github-starred/ultimatepp
2
0
Fork 0
mirror of https://github.com/ultimatepp/ultimatepp.git synced 2026-07-11 22:03:01 -06:00
Code Issues 33 Projects Packages Wiki Activity Actions

.plugin/pcre: updated to version 8.10

Browse source 
git-svn-id: svn://ultimatepp.org/upp/trunk@2541 f0d560ea-af0d-0410-9eb7-867de7ffcac7
This commit is contained in:
unodgs 2010-07-18 10:57:15 +00:00
parent 2e9138455e
commit fc02cb1f0b
27 changed files with 9323 additions and 5440 deletions
Download patch file Download diff file Expand all files Collapse all files

2
uppsrc/plugin/pcre/lib.cpp
View file

@ -25,7 +25,7 @@
#include "lib/pcre_study.c"
#include "lib/pcre_tables.c"
#include "lib/pcre_try_flipped.c"
#include "lib/pcre_ucp_searchfuncs.c"
#include "lib/pcre_ucd.c"
#include "lib/pcre_valid_utf8.c"
#include "lib/pcre_version.c"
#include "lib/pcre_xclass.c"

42
uppsrc/plugin/pcre/lib/config.h
View file

@ -29,7 +29,10 @@ them both to 0; an emulation function will be used. */
/* If you are compiling for a system that uses EBCDIC instead of ASCII
character codes, define this macro as 1. On systems that can use
"configure", this can be done via --enable-ebcdic. */
"configure", this can be done via --enable-ebcdic. PCRE will then assume
that all input strings are in EBCDIC. If you do not define this macro, PCRE
will assume input strings are ASCII or UTF-8 Unicode. It is not possible to
build a version of PCRE that supports both EBCDIC and UTF-8. */
/* #undef EBCDIC */
/* Define to 1 if you have the `bcopy' function. */
@ -120,12 +123,13 @@ them both to 0; an emulation function will be used. */
#define HAVE_STRING_H 1
#endif
/* Define to 1 if you have the `strtoll' function. */
#ifndef HAVE_STRTOLL
#define HAVE_STRTOLL 1
#endif
/* Define to 1 if you have `strtoimax'. */
/* #undef HAVE_STRTOIMAX */
/* Define to 1 if you have the `strtoq' function. */
/* Define to 1 if you have `strtoll'. */
/* #undef HAVE_STRTOLL */
/* Define to 1 if you have `strtoq'. */
#ifndef HAVE_STRTOQ
#define HAVE_STRTOQ 1
#endif
@ -161,7 +165,7 @@ them both to 0; an emulation function will be used. */
#define HAVE_ZLIB_H 1
#endif
/* Define to 1 if you have the `_strtoi64' function. */
/* Define to 1 if you have `_strtoi64'. */
/* #undef HAVE__STRTOI64 */
/* The value of LINK_SIZE determines the number of bytes used to store links
@ -174,6 +178,12 @@ them both to 0; an emulation function will be used. */
#define LINK_SIZE 2
#endif
/* Define to the sub-directory in which libtool stores uninstalled libraries.
*/
#ifndef LT_OBJDIR
#define LT_OBJDIR ".libs/"
#endif
/* The value of MATCH_LIMIT determines the default number of times the
internal match() function can be called during a single execution of
pcre_exec(). There is a runtime interface for setting a different limit.
@ -240,13 +250,16 @@ them both to 0; an emulation function will be used. */
#define PACKAGE_NAME "PCRE"
/* Define to the full name and version of this package. */
#define PACKAGE_STRING "PCRE 7.6"
#define PACKAGE_STRING "PCRE 8.10"
/* Define to the one symbol short name of this package. */
#define PACKAGE_TARNAME "pcre"
/* Define to the home page for this package. */
#define PACKAGE_URL ""
/* Define to the version of this package. */
#define PACKAGE_VERSION "7.6"
#define PACKAGE_VERSION "8.10"
/* If you are compiling for a system other than a Unix-like system or
@ -294,16 +307,23 @@ them both to 0; an emulation function will be used. */
/* Define to enable support for Unicode properties */
/* #undef SUPPORT_UCP */
/* Define to enable support for the UTF-8 Unicode encoding. */
/* Define to enable support for the UTF-8 Unicode encoding. This will work
even in an EBCDIC environment, but it is incompatible with the EBCDIC
macro. That is, PCRE can support *either* EBCDIC code *or* ASCII/UTF-8, but
not both at once. */
/* #undef SUPPORT_UTF8 */
/* Version number of package */
#ifndef VERSION
#define VERSION "7.6"
#define VERSION "8.10"
#endif
/* Define to empty if `const' does not conform to ANSI C. */
/* #undef const */
/* Define to the type of a signed integer type of width exactly 64 bits if
such a type exists and the standard includes do not define it. */
/* #undef int64_t */
/* Define to `unsigned int' if <sys/types.h> does not define. */
/* #undef size_t */

25
uppsrc/plugin/pcre/lib/pcre.h
View file

@ -5,7 +5,7 @@
/* This is the public header file for the PCRE library, to be #included by
applications that call the PCRE functions.
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2010 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -41,10 +41,10 @@ POSSIBILITY OF SUCH DAMAGE.
/* The current PCRE version information. */
#define PCRE_MAJOR 7
#define PCRE_MINOR 6
#define PCRE_PRERELEASE
#define PCRE_DATE 2008-01-28
#define PCRE_MAJOR 8
#define PCRE_MINOR 10
#define PCRE_PRERELEASE
#define PCRE_DATE 2010-06-25
/* 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
@ -95,7 +95,8 @@ it is needed here for malloc. */
extern "C" {
#endif
/* Options */
/* Options. Some are compile-time only, some are run-time only, and some are
both, so we keep them all distinct. */
#define PCRE_CASELESS 0x00000001
#define PCRE_MULTILINE 0x00000002
@ -112,7 +113,8 @@ extern "C" {
#define PCRE_NO_AUTO_CAPTURE 0x00001000
#define PCRE_NO_UTF8_CHECK 0x00002000
#define PCRE_AUTO_CALLOUT 0x00004000
#define PCRE_PARTIAL 0x00008000
#define PCRE_PARTIAL_SOFT 0x00008000
#define PCRE_PARTIAL 0x00008000 /* Backwards compatible synonym */
#define PCRE_DFA_SHORTEST 0x00010000
#define PCRE_DFA_RESTART 0x00020000
#define PCRE_FIRSTLINE 0x00040000
@ -124,6 +126,12 @@ extern "C" {
#define PCRE_NEWLINE_ANYCRLF 0x00500000
#define PCRE_BSR_ANYCRLF 0x00800000
#define PCRE_BSR_UNICODE 0x01000000
#define PCRE_JAVASCRIPT_COMPAT 0x02000000
#define PCRE_NO_START_OPTIMIZE 0x04000000
#define PCRE_NO_START_OPTIMISE 0x04000000
#define PCRE_PARTIAL_HARD 0x08000000
#define PCRE_NOTEMPTY_ATSTART 0x10000000
#define PCRE_UCP 0x20000000
/* Exec-time and get/set-time error codes */
@ -170,6 +178,7 @@ extern "C" {
#define PCRE_INFO_OKPARTIAL 12
#define PCRE_INFO_JCHANGED 13
#define PCRE_INFO_HASCRORLF 14
#define PCRE_INFO_MINLENGTH 15
/* Request types for pcre_config(). Do not re-arrange, in order to remain
compatible. */
@ -192,6 +201,7 @@ these bits, just add new ones on the end, in order to remain compatible. */
#define PCRE_EXTRA_CALLOUT_DATA 0x0004
#define PCRE_EXTRA_TABLES 0x0008
#define PCRE_EXTRA_MATCH_LIMIT_RECURSION 0x0010
#define PCRE_EXTRA_MARK 0x0020
/* Types */
@ -217,6 +227,7 @@ typedef struct pcre_extra {
void *callout_data; /* Data passed back in callouts */
const unsigned char *tables; /* Pointer to character tables */
unsigned long int match_limit_recursion; /* Max recursive calls to match() */
unsigned char **mark; /* For passing back a mark pointer */
} pcre_extra;
/* The structure for passing out data via the pcre_callout_function. We use a

2268
uppsrc/plugin/pcre/lib/pcre_compile.c
View file

File diff suppressed because it is too large Load diff

8
uppsrc/plugin/pcre/lib/pcre_config.c
View file

@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2009 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -62,7 +62,7 @@ Arguments:
Returns: 0 if data returned, negative on error
*/
PCRE_EXP_DEFN int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_config(int what, void *where)
{
switch (what)
@ -104,11 +104,11 @@ switch (what)
break;
case PCRE_CONFIG_MATCH_LIMIT:
*((unsigned int *)where) = MATCH_LIMIT;
*((unsigned long int *)where) = MATCH_LIMIT;
break;
case PCRE_CONFIG_MATCH_LIMIT_RECURSION:
*((unsigned int *)where) = MATCH_LIMIT_RECURSION;
*((unsigned long int *)where) = MATCH_LIMIT_RECURSION;
break;
case PCRE_CONFIG_STACKRECURSE:

847
uppsrc/plugin/pcre/lib/pcre_dfa_exec.c
View file

File diff suppressed because it is too large Load diff

2607
uppsrc/plugin/pcre/lib/pcre_exec.c
View file

File diff suppressed because it is too large Load diff

17
uppsrc/plugin/pcre/lib/pcre_fullinfo.c
View file

@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2009 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -65,7 +65,7 @@ Arguments:
Returns: 0 if data returned, negative on error
*/
PCRE_EXP_DEFN int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_fullinfo(const pcre *argument_re, const pcre_extra *extra_data, int what,
void *where)
{
@ -89,7 +89,7 @@ if (re->magic_number != MAGIC_NUMBER)
switch (what)
{
case PCRE_INFO_OPTIONS:
*((unsigned long int *)where) = re->options & PUBLIC_OPTIONS;
*((unsigned long int *)where) = re->options & PUBLIC_COMPILE_OPTIONS;
break;
case PCRE_INFO_SIZE:
@ -119,10 +119,16 @@ switch (what)
case PCRE_INFO_FIRSTTABLE:
*((const uschar **)where) =
(study != NULL && (study->options & PCRE_STUDY_MAPPED) != 0)?
(study != NULL && (study->flags & PCRE_STUDY_MAPPED) != 0)?
((const pcre_study_data *)extra_data->study_data)->start_bits : NULL;
break;
case PCRE_INFO_MINLENGTH:
*((int *)where) =
(study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0)?
study->minlength : -1;
break;
case PCRE_INFO_LASTLITERAL:
*((int *)where) =
((re->flags & PCRE_REQCHSET) != 0)? re->req_byte : -1;
@ -144,6 +150,9 @@ switch (what)
*((const uschar **)where) = (const uschar *)(_pcre_default_tables);
break;
/* From release 8.00 this will always return TRUE because NOPARTIAL is
no longer ever set (the restrictions have been removed). */
case PCRE_INFO_OKPARTIAL:
*((int *)where) = (re->flags & PCRE_NOPARTIAL) == 0;
break;

