returned. In principle, we could implement several newline "types", each
referring to a different set of newline characters. At present, PCRE supports
-only NLTYPE_FIXED, which gets handled without these functions, and NLTYPE_ALL,
-so for now the type isn't passed into the functions. It can easily be added
-later if required. The full list of Unicode newline characters is taken from
+only NLTYPE_FIXED, which gets handled without these functions, NLTYPE_ANYCRLF,
+and NLTYPE_ANY. The full list of Unicode newline characters is taken from
http://unicode.org/unicode/reports/tr18/. */
Arguments:
ptr pointer to possible newline
+ type the newline type
endptr pointer to the end of the string
lenptr where to return the length
utf8 TRUE if in utf8 mode
*/
BOOL
-_pcre_is_newline(const uschar *ptr, const uschar *endptr, int *lenptr,
- BOOL utf8)
+_pcre_is_newline(const uschar *ptr, int type, const uschar *endptr,
+ int *lenptr, BOOL utf8)
{
int c;
if (utf8) { GETCHAR(c, ptr); } else c = *ptr;
-switch(c)
+
+if (type == NLTYPE_ANYCRLF) switch(c)
+ {
+ case 0x000a: *lenptr = 1; return TRUE; /* LF */
+ case 0x000d: *lenptr = (ptr < endptr - 1 && ptr[1] == 0x0a)? 2 : 1;
+ return TRUE; /* CR */
+ default: return FALSE;
+ }
+
+/* NLTYPE_ANY */
+
+else switch(c)
{
case 0x000a: /* LF */
case 0x000b: /* VT */
Arguments:
ptr pointer to possible newline
+ type the newline type
startptr pointer to the start of the string
lenptr where to return the length
utf8 TRUE if in utf8 mode
*/
BOOL
-_pcre_was_newline(const uschar *ptr, const uschar *startptr, int *lenptr,
- BOOL utf8)
+_pcre_was_newline(const uschar *ptr, int type, const uschar *startptr,
+ int *lenptr, BOOL utf8)
{
int c;
ptr--;
GETCHAR(c, ptr);
}
else c = *ptr;
-switch(c)
+
+if (type == NLTYPE_ANYCRLF) switch(c)
+ {
+ case 0x000a: *lenptr = (ptr > startptr && ptr[-1] == 0x0d)? 2 : 1;
+ return TRUE; /* LF */
+ case 0x000d: *lenptr = 1; return TRUE; /* CR */
+ default: return FALSE;
+ }
+
+else switch(c)
{
case 0x000a: *lenptr = (ptr > startptr && ptr[-1] == 0x0d)? 2 : 1;
return TRUE; /* LF */
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
int
_pcre_ord2utf8(int cvalue, uschar *buffer)
{
+#ifdef SUPPORT_UTF8
register int i, j;
for (i = 0; i < _pcre_utf8_table1_size; i++)
if (cvalue <= _pcre_utf8_table1[i]) break;
}
*buffer = _pcre_utf8_table2[i] | cvalue;
return i + 1;
+#else
+return 0; /* Keep compiler happy; this function won't ever be */
+#endif /* called when SUPPORT_UTF8 is not defined. */
}
/* End of pcre_ord2utf8.c */
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
a negative error number
*/
-PCRE_DATA_SCOPE int
+PCRE_EXP_DEFN int
pcre_refcount(pcre *argument_re, int adjust)
{
real_pcre *re = (real_pcre *)argument_re;
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
character with a value > 255. */
case OP_NCLASS:
+#ifdef SUPPORT_UTF8
if (utf8)
{
start_bits[24] |= 0xf0; /* Bits for 0xc4 - 0xc8 */
memset(start_bits+25, 0xff, 7); /* Bits for 0xc9 - 0xff */
}
+#endif
/* Fall through */
case OP_CLASS:
value is > 127. In fact, there are only two possible starting bytes for
characters in the range 128 - 255. */
+#ifdef SUPPORT_UTF8
if (utf8)
{
for (c = 0; c < 16; c++) start_bits[c] |= tcode[c];
/* In non-UTF-8 mode, the two bit maps are completely compatible. */
else
+#endif
{
for (c = 0; c < 32; c++) start_bits[c] |= tcode[c];
}
NULL on error or if no optimization possible
*/
-PCRE_DATA_SCOPE pcre_extra *
+PCRE_EXP_DEFN pcre_extra *
pcre_study(const pcre *external_re, int options, const char **errorptr)
{
uschar start_bits[32];
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
/* These are the breakpoints for different numbers of bytes in a UTF-8
character. */
+#ifdef SUPPORT_UTF8
+
const int _pcre_utf8_table1[] =
{ 0x7f, 0x7ff, 0xffff, 0x1fffff, 0x3ffffff, 0x7fffffff};
const int _pcre_utt_size = sizeof(_pcre_utt)/sizeof(ucp_type_table);
+#endif /* SUPPORT_UTF8 */
+
/* End of pcre_tables.c */
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
pre-processor time. This hack uses a standard trick for avoiding calling
the STRING macro with an empty argument when doing the test. */
-PCRE_DATA_SCOPE const char *
+PCRE_EXP_DEFN const char *
pcre_version(void)
{
return (XSTRING(Z PCRE_PRERELEASE)[1] == 0)?
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
- Copyright (c) 1997-2006 University of Cambridge
+ Copyright (c) 1997-2007 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
projects / platform / upstream / glib.git / commitdiff