18
uppsrc/plugin/pcre/lib/pcre_get.c
View file

@ -65,7 +65,7 @@ Returns: the number of the named parentheses, or a negative number
(PCRE_ERROR_NOSUBSTRING) if not found
*/
int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_get_stringnumber(const pcre *code, const char *stringname)
{
int rc;
@ -114,7 +114,7 @@ Returns: the length of each entry, or a negative number
(PCRE_ERROR_NOSUBSTRING) if not found
*/
int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_get_stringtable_entries(const pcre *code, const char *stringname,
char **firstptr, char **lastptr)
{
@ -231,7 +231,7 @@ Returns: if successful:
PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
*/
int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_copy_substring(const char *subject, int *ovector, int stringcount,
int stringnumber, char *buffer, int size)
{
@ -276,7 +276,7 @@ Returns: if successful:
PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
*/
int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_copy_named_substring(const pcre *code, const char *subject, int *ovector,
int stringcount, const char *stringname, char *buffer, int size)
{
@ -308,7 +308,7 @@ Returns: if successful: 0
PCRE_ERROR_NOMEMORY (-6) failed to get store
*/
int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_get_substring_list(const char *subject, int *ovector, int stringcount,
const char ***listptr)
{
@ -353,7 +353,7 @@ Argument: the result of a previous pcre_get_substring_list()
Returns: nothing
*/
void
PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
pcre_free_substring_list(const char **pointer)
{
(pcre_free)((void *)pointer);
@ -386,7 +386,7 @@ Returns: if successful:
PCRE_ERROR_NOSUBSTRING (-7) substring not present
*/
int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_get_substring(const char *subject, int *ovector, int stringcount,
int stringnumber, const char **stringptr)
{
@ -433,7 +433,7 @@ Returns: if successful:
PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
*/
int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_get_named_substring(const pcre *code, const char *subject, int *ovector,
int stringcount, const char *stringname, const char **stringptr)
{
@ -456,7 +456,7 @@ Argument: the result of a previous pcre_get_substring()
Returns: nothing
*/
void
PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
pcre_free_substring(const char *pointer)
{
(pcre_free)((void *)pointer);

27
uppsrc/plugin/pcre/lib/pcre_globals.c
View file

@ -43,8 +43,14 @@ PCRE is thread-clean and doesn't use any global variables in the normal sense.
However, it calls memory allocation and freeing functions via the four
indirections below, and it can optionally do callouts, using the fifth
indirection. These values can be changed by the caller, but are shared between
all threads. However, when compiling for Virtual Pascal, things are done
differently, and global variables are not used (see pcre.in). */
all threads.
For MS Visual Studio and Symbian OS, there are problems in initializing these
variables to non-local functions. In these cases, therefore, an indirection via
a local function is used.
Also, when compiling for Virtual Pascal, things are done differently, and
global variables are not used. */
#ifdef HAVE_CONFIG_H
#include "config.h"
@ -52,7 +58,22 @@ differently, and global variables are not used (see pcre.in). */
#include "pcre_internal.h"
#ifndef VPCOMPAT
#if defined _MSC_VER || defined __SYMBIAN32__
static void* LocalPcreMalloc(size_t aSize)
{
return malloc(aSize);
}
static void LocalPcreFree(void* aPtr)
{
free(aPtr);
}
PCRE_EXP_DATA_DEFN void *(*pcre_malloc)(size_t) = LocalPcreMalloc;
PCRE_EXP_DATA_DEFN void (*pcre_free)(void *) = LocalPcreFree;
PCRE_EXP_DATA_DEFN void *(*pcre_stack_malloc)(size_t) = LocalPcreMalloc;
PCRE_EXP_DATA_DEFN void (*pcre_stack_free)(void *) = LocalPcreFree;
PCRE_EXP_DATA_DEFN int (*pcre_callout)(pcre_callout_block *) = NULL;
#elif !defined VPCOMPAT
PCRE_EXP_DATA_DEFN void *(*pcre_malloc)(size_t) = malloc;
PCRE_EXP_DATA_DEFN void (*pcre_free)(void *) = free;
PCRE_EXP_DATA_DEFN void *(*pcre_stack_malloc)(size_t) = malloc;

6
uppsrc/plugin/pcre/lib/pcre_info.c
View file

@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2009 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -72,7 +72,7 @@ Returns: number of capturing subpatterns
or negative values on error
*/
PCRE_EXP_DEFN int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_info(const pcre *argument_re, int *optptr, int *first_byte)
{
real_pcre internal_re;
@ -83,7 +83,7 @@ if (re->magic_number != MAGIC_NUMBER)
re = _pcre_try_flipped(re, &internal_re, NULL, NULL);
if (re == NULL) return PCRE_ERROR_BADMAGIC;
}
if (optptr != NULL) *optptr = (int)(re->options & PUBLIC_OPTIONS);
if (optptr != NULL) *optptr = (int)(re->options & PUBLIC_COMPILE_OPTIONS);
if (first_byte != NULL)
*first_byte = ((re->flags & PCRE_FIRSTSET) != 0)? re->first_byte :
((re->flags & PCRE_STARTLINE) != 0)? -1 : -2;

1045
uppsrc/plugin/pcre/lib/pcre_internal.h
View file

File diff suppressed because it is too large Load diff

8
uppsrc/plugin/pcre/lib/pcre_newline.c
View file

@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2009 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -73,8 +73,7 @@ Returns: TRUE or FALSE
*/
BOOL
_pcre_is_newline(const uschar *ptr, int type, const uschar *endptr,
int *lenptr, BOOL utf8)
_pcre_is_newline(USPTR ptr, int type, USPTR endptr, int *lenptr, BOOL utf8)
{
int c;
if (utf8) { GETCHAR(c, ptr); } else c = *ptr;
@ -123,8 +122,7 @@ Returns: TRUE or FALSE
*/
BOOL
_pcre_was_newline(const uschar *ptr, int type, const uschar *startptr,
int *lenptr, BOOL utf8)
_pcre_was_newline(USPTR ptr, int type, USPTR startptr, int *lenptr, BOOL utf8)
{
int c;
ptr--;

6
uppsrc/plugin/pcre/lib/pcre_ord2utf8.c
View file

@ -78,8 +78,10 @@ for (j = i; j > 0; j--)
*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. */
(void)(cvalue); /* Keep compiler happy; this function won't ever be */
(void)(buffer); /* called when SUPPORT_UTF8 is not defined. */
return 0;
#endif
}
/* End of pcre_ord2utf8.c */

2
uppsrc/plugin/pcre/lib/pcre_refcount.c
View file

@ -68,7 +68,7 @@ Returns: the (possibly updated) count value (a non-negative number), or
a negative error number
*/
PCRE_EXP_DEFN int
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_refcount(pcre *argument_re, int adjust)
{
real_pcre *re = (real_pcre *)argument_re;

172
uppsrc/plugin/pcre/lib/pcre_scanner.h Normal file
View file

@ -0,0 +1,172 @@
// Copyright (c) 2005, Google Inc.
// All rights reserved.
//
// 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 Google Inc. nor the names of its
// 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.
//
// Author: Sanjay Ghemawat
//
// Regular-expression based scanner for parsing an input stream.
//
// Example 1: parse a sequence of "var = number" entries from input:
//
// Scanner scanner(input);
// string var;
// int number;
// scanner.SetSkipExpression("\\s+"); // Skip any white space we encounter
// while (scanner.Consume("(\\w+) = (\\d+)", &var, &number)) {
// ...;
// }
#ifndef _PCRE_SCANNER_H
#define _PCRE_SCANNER_H
#include <assert.h>
#include <string>
#include <vector>
#include <pcrecpp.h>
#include <pcre_stringpiece.h>
namespace pcrecpp {
class PCRECPP_EXP_DEFN Scanner {
public:
Scanner();
explicit Scanner(const std::string& input);
~Scanner();
// Return current line number. The returned line-number is
// one-based. I.e. it returns 1 + the number of consumed newlines.
//
// Note: this method may be slow. It may take time proportional to
// the size of the input.
int LineNumber() const;
// Return the byte-offset that the scanner is looking in the
// input data;
int Offset() const;
// Return true iff the start of the remaining input matches "re"
bool LookingAt(const RE& re) const;
// Return true iff all of the following are true
// a. the start of the remaining input matches "re",
// b. if any arguments are supplied, matched sub-patterns can be
// parsed and stored into the arguments.
// If it returns true, it skips over the matched input and any
// following input that matches the "skip" regular expression.
bool Consume(const RE& re,
const Arg& arg0 = RE::no_arg,
const Arg& arg1 = RE::no_arg,
const Arg& arg2 = RE::no_arg
// TODO: Allow more arguments?
);
// Set the "skip" regular expression. If after consuming some data,
// a prefix of the input matches this RE, it is automatically
// skipped. For example, a programming language scanner would use
// a skip RE that matches white space and comments.
//
// scanner.SetSkipExpression("\\s+|//.*|/[*](.|\n)*?[*]/");
//
// Skipping repeats as long as it succeeds. We used to let people do
// this by writing "(...)*" in the regular expression, but that added
// up to lots of recursive calls within the pcre library, so now we
// control repetition explicitly via the function call API.
//
// You can pass NULL for "re" if you do not want any data to be skipped.
void Skip(const char* re); // DEPRECATED; does *not* repeat
void SetSkipExpression(const char* re);
// Temporarily pause "skip"ing. This
// Skip("Foo"); code ; DisableSkip(); code; EnableSkip()
// is similar to
// Skip("Foo"); code ; Skip(NULL); code ; Skip("Foo");
// but avoids creating/deleting new RE objects.
void DisableSkip();
// Reenable previously paused skipping. Any prefix of the input
// that matches the skip pattern is immediately dropped.
void EnableSkip();
/***** Special wrappers around SetSkip() for some common idioms *****/
// Arranges to skip whitespace, C comments, C++ comments.
// The overall RE is a disjunction of the following REs:
// \\s whitespace
// //.*\n C++ comment
// /[*](.|\n)*?[*]/ C comment (x*? means minimal repetitions of x)
// We get repetition via the semantics of SetSkipExpression, not by using *
void SkipCXXComments() {
SetSkipExpression("\\s|//.*\n|/[*](?:\n|.)*?[*]/");
}
void set_save_comments(bool comments) {
save_comments_ = comments;
}
bool save_comments() {
return save_comments_;
}
// Append to vector ranges the comments found in the
// byte range [start,end] (inclusive) of the input data.
// Only comments that were extracted entirely within that
// range are returned: no range splitting of atomically-extracted
// comments is performed.
void GetComments(int start, int end, std::vector<StringPiece> *ranges);
// Append to vector ranges the comments added
// since the last time this was called. This
// functionality is provided for efficiency when
// interleaving scanning with parsing.
void GetNextComments(std::vector<StringPiece> *ranges);
private:
std::string data_; // All the input data
StringPiece input_; // Unprocessed input
RE* skip_; // If non-NULL, RE for skipping input
bool should_skip_; // If true, use skip_
bool skip_repeat_; // If true, repeat skip_ as long as it works
bool save_comments_; // If true, aggregate the skip expression
// the skipped comments
// TODO: later consider requiring that the StringPieces be added
// in order by their start position
std::vector<StringPiece> *comments_;
// the offset into comments_ that has been returned by GetNextComments
int comments_offset_;
// helper function to consume *skip_ and honour
// save_comments_
void ConsumeSkip();
};
} // namespace pcrecpp
#endif /* _PCRE_SCANNER_H */

725
uppsrc/plugin/pcre/lib/pcre_study.c
View file

@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2010 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -48,34 +48,514 @@ supporting functions. */
#include "pcre_internal.h"
#define SET_BIT(c) start_bits[c/8] |= (1 << (c&7))
/* Returns from set_start_bits() */
enum { SSB_FAIL, SSB_DONE, SSB_CONTINUE };
/*************************************************
* Find the minimum subject length for a group *
*************************************************/
/* Scan a parenthesized group and compute the minimum length of subject that
is needed to match it. This is a lower bound; it does not mean there is a
string of that length that matches. In UTF8 mode, the result is in characters
rather than bytes.
Arguments:
code pointer to start of group (the bracket)
startcode pointer to start of the whole pattern
options the compiling options
Returns: the minimum length
-1 if \C was encountered
-2 internal error (missing capturing bracket)
*/
static int
find_minlength(const uschar *code, const uschar *startcode, int options)
{
int length = -1;
BOOL utf8 = (options & PCRE_UTF8) != 0;
BOOL had_recurse = FALSE;
register int branchlength = 0;
register uschar *cc = (uschar *)code + 1 + LINK_SIZE;
if (*code == OP_CBRA || *code == OP_SCBRA) cc += 2;
/* Scan along the opcodes for this branch. If we get to the end of the
branch, check the length against that of the other branches. */
for (;;)
{
int d, min;
uschar *cs, *ce;
register int op = *cc;
switch (op)
{
case OP_COND:
case OP_SCOND:
/* If there is only one branch in a condition, the implied branch has zero
length, so we don't add anything. This covers the DEFINE "condition"
automatically. */
cs = cc + GET(cc, 1);
if (*cs != OP_ALT)
{
cc = cs + 1 + LINK_SIZE;
break;
}
/* Otherwise we can fall through and treat it the same as any other
subpattern. */
case OP_CBRA:
case OP_SCBRA:
case OP_BRA:
case OP_SBRA:
case OP_ONCE:
d = find_minlength(cc, startcode, options);
if (d < 0) return d;
branchlength += d;
do cc += GET(cc, 1); while (*cc == OP_ALT);
cc += 1 + LINK_SIZE;
break;
/* Reached end of a branch; if it's a ket it is the end of a nested
call. If it's ALT it is an alternation in a nested call. If it is
END it's the end of the outer call. All can be handled by the same code. */
case OP_ALT:
case OP_KET:
case OP_KETRMAX:
case OP_KETRMIN:
case OP_END:
if (length < 0 || (!had_recurse && branchlength < length))
length = branchlength;
if (*cc != OP_ALT) return length;
cc += 1 + LINK_SIZE;
branchlength = 0;
had_recurse = FALSE;
break;
/* Skip over assertive subpatterns */
case OP_ASSERT:
case OP_ASSERT_NOT:
case OP_ASSERTBACK:
case OP_ASSERTBACK_NOT:
do cc += GET(cc, 1); while (*cc == OP_ALT);
/* Fall through */
/* Skip over things that don't match chars */
case OP_REVERSE:
case OP_CREF:
case OP_NCREF:
case OP_RREF:
case OP_NRREF:
case OP_DEF:
case OP_OPT:
case OP_CALLOUT:
case OP_SOD:
case OP_SOM:
case OP_EOD:
case OP_EODN:
case OP_CIRC:
case OP_DOLL:
case OP_NOT_WORD_BOUNDARY:
case OP_WORD_BOUNDARY:
cc += _pcre_OP_lengths[*cc];
break;
/* Skip over a subpattern that has a {0} or {0,x} quantifier */
case OP_BRAZERO:
case OP_BRAMINZERO:
case OP_SKIPZERO:
cc += _pcre_OP_lengths[*cc];
do cc += GET(cc, 1); while (*cc == OP_ALT);
cc += 1 + LINK_SIZE;
break;
/* Handle literal characters and + repetitions */
case OP_CHAR:
case OP_CHARNC:
case OP_NOT:
case OP_PLUS:
case OP_MINPLUS:
case OP_POSPLUS:
case OP_NOTPLUS:
case OP_NOTMINPLUS:
case OP_NOTPOSPLUS:
branchlength++;
cc += 2;
#ifdef SUPPORT_UTF8
if (utf8 && cc[-1] >= 0xc0) cc += _pcre_utf8_table4[cc[-1] & 0x3f];
#endif
break;
case OP_TYPEPLUS:
case OP_TYPEMINPLUS:
case OP_TYPEPOSPLUS:
branchlength++;
cc += (cc[1] == OP_PROP || cc[1] == OP_NOTPROP)? 4 : 2;
break;
/* Handle exact repetitions. The count is already in characters, but we
need to skip over a multibyte character in UTF8 mode. */
case OP_EXACT:
case OP_NOTEXACT:
branchlength += GET2(cc,1);
cc += 4;
#ifdef SUPPORT_UTF8
if (utf8 && cc[-1] >= 0xc0) cc += _pcre_utf8_table4[cc[-1] & 0x3f];
#endif
break;
case OP_TYPEEXACT:
branchlength += GET2(cc,1);
cc += (cc[3] == OP_PROP || cc[3] == OP_NOTPROP)? 6 : 4;
break;
/* Handle single-char non-literal matchers */
case OP_PROP:
case OP_NOTPROP:
cc += 2;
/* Fall through */
case OP_NOT_DIGIT:
case OP_DIGIT:
case OP_NOT_WHITESPACE:
case OP_WHITESPACE:
case OP_NOT_WORDCHAR:
case OP_WORDCHAR:
case OP_ANY:
case OP_ALLANY:
case OP_EXTUNI:
case OP_HSPACE:
case OP_NOT_HSPACE:
case OP_VSPACE:
case OP_NOT_VSPACE:
branchlength++;
cc++;
break;
/* "Any newline" might match two characters */
case OP_ANYNL:
branchlength += 2;
cc++;
break;
/* The single-byte matcher means we can't proceed in UTF-8 mode */
case OP_ANYBYTE:
#ifdef SUPPORT_UTF8
if (utf8) return -1;
#endif
branchlength++;
cc++;
break;
/* For repeated character types, we have to test for \p and \P, which have
an extra two bytes of parameters. */
case OP_TYPESTAR:
case OP_TYPEMINSTAR:
case OP_TYPEQUERY:
case OP_TYPEMINQUERY:
case OP_TYPEPOSSTAR:
case OP_TYPEPOSQUERY:
if (cc[1] == OP_PROP || cc[1] == OP_NOTPROP) cc += 2;
cc += _pcre_OP_lengths[op];
break;
case OP_TYPEUPTO:
case OP_TYPEMINUPTO:
case OP_TYPEPOSUPTO:
if (cc[3] == OP_PROP || cc[3] == OP_NOTPROP) cc += 2;
cc += _pcre_OP_lengths[op];
break;
/* Check a class for variable quantification */
#ifdef SUPPORT_UTF8
case OP_XCLASS:
cc += GET(cc, 1) - 33;
/* Fall through */
#endif
case OP_CLASS:
case OP_NCLASS:
cc += 33;
switch (*cc)
{
case OP_CRPLUS:
case OP_CRMINPLUS:
branchlength++;
/* Fall through */
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRQUERY:
case OP_CRMINQUERY:
cc++;
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
branchlength += GET2(cc,1);
cc += 5;
break;
default:
branchlength++;
break;
}
break;
/* Backreferences and subroutine calls are treated in the same way: we find
the minimum length for the subpattern. A recursion, however, causes an
a flag to be set that causes the length of this branch to be ignored. The
logic is that a recursion can only make sense if there is another
alternation that stops the recursing. That will provide the minimum length
(when no recursion happens). A backreference within the group that it is
referencing behaves in the same way.
If PCRE_JAVASCRIPT_COMPAT is set, a backreference to an unset bracket
matches an empty string (by default it causes a matching failure), so in
that case we must set the minimum length to zero. */
case OP_REF:
if ((options & PCRE_JAVASCRIPT_COMPAT) == 0)
{
ce = cs = (uschar *)_pcre_find_bracket(startcode, utf8, GET2(cc, 1));
if (cs == NULL) return -2;
do ce += GET(ce, 1); while (*ce == OP_ALT);
if (cc > cs && cc < ce)
{
d = 0;
had_recurse = TRUE;
}
else d = find_minlength(cs, startcode, options);
}
else d = 0;
cc += 3;
/* Handle repeated back references */
switch (*cc)
{
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRQUERY:
case OP_CRMINQUERY:
min = 0;
cc++;
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
min = GET2(cc, 1);
cc += 5;
break;
default:
min = 1;
break;
}
branchlength += min * d;
break;
case OP_RECURSE:
cs = ce = (uschar *)startcode + GET(cc, 1);
if (cs == NULL) return -2;
do ce += GET(ce, 1); while (*ce == OP_ALT);
if (cc > cs && cc < ce)
had_recurse = TRUE;
else
branchlength += find_minlength(cs, startcode, options);
cc += 1 + LINK_SIZE;
break;
/* Anything else does not or need not match a character. We can get the
item's length from the table, but for those that can match zero occurrences
of a character, we must take special action for UTF-8 characters. */
case OP_UPTO:
case OP_NOTUPTO:
case OP_MINUPTO:
case OP_NOTMINUPTO:
case OP_POSUPTO:
case OP_STAR:
case OP_MINSTAR:
case OP_NOTMINSTAR:
case OP_POSSTAR:
case OP_NOTPOSSTAR:
case OP_QUERY:
case OP_MINQUERY:
case OP_NOTMINQUERY:
case OP_POSQUERY:
case OP_NOTPOSQUERY:
cc += _pcre_OP_lengths[op];
#ifdef SUPPORT_UTF8
if (utf8 && cc[-1] >= 0xc0) cc += _pcre_utf8_table4[cc[-1] & 0x3f];
#endif
break;
/* Skip these, but we need to add in the name length. */
case OP_MARK:
case OP_PRUNE_ARG:
case OP_SKIP_ARG:
case OP_THEN_ARG:
cc += _pcre_OP_lengths[op] + cc[1];
break;
/* For the record, these are the opcodes that are matched by "default":
OP_ACCEPT, OP_CLOSE, OP_COMMIT, OP_FAIL, OP_PRUNE, OP_SET_SOM, OP_SKIP,
OP_THEN. */
default:
cc += _pcre_OP_lengths[op];
break;
}
}
/* Control never gets here */
}
/*************************************************
* Set a bit and maybe its alternate case *
*************************************************/
/* Given a character, set its bit in the table, and also the bit for the other
version of a letter if we are caseless.
/* Given a character, set its first byte's bit in the table, and also the
corresponding bit for the other version of a letter if we are caseless. In
UTF-8 mode, for characters greater than 127, we can only do the caseless thing
when Unicode property support is available.
Arguments:
start_bits points to the bit map
c is the character
p points to the character
caseless the caseless flag
cd the block with char table pointers
utf8 TRUE for UTF-8 mode
Returns: nothing
Returns: pointer after the character
*/
static const uschar *
set_table_bit(uschar *start_bits, const uschar *p, BOOL caseless,
compile_data *cd, BOOL utf8)
{
unsigned int c = *p;
SET_BIT(c);
#ifdef SUPPORT_UTF8
if (utf8 && c > 127)
{
GETCHARINC(c, p);
#ifdef SUPPORT_UCP
if (caseless)
{
uschar buff[8];
c = UCD_OTHERCASE(c);
(void)_pcre_ord2utf8(c, buff);
SET_BIT(buff[0]);
}
#endif
return p;
}
#endif
/* Not UTF-8 mode, or character is less than 127. */
if (caseless && (cd->ctypes[c] & ctype_letter) != 0) SET_BIT(cd->fcc[c]);
return p + 1;
}
/*************************************************
* Set bits for a positive character type *
*************************************************/
/* This function sets starting bits for a character type. In UTF-8 mode, we can
only do a direct setting for bytes less than 128, as otherwise there can be
confusion with bytes in the middle of UTF-8 characters. In a "traditional"
environment, the tables will only recognize ASCII characters anyway, but in at
least one Windows environment, some higher bytes bits were set in the tables.
So we deal with that case by considering the UTF-8 encoding.
Arguments:
start_bits the starting bitmap
cbit type the type of character wanted
table_limit 32 for non-UTF-8; 16 for UTF-8
cd the block with char table pointers
Returns: nothing
*/
static void
set_bit(uschar *start_bits, unsigned int c, BOOL caseless, compile_data *cd)
set_type_bits(uschar *start_bits, int cbit_type, int table_limit,
compile_data *cd)
{
start_bits[c/8] |= (1 << (c&7));
if (caseless && (cd->ctypes[c] & ctype_letter) != 0)
start_bits[cd->fcc[c]/8] |= (1 << (cd->fcc[c]&7));
register int c;
for (c = 0; c < table_limit; c++) start_bits[c] |= cd->cbits[c+cbit_type];
if (table_limit == 32) return;
for (c = 128; c < 256; c++)
{
if ((cd->cbits[c/8] & (1 << (c&7))) != 0)
{
uschar buff[8];
(void)_pcre_ord2utf8(c, buff);
SET_BIT(buff[0]);
}
}
}
/*************************************************
* Set bits for a negative character type *
*************************************************/
/* This function sets starting bits for a negative character type such as \D.
In UTF-8 mode, we can only do a direct setting for bytes less than 128, as
otherwise there can be confusion with bytes in the middle of UTF-8 characters.
Unlike in the positive case, where we can set appropriate starting bits for
specific high-valued UTF-8 characters, in this case we have to set the bits for
all high-valued characters. The lowest is 0xc2, but we overkill by starting at
0xc0 (192) for simplicity.
Arguments:
start_bits the starting bitmap
cbit type the type of character wanted
table_limit 32 for non-UTF-8; 16 for UTF-8
cd the block with char table pointers
Returns: nothing
*/
static void
set_nottype_bits(uschar *start_bits, int cbit_type, int table_limit,
compile_data *cd)
{
register int c;
for (c = 0; c < table_limit; c++) start_bits[c] |= ~cd->cbits[c+cbit_type];
if (table_limit != 32) for (c = 24; c < 32; c++) start_bits[c] = 0xff;
}
@ -110,6 +590,7 @@ set_start_bits(const uschar *code, uschar *start_bits, BOOL caseless,
{
register int c;
int yield = SSB_DONE;
int table_limit = utf8? 16:32;
#if 0
/* ========================================================================= */
@ -217,6 +698,14 @@ do
tcode += 1 + LINK_SIZE;
break;
/* SKIPZERO skips the bracket. */
case OP_SKIPZERO:
tcode++;
do tcode += GET(tcode,1); while (*tcode == OP_ALT);
tcode += 1 + LINK_SIZE;
break;
/* Single-char * or ? sets the bit and tries the next item */
case OP_STAR:
@ -225,12 +714,7 @@ do
case OP_QUERY:
case OP_MINQUERY:
case OP_POSQUERY:
set_bit(start_bits, tcode[1], caseless, cd);
tcode += 2;
#ifdef SUPPORT_UTF8
if (utf8 && tcode[-1] >= 0xc0)
tcode += _pcre_utf8_table4[tcode[-1] & 0x3f];
#endif
tcode = set_table_bit(start_bits, tcode + 1, caseless, cd, utf8);
break;
/* Single-char upto sets the bit and tries the next */
@ -238,12 +722,7 @@ do
case OP_UPTO:
case OP_MINUPTO:
case OP_POSUPTO:
set_bit(start_bits, tcode[3], caseless, cd);
tcode += 4;
#ifdef SUPPORT_UTF8
if (utf8 && tcode[-1] >= 0xc0)
tcode += _pcre_utf8_table4[tcode[-1] & 0x3f];
#endif
tcode = set_table_bit(start_bits, tcode + 3, caseless, cd, utf8);
break;
/* At least one single char sets the bit and stops */
@ -256,59 +735,86 @@ do
case OP_PLUS:
case OP_MINPLUS:
case OP_POSPLUS:
set_bit(start_bits, tcode[1], caseless, cd);
(void)set_table_bit(start_bits, tcode + 1, caseless, cd, utf8);
try_next = FALSE;
break;
/* Single character type sets the bits and stops */
/* Special spacing and line-terminating items. These recognize specific
lists of characters. The difference between VSPACE and ANYNL is that the
latter can match the two-character CRLF sequence, but that is not
relevant for finding the first character, so their code here is
identical. */
case OP_HSPACE:
SET_BIT(0x09);
SET_BIT(0x20);
if (utf8)
{
SET_BIT(0xC2); /* For U+00A0 */
SET_BIT(0xE1); /* For U+1680, U+180E */
SET_BIT(0xE2); /* For U+2000 - U+200A, U+202F, U+205F */
SET_BIT(0xE3); /* For U+3000 */
}
else SET_BIT(0xA0);
try_next = FALSE;
break;
case OP_ANYNL:
case OP_VSPACE:
SET_BIT(0x0A);
SET_BIT(0x0B);
SET_BIT(0x0C);
SET_BIT(0x0D);
if (utf8)
{
SET_BIT(0xC2); /* For U+0085 */
SET_BIT(0xE2); /* For U+2028, U+2029 */
}
else SET_BIT(0x85);
try_next = FALSE;
break;
/* Single character types set the bits and stop. Note that if PCRE_UCP
is set, we do not see these op codes because \d etc are converted to
properties. Therefore, these apply in the case when only characters less
than 256 are recognized to match the types. */
case OP_NOT_DIGIT:
for (c = 0; c < 32; c++)
start_bits[c] |= ~cd->cbits[c+cbit_digit];
set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
try_next = FALSE;
break;
case OP_DIGIT:
for (c = 0; c < 32; c++)
start_bits[c] |= cd->cbits[c+cbit_digit];
set_type_bits(start_bits, cbit_digit, table_limit, cd);
try_next = FALSE;
break;
/* The cbit_space table has vertical tab as whitespace; we have to
discard it. */
ensure it is set as not whitespace. */
case OP_NOT_WHITESPACE:
for (c = 0; c < 32; c++)
{
int d = cd->cbits[c+cbit_space];
if (c == 1) d &= ~0x08;
start_bits[c] |= ~d;
}
set_nottype_bits(start_bits, cbit_space, table_limit, cd);
start_bits[1] |= 0x08;
try_next = FALSE;
break;
/* The cbit_space table has vertical tab as whitespace; we have to
discard it. */
not set it from the table. */
case OP_WHITESPACE:
for (c = 0; c < 32; c++)
{
int d = cd->cbits[c+cbit_space];
if (c == 1) d &= ~0x08;
start_bits[c] |= d;
}
c = start_bits[1]; /* Save in case it was already set */
set_type_bits(start_bits, cbit_space, table_limit, cd);
start_bits[1] = (start_bits[1] & ~0x08) | c;
try_next = FALSE;
break;
case OP_NOT_WORDCHAR:
for (c = 0; c < 32; c++)
start_bits[c] |= ~cd->cbits[c+cbit_word];
set_nottype_bits(start_bits, cbit_word, table_limit, cd);
try_next = FALSE;
break;
case OP_WORDCHAR:
for (c = 0; c < 32; c++)
start_bits[c] |= cd->cbits[c+cbit_word];
set_type_bits(start_bits, cbit_word, table_limit, cd);
try_next = FALSE;
break;
@ -317,6 +823,7 @@ do
case OP_TYPEPLUS:
case OP_TYPEMINPLUS:
case OP_TYPEPOSPLUS:
tcode++;
break;
@ -340,51 +847,69 @@ do
case OP_TYPEPOSQUERY:
switch(tcode[1])
{
default:
case OP_ANY:
case OP_ALLANY:
return SSB_FAIL;
case OP_HSPACE:
SET_BIT(0x09);
SET_BIT(0x20);
if (utf8)
{
SET_BIT(0xC2); /* For U+00A0 */
SET_BIT(0xE1); /* For U+1680, U+180E */
SET_BIT(0xE2); /* For U+2000 - U+200A, U+202F, U+205F */
SET_BIT(0xE3); /* For U+3000 */
}
else SET_BIT(0xA0);
break;
case OP_ANYNL:
case OP_VSPACE:
SET_BIT(0x0A);
SET_BIT(0x0B);
SET_BIT(0x0C);
SET_BIT(0x0D);
if (utf8)
{
SET_BIT(0xC2); /* For U+0085 */
SET_BIT(0xE2); /* For U+2028, U+2029 */
}
else SET_BIT(0x85);
break;
case OP_NOT_DIGIT:
for (c = 0; c < 32; c++)
start_bits[c] |= ~cd->cbits[c+cbit_digit];
set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
break;
case OP_DIGIT:
for (c = 0; c < 32; c++)
start_bits[c] |= cd->cbits[c+cbit_digit];
set_type_bits(start_bits, cbit_digit, table_limit, cd);
break;
/* The cbit_space table has vertical tab as whitespace; we have to
discard it. */
ensure it gets set as not whitespace. */
case OP_NOT_WHITESPACE:
for (c = 0; c < 32; c++)
{
int d = cd->cbits[c+cbit_space];
if (c == 1) d &= ~0x08;
start_bits[c] |= ~d;
}
set_nottype_bits(start_bits, cbit_space, table_limit, cd);
start_bits[1] |= 0x08;
break;
/* The cbit_space table has vertical tab as whitespace; we have to
discard it. */
avoid setting it. */
case OP_WHITESPACE:
for (c = 0; c < 32; c++)
{
int d = cd->cbits[c+cbit_space];
if (c == 1) d &= ~0x08;
start_bits[c] |= d;
}
c = start_bits[1]; /* Save in case it was already set */
set_type_bits(start_bits, cbit_space, table_limit, cd);
start_bits[1] = (start_bits[1] & ~0x08) | c;
break;
case OP_NOT_WORDCHAR:
for (c = 0; c < 32; c++)
start_bits[c] |= ~cd->cbits[c+cbit_word];
set_nottype_bits(start_bits, cbit_word, table_limit, cd);
break;
case OP_WORDCHAR:
for (c = 0; c < 32; c++)
start_bits[c] |= cd->cbits[c+cbit_word];
set_type_bits(start_bits, cbit_word, table_limit, cd);
break;
}
@ -491,13 +1016,15 @@ Arguments:
set NULL unless error
Returns: pointer to a pcre_extra block, with study_data filled in and the
appropriate flag set;
appropriate flags set;
NULL on error or if no optimization possible
*/
PCRE_EXP_DEFN pcre_extra *
PCRE_EXP_DEFN pcre_extra * PCRE_CALL_CONVENTION
pcre_study(const pcre *external_re, int options, const char **errorptr)
{
int min;
BOOL bits_set = FALSE;
uschar start_bits[32];
pcre_extra *extra;
pcre_study_data *study;
@ -524,30 +1051,39 @@ code = (uschar *)re + re->name_table_offset +
(re->name_count * re->name_entry_size);
/* For an anchored pattern, or an unanchored pattern that has a first char, or
a multiline pattern that matches only at "line starts", no further processing
at present. */
a multiline pattern that matches only at "line starts", there is no point in
seeking a list of starting bytes. */
if ((re->options & PCRE_ANCHORED) != 0 ||
(re->flags & (PCRE_FIRSTSET|PCRE_STARTLINE)) != 0)
return NULL;
if ((re->options & PCRE_ANCHORED) == 0 &&
(re->flags & (PCRE_FIRSTSET|PCRE_STARTLINE)) == 0)
{
/* Set the character tables in the block that is passed around */
/* Set the character tables in the block that is passed around */
tables = re->tables;
if (tables == NULL)
(void)pcre_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,
(void *)(&tables));
tables = re->tables;
if (tables == NULL)
(void)pcre_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,
(void *)(&tables));
compile_block.lcc = tables + lcc_offset;
compile_block.fcc = tables + fcc_offset;
compile_block.cbits = tables + cbits_offset;
compile_block.ctypes = tables + ctypes_offset;
compile_block.lcc = tables + lcc_offset;
compile_block.fcc = tables + fcc_offset;
compile_block.cbits = tables + cbits_offset;
compile_block.ctypes = tables + ctypes_offset;
/* See if we can find a fixed set of initial characters for the pattern. */
/* See if we can find a fixed set of initial characters for the pattern. */
memset(start_bits, 0, 32 * sizeof(uschar));
bits_set = set_start_bits(code, start_bits,
(re->options & PCRE_CASELESS) != 0, (re->options & PCRE_UTF8) != 0,
&compile_block) == SSB_DONE;
}
memset(start_bits, 0, 32 * sizeof(uschar));
if (set_start_bits(code, start_bits, (re->options & PCRE_CASELESS) != 0,
(re->options & PCRE_UTF8) != 0, &compile_block) != SSB_DONE) return NULL;
/* Find the minimum length of subject string. */
min = find_minlength(code, code, re->options);
/* Return NULL if no optimization is possible. */
if (!bits_set && min < 0) return NULL;
/* Get a pcre_extra block and a pcre_study_data block. The study data is put in
the latter, which is pointed to by the former, which may also get additional
@ -570,8 +1106,19 @@ extra->flags = PCRE_EXTRA_STUDY_DATA;
extra->study_data = study;
study->size = sizeof(pcre_study_data);
study->options = PCRE_STUDY_MAPPED;
memcpy(study->start_bits, start_bits, sizeof(start_bits));
study->flags = 0;
if (bits_set)
{
study->flags |= PCRE_STUDY_MAPPED;
memcpy(study->start_bits, start_bits, sizeof(start_bits));
}
if (min >= 0)
{
study->flags |= PCRE_STUDY_MINLEN;
study->minlength = min;
}
return extra;
}

641
uppsrc/plugin/pcre/lib/pcre_tables.c
View file

@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2009 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -87,6 +87,19 @@ const uschar _pcre_utf8_table4[] = {
2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 };
/* Table to translate from particular type value to the general value. */
const int _pcre_ucp_gentype[] = {
ucp_C, ucp_C, ucp_C, ucp_C, ucp_C, /* Cc, Cf, Cn, Co, Cs */
ucp_L, ucp_L, ucp_L, ucp_L, ucp_L, /* Ll, Lu, Lm, Lo, Lt */
ucp_M, ucp_M, ucp_M, /* Mc, Me, Mn */
ucp_N, ucp_N, ucp_N, /* Nd, Nl, No */
ucp_P, ucp_P, ucp_P, ucp_P, ucp_P, /* Pc, Pd, Pe, Pf, Pi */
ucp_P, ucp_P, /* Ps, Po */
ucp_S, ucp_S, ucp_S, ucp_S, /* Sc, Sk, Sm, So */
ucp_Z, ucp_Z, ucp_Z /* Zl, Zp, Zs */
};
/* The pcre_utt[] table below translates Unicode property names into type and
code values. It is searched by binary chop, so must be in collating sequence of
name. Originally, the table contained pointers to the name strings in the first
@ -94,221 +107,425 @@ field of each entry. However, that leads to a large number of relocations when
a shared library is dynamically loaded. A significant reduction is made by
putting all the names into a single, large string and then using offsets in the
table itself. Maintenance is more error-prone, but frequent changes to this
data is unlikely. */
data are unlikely.
July 2008: There is now a script called maint/GenerateUtt.py that can be used
to generate this data instead of maintaining it entirely by hand.
The script was updated in March 2009 to generate a new EBCDIC-compliant
version. Like all other character and string literals that are compared against
the regular expression pattern, we must use STR_ macros instead of literal
strings to make sure that UTF-8 support works on EBCDIC platforms. */
#define STRING_Any0 STR_A STR_n STR_y "\0"
#define STRING_Arabic0 STR_A STR_r STR_a STR_b STR_i STR_c "\0"
#define STRING_Armenian0 STR_A STR_r STR_m STR_e STR_n STR_i STR_a STR_n "\0"
#define STRING_Avestan0 STR_A STR_v STR_e STR_s STR_t STR_a STR_n "\0"
#define STRING_Balinese0 STR_B STR_a STR_l STR_i STR_n STR_e STR_s STR_e "\0"
#define STRING_Bamum0 STR_B STR_a STR_m STR_u STR_m "\0"
#define STRING_Bengali0 STR_B STR_e STR_n STR_g STR_a STR_l STR_i "\0"
#define STRING_Bopomofo0 STR_B STR_o STR_p STR_o STR_m STR_o STR_f STR_o "\0"
#define STRING_Braille0 STR_B STR_r STR_a STR_i STR_l STR_l STR_e "\0"
#define STRING_Buginese0 STR_B STR_u STR_g STR_i STR_n STR_e STR_s STR_e "\0"
#define STRING_Buhid0 STR_B STR_u STR_h STR_i STR_d "\0"
#define STRING_C0 STR_C "\0"
#define STRING_Canadian_Aboriginal0 STR_C STR_a STR_n STR_a STR_d STR_i STR_a STR_n STR_UNDERSCORE STR_A STR_b STR_o STR_r STR_i STR_g STR_i STR_n STR_a STR_l "\0"
#define STRING_Carian0 STR_C STR_a STR_r STR_i STR_a STR_n "\0"
#define STRING_Cc0 STR_C STR_c "\0"
#define STRING_Cf0 STR_C STR_f "\0"
#define STRING_Cham0 STR_C STR_h STR_a STR_m "\0"
#define STRING_Cherokee0 STR_C STR_h STR_e STR_r STR_o STR_k STR_e STR_e "\0"
#define STRING_Cn0 STR_C STR_n "\0"
#define STRING_Co0 STR_C STR_o "\0"
#define STRING_Common0 STR_C STR_o STR_m STR_m STR_o STR_n "\0"
#define STRING_Coptic0 STR_C STR_o STR_p STR_t STR_i STR_c "\0"
#define STRING_Cs0 STR_C STR_s "\0"
#define STRING_Cuneiform0 STR_C STR_u STR_n STR_e STR_i STR_f STR_o STR_r STR_m "\0"
#define STRING_Cypriot0 STR_C STR_y STR_p STR_r STR_i STR_o STR_t "\0"
#define STRING_Cyrillic0 STR_C STR_y STR_r STR_i STR_l STR_l STR_i STR_c "\0"
#define STRING_Deseret0 STR_D STR_e STR_s STR_e STR_r STR_e STR_t "\0"
#define STRING_Devanagari0 STR_D STR_e STR_v STR_a STR_n STR_a STR_g STR_a STR_r STR_i "\0"
#define STRING_Egyptian_Hieroglyphs0 STR_E STR_g STR_y STR_p STR_t STR_i STR_a STR_n STR_UNDERSCORE STR_H STR_i STR_e STR_r STR_o STR_g STR_l STR_y STR_p STR_h STR_s "\0"
#define STRING_Ethiopic0 STR_E STR_t STR_h STR_i STR_o STR_p STR_i STR_c "\0"
#define STRING_Georgian0 STR_G STR_e STR_o STR_r STR_g STR_i STR_a STR_n "\0"
#define STRING_Glagolitic0 STR_G STR_l STR_a STR_g STR_o STR_l STR_i STR_t STR_i STR_c "\0"
#define STRING_Gothic0 STR_G STR_o STR_t STR_h STR_i STR_c "\0"
#define STRING_Greek0 STR_G STR_r STR_e STR_e STR_k "\0"
#define STRING_Gujarati0 STR_G STR_u STR_j STR_a STR_r STR_a STR_t STR_i "\0"
#define STRING_Gurmukhi0 STR_G STR_u STR_r STR_m STR_u STR_k STR_h STR_i "\0"
#define STRING_Han0 STR_H STR_a STR_n "\0"
#define STRING_Hangul0 STR_H STR_a STR_n STR_g STR_u STR_l "\0"
#define STRING_Hanunoo0 STR_H STR_a STR_n STR_u STR_n STR_o STR_o "\0"
#define STRING_Hebrew0 STR_H STR_e STR_b STR_r STR_e STR_w "\0"
#define STRING_Hiragana0 STR_H STR_i STR_r STR_a STR_g STR_a STR_n STR_a "\0"
#define STRING_Imperial_Aramaic0 STR_I STR_m STR_p STR_e STR_r STR_i STR_a STR_l STR_UNDERSCORE STR_A STR_r STR_a STR_m STR_a STR_i STR_c "\0"
#define STRING_Inherited0 STR_I STR_n STR_h STR_e STR_r STR_i STR_t STR_e STR_d "\0"
#define STRING_Inscriptional_Pahlavi0 STR_I STR_n STR_s STR_c STR_r STR_i STR_p STR_t STR_i STR_o STR_n STR_a STR_l STR_UNDERSCORE STR_P STR_a STR_h STR_l STR_a STR_v STR_i "\0"
#define STRING_Inscriptional_Parthian0 STR_I STR_n STR_s STR_c STR_r STR_i STR_p STR_t STR_i STR_o STR_n STR_a STR_l STR_UNDERSCORE STR_P STR_a STR_r STR_t STR_h STR_i STR_a STR_n "\0"
#define STRING_Javanese0 STR_J STR_a STR_v STR_a STR_n STR_e STR_s STR_e "\0"
#define STRING_Kaithi0 STR_K STR_a STR_i STR_t STR_h STR_i "\0"
#define STRING_Kannada0 STR_K STR_a STR_n STR_n STR_a STR_d STR_a "\0"
#define STRING_Katakana0 STR_K STR_a STR_t STR_a STR_k STR_a STR_n STR_a "\0"
#define STRING_Kayah_Li0 STR_K STR_a STR_y STR_a STR_h STR_UNDERSCORE STR_L STR_i "\0"
#define STRING_Kharoshthi0 STR_K STR_h STR_a STR_r STR_o STR_s STR_h STR_t STR_h STR_i "\0"
#define STRING_Khmer0 STR_K STR_h STR_m STR_e STR_r "\0"
#define STRING_L0 STR_L "\0"
#define STRING_L_AMPERSAND0 STR_L STR_AMPERSAND "\0"
#define STRING_Lao0 STR_L STR_a STR_o "\0"
#define STRING_Latin0 STR_L STR_a STR_t STR_i STR_n "\0"
#define STRING_Lepcha0 STR_L STR_e STR_p STR_c STR_h STR_a "\0"
#define STRING_Limbu0 STR_L STR_i STR_m STR_b STR_u "\0"
#define STRING_Linear_B0 STR_L STR_i STR_n STR_e STR_a STR_r STR_UNDERSCORE STR_B "\0"
#define STRING_Lisu0 STR_L STR_i STR_s STR_u "\0"
#define STRING_Ll0 STR_L STR_l "\0"
#define STRING_Lm0 STR_L STR_m "\0"
#define STRING_Lo0 STR_L STR_o "\0"
#define STRING_Lt0 STR_L STR_t "\0"
#define STRING_Lu0 STR_L STR_u "\0"
#define STRING_Lycian0 STR_L STR_y STR_c STR_i STR_a STR_n "\0"
#define STRING_Lydian0 STR_L STR_y STR_d STR_i STR_a STR_n "\0"
#define STRING_M0 STR_M "\0"
#define STRING_Malayalam0 STR_M STR_a STR_l STR_a STR_y STR_a STR_l STR_a STR_m "\0"
#define STRING_Mc0 STR_M STR_c "\0"
#define STRING_Me0 STR_M STR_e "\0"
#define STRING_Meetei_Mayek0 STR_M STR_e STR_e STR_t STR_e STR_i STR_UNDERSCORE STR_M STR_a STR_y STR_e STR_k "\0"
#define STRING_Mn0 STR_M STR_n "\0"
#define STRING_Mongolian0 STR_M STR_o STR_n STR_g STR_o STR_l STR_i STR_a STR_n "\0"
#define STRING_Myanmar0 STR_M STR_y STR_a STR_n STR_m STR_a STR_r "\0"
#define STRING_N0 STR_N "\0"
#define STRING_Nd0 STR_N STR_d "\0"
#define STRING_New_Tai_Lue0 STR_N STR_e STR_w STR_UNDERSCORE STR_T STR_a STR_i STR_UNDERSCORE STR_L STR_u STR_e "\0"
#define STRING_Nko0 STR_N STR_k STR_o "\0"
#define STRING_Nl0 STR_N STR_l "\0"
#define STRING_No0 STR_N STR_o "\0"
#define STRING_Ogham0 STR_O STR_g STR_h STR_a STR_m "\0"
#define STRING_Ol_Chiki0 STR_O STR_l STR_UNDERSCORE STR_C STR_h STR_i STR_k STR_i "\0"
#define STRING_Old_Italic0 STR_O STR_l STR_d STR_UNDERSCORE STR_I STR_t STR_a STR_l STR_i STR_c "\0"
#define STRING_Old_Persian0 STR_O STR_l STR_d STR_UNDERSCORE STR_P STR_e STR_r STR_s STR_i STR_a STR_n "\0"
#define STRING_Old_South_Arabian0 STR_O STR_l STR_d STR_UNDERSCORE STR_S STR_o STR_u STR_t STR_h STR_UNDERSCORE STR_A STR_r STR_a STR_b STR_i STR_a STR_n "\0"
#define STRING_Old_Turkic0 STR_O STR_l STR_d STR_UNDERSCORE STR_T STR_u STR_r STR_k STR_i STR_c "\0"
#define STRING_Oriya0 STR_O STR_r STR_i STR_y STR_a "\0"
#define STRING_Osmanya0 STR_O STR_s STR_m STR_a STR_n STR_y STR_a "\0"
#define STRING_P0 STR_P "\0"
#define STRING_Pc0 STR_P STR_c "\0"
#define STRING_Pd0 STR_P STR_d "\0"
#define STRING_Pe0 STR_P STR_e "\0"
#define STRING_Pf0 STR_P STR_f "\0"
#define STRING_Phags_Pa0 STR_P STR_h STR_a STR_g STR_s STR_UNDERSCORE STR_P STR_a "\0"
#define STRING_Phoenician0 STR_P STR_h STR_o STR_e STR_n STR_i STR_c STR_i STR_a STR_n "\0"
#define STRING_Pi0 STR_P STR_i "\0"
#define STRING_Po0 STR_P STR_o "\0"
#define STRING_Ps0 STR_P STR_s "\0"
#define STRING_Rejang0 STR_R STR_e STR_j STR_a STR_n STR_g "\0"
#define STRING_Runic0 STR_R STR_u STR_n STR_i STR_c "\0"
#define STRING_S0 STR_S "\0"
#define STRING_Samaritan0 STR_S STR_a STR_m STR_a STR_r STR_i STR_t STR_a STR_n "\0"
#define STRING_Saurashtra0 STR_S STR_a STR_u STR_r STR_a STR_s STR_h STR_t STR_r STR_a "\0"
#define STRING_Sc0 STR_S STR_c "\0"
#define STRING_Shavian0 STR_S STR_h STR_a STR_v STR_i STR_a STR_n "\0"
#define STRING_Sinhala0 STR_S STR_i STR_n STR_h STR_a STR_l STR_a "\0"
#define STRING_Sk0 STR_S STR_k "\0"
#define STRING_Sm0 STR_S STR_m "\0"
#define STRING_So0 STR_S STR_o "\0"
#define STRING_Sundanese0 STR_S STR_u STR_n STR_d STR_a STR_n STR_e STR_s STR_e "\0"
#define STRING_Syloti_Nagri0 STR_S STR_y STR_l STR_o STR_t STR_i STR_UNDERSCORE STR_N STR_a STR_g STR_r STR_i "\0"
#define STRING_Syriac0 STR_S STR_y STR_r STR_i STR_a STR_c "\0"
#define STRING_Tagalog0 STR_T STR_a STR_g STR_a STR_l STR_o STR_g "\0"
#define STRING_Tagbanwa0 STR_T STR_a STR_g STR_b STR_a STR_n STR_w STR_a "\0"
#define STRING_Tai_Le0 STR_T STR_a STR_i STR_UNDERSCORE STR_L STR_e "\0"
#define STRING_Tai_Tham0 STR_T STR_a STR_i STR_UNDERSCORE STR_T STR_h STR_a STR_m "\0"
#define STRING_Tai_Viet0 STR_T STR_a STR_i STR_UNDERSCORE STR_V STR_i STR_e STR_t "\0"
#define STRING_Tamil0 STR_T STR_a STR_m STR_i STR_l "\0"
#define STRING_Telugu0 STR_T STR_e STR_l STR_u STR_g STR_u "\0"
#define STRING_Thaana0 STR_T STR_h STR_a STR_a STR_n STR_a "\0"
#define STRING_Thai0 STR_T STR_h STR_a STR_i "\0"
#define STRING_Tibetan0 STR_T STR_i STR_b STR_e STR_t STR_a STR_n "\0"
#define STRING_Tifinagh0 STR_T STR_i STR_f STR_i STR_n STR_a STR_g STR_h "\0"
#define STRING_Ugaritic0 STR_U STR_g STR_a STR_r STR_i STR_t STR_i STR_c "\0"
#define STRING_Vai0 STR_V STR_a STR_i "\0"
#define STRING_Xan0 STR_X STR_a STR_n "\0"
#define STRING_Xps0 STR_X STR_p STR_s "\0"
#define STRING_Xsp0 STR_X STR_s STR_p "\0"
#define STRING_Xwd0 STR_X STR_w STR_d "\0"
#define STRING_Yi0 STR_Y STR_i "\0"
#define STRING_Z0 STR_Z "\0"
#define STRING_Zl0 STR_Z STR_l "\0"
#define STRING_Zp0 STR_Z STR_p "\0"
#define STRING_Zs0 STR_Z STR_s "\0"
const char _pcre_utt_names[] =
"Any\0"
"Arabic\0"
"Armenian\0"
"Balinese\0"
"Bengali\0"
"Bopomofo\0"
"Braille\0"
"Buginese\0"
"Buhid\0"
"C\0"
"Canadian_Aboriginal\0"
"Cc\0"
"Cf\0"
"Cherokee\0"
"Cn\0"
"Co\0"
"Common\0"
"Coptic\0"
"Cs\0"
"Cuneiform\0"
"Cypriot\0"
"Cyrillic\0"
"Deseret\0"
"Devanagari\0"
"Ethiopic\0"
"Georgian\0"
"Glagolitic\0"
"Gothic\0"
"Greek\0"
"Gujarati\0"
"Gurmukhi\0"
"Han\0"
"Hangul\0"
"Hanunoo\0"
"Hebrew\0"
"Hiragana\0"
"Inherited\0"
"Kannada\0"
"Katakana\0"
"Kharoshthi\0"
"Khmer\0"
"L\0"
"L&\0"
"Lao\0"
"Latin\0"
"Limbu\0"
"Linear_B\0"
"Ll\0"
"Lm\0"
"Lo\0"
"Lt\0"
"Lu\0"
"M\0"
"Malayalam\0"
"Mc\0"
"Me\0"
"Mn\0"
"Mongolian\0"
"Myanmar\0"
"N\0"
"Nd\0"
"New_Tai_Lue\0"
"Nko\0"
"Nl\0"
"No\0"
"Ogham\0"
"Old_Italic\0"
"Old_Persian\0"
"Oriya\0"
"Osmanya\0"
"P\0"
"Pc\0"
"Pd\0"
"Pe\0"
"Pf\0"
"Phags_Pa\0"
"Phoenician\0"
"Pi\0"
"Po\0"
"Ps\0"
"Runic\0"
"S\0"
"Sc\0"
"Shavian\0"
"Sinhala\0"
"Sk\0"
"Sm\0"
"So\0"
"Syloti_Nagri\0"
"Syriac\0"
"Tagalog\0"
"Tagbanwa\0"
"Tai_Le\0"
"Tamil\0"
"Telugu\0"
"Thaana\0"
"Thai\0"
"Tibetan\0"
"Tifinagh\0"
"Ugaritic\0"
"Yi\0"
"Z\0"
"Zl\0"
"Zp\0"
"Zs\0";
STRING_Any0
STRING_Arabic0
STRING_Armenian0
STRING_Avestan0
STRING_Balinese0
STRING_Bamum0
STRING_Bengali0
STRING_Bopomofo0
STRING_Braille0
STRING_Buginese0
STRING_Buhid0
STRING_C0
STRING_Canadian_Aboriginal0
STRING_Carian0
STRING_Cc0
STRING_Cf0
STRING_Cham0
STRING_Cherokee0
STRING_Cn0
STRING_Co0
STRING_Common0
STRING_Coptic0
STRING_Cs0
STRING_Cuneiform0
STRING_Cypriot0
STRING_Cyrillic0
STRING_Deseret0
STRING_Devanagari0
STRING_Egyptian_Hieroglyphs0
STRING_Ethiopic0
STRING_Georgian0
STRING_Glagolitic0
STRING_Gothic0
STRING_Greek0
STRING_Gujarati0
STRING_Gurmukhi0
STRING_Han0
STRING_Hangul0
STRING_Hanunoo0
STRING_Hebrew0
STRING_Hiragana0
STRING_Imperial_Aramaic0
STRING_Inherited0
STRING_Inscriptional_Pahlavi0
STRING_Inscriptional_Parthian0
STRING_Javanese0
STRING_Kaithi0
STRING_Kannada0
STRING_Katakana0
STRING_Kayah_Li0
STRING_Kharoshthi0
STRING_Khmer0
STRING_L0
STRING_L_AMPERSAND0
STRING_Lao0
STRING_Latin0
STRING_Lepcha0
STRING_Limbu0
STRING_Linear_B0
STRING_Lisu0
STRING_Ll0
STRING_Lm0
STRING_Lo0
STRING_Lt0
STRING_Lu0
STRING_Lycian0
STRING_Lydian0
STRING_M0
STRING_Malayalam0
STRING_Mc0
STRING_Me0
STRING_Meetei_Mayek0
STRING_Mn0
STRING_Mongolian0
STRING_Myanmar0
STRING_N0
STRING_Nd0
STRING_New_Tai_Lue0
STRING_Nko0
STRING_Nl0
STRING_No0
STRING_Ogham0
STRING_Ol_Chiki0
STRING_Old_Italic0
STRING_Old_Persian0
STRING_Old_South_Arabian0
STRING_Old_Turkic0
STRING_Oriya0
STRING_Osmanya0
STRING_P0
STRING_Pc0
STRING_Pd0
STRING_Pe0
STRING_Pf0
STRING_Phags_Pa0
STRING_Phoenician0
STRING_Pi0
STRING_Po0
STRING_Ps0
STRING_Rejang0
STRING_Runic0
STRING_S0
STRING_Samaritan0
STRING_Saurashtra0
STRING_Sc0
STRING_Shavian0
STRING_Sinhala0
STRING_Sk0
STRING_Sm0
STRING_So0
STRING_Sundanese0
STRING_Syloti_Nagri0
STRING_Syriac0
STRING_Tagalog0
STRING_Tagbanwa0
STRING_Tai_Le0
STRING_Tai_Tham0
STRING_Tai_Viet0
STRING_Tamil0
STRING_Telugu0
STRING_Thaana0
STRING_Thai0
STRING_Tibetan0
STRING_Tifinagh0
STRING_Ugaritic0
STRING_Vai0
STRING_Xan0
STRING_Xps0
STRING_Xsp0
STRING_Xwd0
STRING_Yi0
STRING_Z0
STRING_Zl0
STRING_Zp0
STRING_Zs0;
const ucp_type_table _pcre_utt[] = {
{ 0, PT_ANY, 0 },
{ 4, PT_SC, ucp_Arabic },
{ 11, PT_SC, ucp_Armenian },
{ 20, PT_SC, ucp_Balinese },
{ 29, PT_SC, ucp_Bengali },
{ 37, PT_SC, ucp_Bopomofo },
{ 46, PT_SC, ucp_Braille },
{ 54, PT_SC, ucp_Buginese },
{ 63, PT_SC, ucp_Buhid },
{ 69, PT_GC, ucp_C },
{ 71, PT_SC, ucp_Canadian_Aboriginal },
{ 91, PT_PC, ucp_Cc },
{ 94, PT_PC, ucp_Cf },
{ 97, PT_SC, ucp_Cherokee },
{ 106, PT_PC, ucp_Cn },
{ 109, PT_PC, ucp_Co },
{ 112, PT_SC, ucp_Common },
{ 119, PT_SC, ucp_Coptic },
{ 126, PT_PC, ucp_Cs },
{ 129, PT_SC, ucp_Cuneiform },
{ 139, PT_SC, ucp_Cypriot },
{ 147, PT_SC, ucp_Cyrillic },
{ 156, PT_SC, ucp_Deseret },
{ 164, PT_SC, ucp_Devanagari },
{ 175, PT_SC, ucp_Ethiopic },
{ 184, PT_SC, ucp_Georgian },
{ 193, PT_SC, ucp_Glagolitic },
{ 204, PT_SC, ucp_Gothic },
{ 211, PT_SC, ucp_Greek },
{ 217, PT_SC, ucp_Gujarati },
{ 226, PT_SC, ucp_Gurmukhi },
{ 235, PT_SC, ucp_Han },
{ 239, PT_SC, ucp_Hangul },
{ 246, PT_SC, ucp_Hanunoo },
{ 254, PT_SC, ucp_Hebrew },
{ 261, PT_SC, ucp_Hiragana },
{ 270, PT_SC, ucp_Inherited },
{ 280, PT_SC, ucp_Kannada },
{ 288, PT_SC, ucp_Katakana },
{ 297, PT_SC, ucp_Kharoshthi },
{ 308, PT_SC, ucp_Khmer },
{ 314, PT_GC, ucp_L },
{ 316, PT_LAMP, 0 },
{ 319, PT_SC, ucp_Lao },
{ 323, PT_SC, ucp_Latin },
{ 329, PT_SC, ucp_Limbu },
{ 335, PT_SC, ucp_Linear_B },
{ 344, PT_PC, ucp_Ll },
{ 347, PT_PC, ucp_Lm },
{ 350, PT_PC, ucp_Lo },
{ 353, PT_PC, ucp_Lt },
{ 356, PT_PC, ucp_Lu },
{ 359, PT_GC, ucp_M },
{ 361, PT_SC, ucp_Malayalam },
{ 371, PT_PC, ucp_Mc },
{ 374, PT_PC, ucp_Me },
{ 377, PT_PC, ucp_Mn },
{ 380, PT_SC, ucp_Mongolian },
{ 390, PT_SC, ucp_Myanmar },
{ 398, PT_GC, ucp_N },
{ 400, PT_PC, ucp_Nd },
{ 403, PT_SC, ucp_New_Tai_Lue },
{ 415, PT_SC, ucp_Nko },
{ 419, PT_PC, ucp_Nl },
{ 422, PT_PC, ucp_No },
{ 425, PT_SC, ucp_Ogham },
{ 431, PT_SC, ucp_Old_Italic },
{ 442, PT_SC, ucp_Old_Persian },
{ 454, PT_SC, ucp_Oriya },
{ 460, PT_SC, ucp_Osmanya },
{ 468, PT_GC, ucp_P },
{ 470, PT_PC, ucp_Pc },
{ 473, PT_PC, ucp_Pd },
{ 476, PT_PC, ucp_Pe },
{ 479, PT_PC, ucp_Pf },
{ 482, PT_SC, ucp_Phags_Pa },
{ 491, PT_SC, ucp_Phoenician },
{ 502, PT_PC, ucp_Pi },
{ 505, PT_PC, ucp_Po },
{ 508, PT_PC, ucp_Ps },
{ 511, PT_SC, ucp_Runic },
{ 517, PT_GC, ucp_S },
{ 519, PT_PC, ucp_Sc },
{ 522, PT_SC, ucp_Shavian },
{ 530, PT_SC, ucp_Sinhala },
{ 538, PT_PC, ucp_Sk },
{ 541, PT_PC, ucp_Sm },
{ 544, PT_PC, ucp_So },
{ 547, PT_SC, ucp_Syloti_Nagri },
{ 560, PT_SC, ucp_Syriac },
{ 567, PT_SC, ucp_Tagalog },
{ 575, PT_SC, ucp_Tagbanwa },
{ 584, PT_SC, ucp_Tai_Le },
{ 591, PT_SC, ucp_Tamil },
{ 597, PT_SC, ucp_Telugu },
{ 604, PT_SC, ucp_Thaana },
{ 611, PT_SC, ucp_Thai },
{ 616, PT_SC, ucp_Tibetan },
{ 624, PT_SC, ucp_Tifinagh },
{ 633, PT_SC, ucp_Ugaritic },
{ 642, PT_SC, ucp_Yi },
{ 645, PT_GC, ucp_Z },
{ 647, PT_PC, ucp_Zl },
{ 650, PT_PC, ucp_Zp },
{ 653, PT_PC, ucp_Zs }
{ 0, PT_ANY, 0 },
{ 4, PT_SC, ucp_Arabic },
{ 11, PT_SC, ucp_Armenian },
{ 20, PT_SC, ucp_Avestan },
{ 28, PT_SC, ucp_Balinese },
{ 37, PT_SC, ucp_Bamum },
{ 43, PT_SC, ucp_Bengali },
{ 51, PT_SC, ucp_Bopomofo },
{ 60, PT_SC, ucp_Braille },
{ 68, PT_SC, ucp_Buginese },
{ 77, PT_SC, ucp_Buhid },
{ 83, PT_GC, ucp_C },
{ 85, PT_SC, ucp_Canadian_Aboriginal },
{ 105, PT_SC, ucp_Carian },
{ 112, PT_PC, ucp_Cc },
{ 115, PT_PC, ucp_Cf },
{ 118, PT_SC, ucp_Cham },
{ 123, PT_SC, ucp_Cherokee },
{ 132, PT_PC, ucp_Cn },
{ 135, PT_PC, ucp_Co },
{ 138, PT_SC, ucp_Common },
{ 145, PT_SC, ucp_Coptic },
{ 152, PT_PC, ucp_Cs },
{ 155, PT_SC, ucp_Cuneiform },
{ 165, PT_SC, ucp_Cypriot },
{ 173, PT_SC, ucp_Cyrillic },
{ 182, PT_SC, ucp_Deseret },
{ 190, PT_SC, ucp_Devanagari },
{ 201, PT_SC, ucp_Egyptian_Hieroglyphs },
{ 222, PT_SC, ucp_Ethiopic },
{ 231, PT_SC, ucp_Georgian },
{ 240, PT_SC, ucp_Glagolitic },
{ 251, PT_SC, ucp_Gothic },
{ 258, PT_SC, ucp_Greek },
{ 264, PT_SC, ucp_Gujarati },
{ 273, PT_SC, ucp_Gurmukhi },
{ 282, PT_SC, ucp_Han },
{ 286, PT_SC, ucp_Hangul },
{ 293, PT_SC, ucp_Hanunoo },
{ 301, PT_SC, ucp_Hebrew },
{ 308, PT_SC, ucp_Hiragana },
{ 317, PT_SC, ucp_Imperial_Aramaic },
{ 334, PT_SC, ucp_Inherited },
{ 344, PT_SC, ucp_Inscriptional_Pahlavi },
{ 366, PT_SC, ucp_Inscriptional_Parthian },
{ 389, PT_SC, ucp_Javanese },
{ 398, PT_SC, ucp_Kaithi },
{ 405, PT_SC, ucp_Kannada },
{ 413, PT_SC, ucp_Katakana },
{ 422, PT_SC, ucp_Kayah_Li },
{ 431, PT_SC, ucp_Kharoshthi },
{ 442, PT_SC, ucp_Khmer },
{ 448, PT_GC, ucp_L },
{ 450, PT_LAMP, 0 },
{ 453, PT_SC, ucp_Lao },
{ 457, PT_SC, ucp_Latin },
{ 463, PT_SC, ucp_Lepcha },
{ 470, PT_SC, ucp_Limbu },
{ 476, PT_SC, ucp_Linear_B },
{ 485, PT_SC, ucp_Lisu },
{ 490, PT_PC, ucp_Ll },
{ 493, PT_PC, ucp_Lm },
{ 496, PT_PC, ucp_Lo },
{ 499, PT_PC, ucp_Lt },
{ 502, PT_PC, ucp_Lu },
{ 505, PT_SC, ucp_Lycian },
{ 512, PT_SC, ucp_Lydian },
{ 519, PT_GC, ucp_M },
{ 521, PT_SC, ucp_Malayalam },
{ 531, PT_PC, ucp_Mc },
{ 534, PT_PC, ucp_Me },
{ 537, PT_SC, ucp_Meetei_Mayek },
{ 550, PT_PC, ucp_Mn },
{ 553, PT_SC, ucp_Mongolian },
{ 563, PT_SC, ucp_Myanmar },
{ 571, PT_GC, ucp_N },
{ 573, PT_PC, ucp_Nd },
{ 576, PT_SC, ucp_New_Tai_Lue },
{ 588, PT_SC, ucp_Nko },
{ 592, PT_PC, ucp_Nl },
{ 595, PT_PC, ucp_No },
{ 598, PT_SC, ucp_Ogham },
{ 604, PT_SC, ucp_Ol_Chiki },
{ 613, PT_SC, ucp_Old_Italic },
{ 624, PT_SC, ucp_Old_Persian },
{ 636, PT_SC, ucp_Old_South_Arabian },
{ 654, PT_SC, ucp_Old_Turkic },
{ 665, PT_SC, ucp_Oriya },
{ 671, PT_SC, ucp_Osmanya },
{ 679, PT_GC, ucp_P },
{ 681, PT_PC, ucp_Pc },
{ 684, PT_PC, ucp_Pd },
{ 687, PT_PC, ucp_Pe },
{ 690, PT_PC, ucp_Pf },
{ 693, PT_SC, ucp_Phags_Pa },
{ 702, PT_SC, ucp_Phoenician },
{ 713, PT_PC, ucp_Pi },
{ 716, PT_PC, ucp_Po },
{ 719, PT_PC, ucp_Ps },
{ 722, PT_SC, ucp_Rejang },
{ 729, PT_SC, ucp_Runic },
{ 735, PT_GC, ucp_S },
{ 737, PT_SC, ucp_Samaritan },
{ 747, PT_SC, ucp_Saurashtra },
{ 758, PT_PC, ucp_Sc },
{ 761, PT_SC, ucp_Shavian },
{ 769, PT_SC, ucp_Sinhala },
{ 777, PT_PC, ucp_Sk },
{ 780, PT_PC, ucp_Sm },
{ 783, PT_PC, ucp_So },
{ 786, PT_SC, ucp_Sundanese },
{ 796, PT_SC, ucp_Syloti_Nagri },
{ 809, PT_SC, ucp_Syriac },
{ 816, PT_SC, ucp_Tagalog },
{ 824, PT_SC, ucp_Tagbanwa },
{ 833, PT_SC, ucp_Tai_Le },
{ 840, PT_SC, ucp_Tai_Tham },
{ 849, PT_SC, ucp_Tai_Viet },
{ 858, PT_SC, ucp_Tamil },
{ 864, PT_SC, ucp_Telugu },
{ 871, PT_SC, ucp_Thaana },
{ 878, PT_SC, ucp_Thai },
{ 883, PT_SC, ucp_Tibetan },
{ 891, PT_SC, ucp_Tifinagh },
{ 900, PT_SC, ucp_Ugaritic },
{ 909, PT_SC, ucp_Vai },
{ 913, PT_ALNUM, 0 },
{ 917, PT_PXSPACE, 0 },
{ 921, PT_SPACE, 0 },
{ 925, PT_WORD, 0 },
{ 929, PT_SC, ucp_Yi },
{ 932, PT_GC, ucp_Z },
{ 934, PT_PC, ucp_Zl },
{ 937, PT_PC, ucp_Zp },
{ 940, PT_PC, ucp_Zs }
};
const int _pcre_utt_size = sizeof(_pcre_utt)/sizeof(ucp_type_table);

6
uppsrc/plugin/pcre/lib/pcre_try_flipped.c
View file

@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2009 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -128,7 +128,9 @@ if (study != NULL)
{
*internal_study = *study; /* To copy other fields */
internal_study->size = byteflip(study->size, sizeof(study->size));
internal_study->options = byteflip(study->options, sizeof(study->options));
internal_study->flags = byteflip(study->flags, sizeof(study->flags));
internal_study->minlength = byteflip(study->minlength,
sizeof(study->minlength));
}
return internal_re;

2829
uppsrc/plugin/pcre/lib/pcre_ucd.c Normal file
View file

File diff suppressed because it is too large Load diff

179
uppsrc/plugin/pcre/lib/pcre_ucp_searchfuncs.c
View file

@ -1,179 +0,0 @@
/*************************************************
* Perl-Compatible Regular Expressions *
*************************************************/
/* 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.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
-----------------------------------------------------------------------------
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 names of its
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.
-----------------------------------------------------------------------------
*/
/* This module contains code for searching the table of Unicode character
properties. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
#include "ucp.h" /* Category definitions */
#include "ucpinternal.h" /* Internal table details */
#include "ucptable.h" /* The table itself */
/* Table to translate from particular type value to the general value. */
static const int ucp_gentype[] = {
ucp_C, ucp_C, ucp_C, ucp_C, ucp_C, /* Cc, Cf, Cn, Co, Cs */
ucp_L, ucp_L, ucp_L, ucp_L, ucp_L, /* Ll, Lu, Lm, Lo, Lt */
ucp_M, ucp_M, ucp_M, /* Mc, Me, Mn */
ucp_N, ucp_N, ucp_N, /* Nd, Nl, No */
ucp_P, ucp_P, ucp_P, ucp_P, ucp_P, /* Pc, Pd, Pe, Pf, Pi */
ucp_P, ucp_P, /* Ps, Po */
ucp_S, ucp_S, ucp_S, ucp_S, /* Sc, Sk, Sm, So */
ucp_Z, ucp_Z, ucp_Z /* Zl, Zp, Zs */
};
/*************************************************
* Search table and return type *
*************************************************/
/* Three values are returned: the category is ucp_C, ucp_L, etc. The detailed
character type is ucp_Lu, ucp_Nd, etc. The script is ucp_Latin, etc.
Arguments:
c the character value
type_ptr the detailed character type is returned here
script_ptr the script is returned here
Returns: the character type category
*/
int
_pcre_ucp_findprop(const unsigned int c, int *type_ptr, int *script_ptr)
{
int bot = 0;
int top = sizeof(ucp_table)/sizeof(cnode);
int mid;
/* The table is searched using a binary chop. You might think that using
intermediate variables to hold some of the common expressions would speed
things up, but tests with gcc 3.4.4 on Linux showed that, on the contrary, it
makes things a lot slower. */
for (;;)
{
if (top <= bot)
{
*type_ptr = ucp_Cn;
*script_ptr = ucp_Common;
return ucp_C;
}
mid = (bot + top) >> 1;
if (c == (ucp_table[mid].f0 & f0_charmask)) break;
if (c < (ucp_table[mid].f0 & f0_charmask)) top = mid;
else
{
if ((ucp_table[mid].f0 & f0_rangeflag) != 0 &&
c <= (ucp_table[mid].f0 & f0_charmask) +
(ucp_table[mid].f1 & f1_rangemask)) break;
bot = mid + 1;
}
}
/* Found an entry in the table. Set the script and detailed type values, and
return the general type. */
*script_ptr = (ucp_table[mid].f0 & f0_scriptmask) >> f0_scriptshift;
*type_ptr = (ucp_table[mid].f1 & f1_typemask) >> f1_typeshift;
return ucp_gentype[*type_ptr];
}
/*************************************************
* Search table and return other case *
*************************************************/
/* If the given character is a letter, and there is another case for the
letter, return the other case. Otherwise, return -1.
Arguments:
c the character value
Returns: the other case or NOTACHAR if none
*/
unsigned int
_pcre_ucp_othercase(const unsigned int c)
{
int bot = 0;
int top = sizeof(ucp_table)/sizeof(cnode);
int mid, offset;
/* The table is searched using a binary chop. You might think that using
intermediate variables to hold some of the common expressions would speed
things up, but tests with gcc 3.4.4 on Linux showed that, on the contrary, it
makes things a lot slower. */
for (;;)
{
if (top <= bot) return -1;
mid = (bot + top) >> 1;
if (c == (ucp_table[mid].f0 & f0_charmask)) break;
if (c < (ucp_table[mid].f0 & f0_charmask)) top = mid;
else
{
if ((ucp_table[mid].f0 & f0_rangeflag) != 0 &&
c <= (ucp_table[mid].f0 & f0_charmask) +
(ucp_table[mid].f1 & f1_rangemask)) break;
bot = mid + 1;
}
}
/* Found an entry in the table. Return NOTACHAR for a range entry. Otherwise
return the other case if there is one, else NOTACHAR. */
if ((ucp_table[mid].f0 & f0_rangeflag) != 0) return NOTACHAR;
offset = ucp_table[mid].f1 & f1_casemask;
if ((offset & f1_caseneg) != 0) offset |= f1_caseneg;
return (offset == 0)? NOTACHAR : c + offset;
}
/* End of pcre_ucp_searchfuncs.c */

9
uppsrc/plugin/pcre/lib/pcre_valid_utf8.c
View file

@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2009 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -75,10 +75,10 @@ Returns: < 0 if the string is a valid UTF-8 string
*/
int
_pcre_valid_utf8(const uschar *string, int length)
_pcre_valid_utf8(USPTR string, int length)
{
#ifdef SUPPORT_UTF8
register const uschar *p;
register USPTR p;
if (length < 0)
{
@ -154,6 +154,9 @@ for (p = string; length-- > 0; p++)
if ((*(++p) & 0xc0) != 0x80) return p - string;
}
}
#else
(void)(string); /* Keep picky compilers happy */
(void)(length);
#endif
return -1;

2
uppsrc/plugin/pcre/lib/pcre_version.c
View file

@ -79,7 +79,7 @@ I could find no way of detecting that a macro is defined as an empty string at
pre-processor time. This hack uses a standard trick for avoiding calling
the STRING macro with an empty argument when doing the test. */
PCRE_EXP_DEFN const char *
PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION
pcre_version(void)
{
return (XSTRING(Z PCRE_PRERELEASE)[1] == 0)?

48
uppsrc/plugin/pcre/lib/pcre_xclass.c
View file

@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2008 University of Cambridge
Copyright (c) 1997-2010 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@ -39,8 +39,7 @@ POSSIBILITY OF SUCH DAMAGE.
/* This module contains an internal function that is used to match an extended
class (one that contains characters whose values are > 255). It is used by both
pcre_exec() and pcre_def_exec(). */
class. It is used by both pcre_exec() and pcre_def_exec(). */
#ifdef HAVE_CONFIG_H
@ -55,7 +54,7 @@ pcre_exec() and pcre_def_exec(). */
*************************************************/
/* This function is called to match a character against an extended class that
might contain values > 255.
might contain values > 255 and/or Unicode properties.
Arguments:
c the character
@ -104,8 +103,7 @@ while ((t = *data++) != XCL_END)
#ifdef SUPPORT_UCP
else /* XCL_PROP & XCL_NOTPROP */
{
int chartype, script;
int category = _pcre_ucp_findprop(c, &chartype, &script);
const ucd_record *prop = GET_UCD(c);
switch(*data)
{
@ -114,20 +112,48 @@ while ((t = *data++) != XCL_END)
break;
case PT_LAMP:
if ((chartype == ucp_Lu || chartype == ucp_Ll || chartype == ucp_Lt) ==
(t == XCL_PROP)) return !negated;
if ((prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
prop->chartype == ucp_Lt) == (t == XCL_PROP)) return !negated;
break;
case PT_GC:
if ((data[1] == category) == (t == XCL_PROP)) return !negated;
if ((data[1] == _pcre_ucp_gentype[prop->chartype]) == (t == XCL_PROP))
return !negated;
break;
case PT_PC:
if ((data[1] == chartype) == (t == XCL_PROP)) return !negated;
if ((data[1] == prop->chartype) == (t == XCL_PROP)) return !negated;
break;
case PT_SC:
if ((data[1] == script) == (t == XCL_PROP)) return !negated;
if ((data[1] == prop->script) == (t == XCL_PROP)) return !negated;
break;
case PT_ALNUM:
if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N) == (t == XCL_PROP))
return !negated;
break;
case PT_SPACE: /* Perl space */
if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
== (t == XCL_PROP))
return !negated;
break;
case PT_PXSPACE: /* POSIX space */
if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
c == CHAR_FF || c == CHAR_CR) == (t == XCL_PROP))
return !negated;
break;
case PT_WORD:
if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N || c == CHAR_UNDERSCORE)
== (t == XCL_PROP))
return !negated;
break;
/* This should never occur, but compilers may mutter if there is no

44
uppsrc/plugin/pcre/lib/ucp.h
View file

@ -6,9 +6,8 @@
#define _UCP_H
/* This file contains definitions of the property values that are returned by
the function _pcre_ucp_findprop(). New values that are added for new releases
of Unicode should always be at the end of each enum, for backwards
compatibility. */
the UCD access macros. New values that are added for new releases of Unicode
should always be at the end of each enum, for backwards compatibility. */
/* These are the general character categories. */
@ -121,11 +120,40 @@ enum {
ucp_Tifinagh,
ucp_Ugaritic,
ucp_Yi,
ucp_Balinese, /* New for Unicode 5.0.0 */
ucp_Cuneiform, /* New for Unicode 5.0.0 */
ucp_Nko, /* New for Unicode 5.0.0 */
ucp_Phags_Pa, /* New for Unicode 5.0.0 */
ucp_Phoenician /* New for Unicode 5.0.0 */
/* New for Unicode 5.0: */
ucp_Balinese,
ucp_Cuneiform,
ucp_Nko,
ucp_Phags_Pa,
ucp_Phoenician,
/* New for Unicode 5.1: */
ucp_Carian,
ucp_Cham,
ucp_Kayah_Li,
ucp_Lepcha,
ucp_Lycian,
ucp_Lydian,
ucp_Ol_Chiki,
ucp_Rejang,
ucp_Saurashtra,
ucp_Sundanese,
ucp_Vai,
/* New for Unicode 5.2: */
ucp_Avestan,
ucp_Bamum,
ucp_Egyptian_Hieroglyphs,
ucp_Imperial_Aramaic,
ucp_Inscriptional_Pahlavi,
ucp_Inscriptional_Parthian,
ucp_Javanese,
ucp_Kaithi,
ucp_Lisu,
ucp_Meetei_Mayek,
ucp_Old_South_Arabian,
ucp_Old_Turkic,
ucp_Samaritan,
ucp_Tai_Tham,
ucp_Tai_Viet
};
#endif

92
uppsrc/plugin/pcre/lib/ucpinternal.h
View file

@ -1,92 +0,0 @@
/*************************************************
* Unicode Property Table handler *
*************************************************/
#ifndef _UCPINTERNAL_H
#define _UCPINTERNAL_H
/* Internal header file defining the layout of the bits in each pair of 32-bit
words that form a data item in the table. */
typedef struct cnode {
pcre_uint32 f0;
pcre_uint32 f1;
} cnode;
/* Things for the f0 field */
#define f0_scriptmask 0xff000000 /* Mask for script field */
#define f0_scriptshift 24 /* Shift for script value */
#define f0_rangeflag 0x00f00000 /* Flag for a range item */
#define f0_charmask 0x001fffff /* Mask for code point value */
/* Things for the f1 field */
#define f1_typemask 0xfc000000 /* Mask for char type field */
#define f1_typeshift 26 /* Shift for the type field */
#define f1_rangemask 0x0000ffff /* Mask for a range offset */
#define f1_casemask 0x0000ffff /* Mask for a case offset */
#define f1_caseneg 0xffff8000 /* Bits for negation */
/* The data consists of a vector of structures of type cnode. The two unsigned
32-bit integers are used as follows:
(f0) (1) The most significant byte holds the script number. The numbers are
defined by the enum in ucp.h.
(2) The 0x00800000 bit is set if this entry defines a range of characters.
It is not set if this entry defines a single character
(3) The 0x00600000 bits are spare.
(4) The 0x001fffff bits contain the code point. No Unicode code point will
ever be greater than 0x0010ffff, so this should be OK for ever.
(f1) (1) The 0xfc000000 bits contain the character type number. The numbers are
defined by an enum in ucp.h.
(2) The 0x03ff0000 bits are spare.
(3) The 0x0000ffff bits contain EITHER the unsigned offset to the top of
range if this entry defines a range, OR the *signed* offset to the
character's "other case" partner if this entry defines a single
character. There is no partner if the value is zero.
-------------------------------------------------------------------------------
| script (8) |.|.|.| codepoint (21) || type (6) |.|.| spare (8) | offset (16) |
-------------------------------------------------------------------------------
| | | | |
| | |-> spare | |-> spare
| | |
| |-> spare |-> spare
|
|-> range flag
The upper/lower casing information is set only for characters that come in
pairs. The non-one-to-one mappings in the Unicode data are ignored.
When searching the data, proceed as follows:
(1) Set up for a binary chop search.
(2) If the top is not greater than the bottom, the character is not in the
table. Its type must therefore be "Cn" ("Undefined").
(3) Find the middle vector element.
(4) Extract the code point and compare. If equal, we are done.
(5) If the test character is smaller, set the top to the current point, and
goto (2).
(6) If the current entry defines a range, compute the last character by adding
the offset, and see if the test character is within the range. If it is,
we are done.
(7) Otherwise, set the bottom to one element past the current point and goto
(2).
*/
#endif /* _UCPINTERNAL_H */
/* End of ucpinternal.h */

3088
uppsrc/plugin/pcre/lib/ucptable.h
View file

File diff suppressed because it is too large Load diff