Skip to content

G
git2
Commit 9ceafb57 by Edward Thomson
Options

regexec: use pcre as our fallback/builtin regex 

Use PCRE 8.42 as the builtin regex implementation, using its POSIX
compatibility layer.  PCRE uses ASCII by default and the users locale
will not influence its behavior, so its `regcomp` implementation is
similar to `regcomp_l` with a C locale.
parent d3a440ca
Showing with 8145 additions and 3761 deletions
deps/pcre/CMakeLists.txt 0 → 100644
INCLUDE(CheckIncludeFile)
INCLUDE(CheckFunctionExists)
INCLUDE(CheckTypeSize)
CHECK_INCLUDE_FILE(dirent.h HAVE_DIRENT_H)
CHECK_INCLUDE_FILE(stdint.h HAVE_STDINT_H)
CHECK_INCLUDE_FILE(inttypes.h HAVE_INTTYPES_H)
CHECK_INCLUDE_FILE(sys/stat.h HAVE_SYS_STAT_H)
CHECK_INCLUDE_FILE(sys/types.h HAVE_SYS_TYPES_H)
CHECK_INCLUDE_FILE(unistd.h HAVE_UNISTD_H)
CHECK_INCLUDE_FILE(windows.h HAVE_WINDOWS_H)
CHECK_FUNCTION_EXISTS(bcopy HAVE_BCOPY)
CHECK_FUNCTION_EXISTS(memmove HAVE_MEMMOVE)
CHECK_FUNCTION_EXISTS(strerror HAVE_STRERROR)
CHECK_FUNCTION_EXISTS(strtoll HAVE_STRTOLL)
CHECK_FUNCTION_EXISTS(strtoq HAVE_STRTOQ)
CHECK_FUNCTION_EXISTS(_strtoi64 HAVE__STRTOI64)
CHECK_TYPE_SIZE("long long" LONG_LONG)
CHECK_TYPE_SIZE("unsigned long long" UNSIGNED_LONG_LONG)
# User-configurable options
SET(SUPPORT_PCRE8 1)
SET(PCRE_LINK_SIZE "2")
SET(PCRE_PARENS_NEST_LIMIT "250")
SET(PCRE_MATCH_LIMIT "10000000")
SET(PCRE_MATCH_LIMIT_RECURSION "MATCH_LIMIT")
SET(PCRE_NEWLINE "LF")
SET(NO_RECURSE 1)
SET(PCRE_POSIX_MALLOC_THRESHOLD "10")
SET(BSR_ANYCRLF 0)
IF (MINGW)
OPTION(NON_STANDARD_LIB_PREFIX
"ON=Shared libraries built in mingw will be named pcre.dll, etc., instead of libpcre.dll, etc."
OFF)
OPTION(NON_STANDARD_LIB_SUFFIX
"ON=Shared libraries built in mingw will be named libpcre-0.dll, etc., instead of libpcre.dll, etc."
OFF)
ENDIF(MINGW)
# Prepare build configuration
SET(pcre_have_long_long 0)
SET(pcre_have_ulong_long 0)
IF(HAVE_LONG_LONG)
SET(pcre_have_long_long 1)
ENDIF(HAVE_LONG_LONG)
IF(HAVE_UNSIGNED_LONG_LONG)
SET(pcre_have_ulong_long 1)
ENDIF(HAVE_UNSIGNED_LONG_LONG)
SET(NEWLINE "")
IF(PCRE_NEWLINE STREQUAL "LF")
SET(NEWLINE "10")
ENDIF(PCRE_NEWLINE STREQUAL "LF")
IF(PCRE_NEWLINE STREQUAL "CR")
SET(NEWLINE "13")
ENDIF(PCRE_NEWLINE STREQUAL "CR")
IF(PCRE_NEWLINE STREQUAL "CRLF")
SET(NEWLINE "3338")
ENDIF(PCRE_NEWLINE STREQUAL "CRLF")
IF(PCRE_NEWLINE STREQUAL "ANY")
SET(NEWLINE "-1")
ENDIF(PCRE_NEWLINE STREQUAL "ANY")
IF(PCRE_NEWLINE STREQUAL "ANYCRLF")
SET(NEWLINE "-2")
ENDIF(PCRE_NEWLINE STREQUAL "ANYCRLF")
IF(NEWLINE STREQUAL "")
MESSAGE(FATAL_ERROR "The PCRE_NEWLINE variable must be set to one of the following values: \"LF\", \"CR\", \"CRLF\", \"ANY\", \"ANYCRLF\".")
ENDIF(NEWLINE STREQUAL "")
# Output files
CONFIGURE_FILE(config.h.in
${PROJECT_BINARY_DIR}/src/pcre/config.h
@ONLY)
# Source code
SET(PCRE_HEADERS ${PROJECT_BINARY_DIR}/src/pcre/config.h)
SET(PCRE_SOURCES
pcre_byte_order.c
pcre_chartables.c
pcre_compile.c
pcre_config.c
pcre_dfa_exec.c
pcre_exec.c
pcre_fullinfo.c
pcre_get.c
pcre_globals.c
pcre_jit_compile.c
pcre_maketables.c
pcre_newline.c
pcre_ord2utf8.c
pcre_refcount.c
pcre_string_utils.c
pcre_study.c
pcre_tables.c
pcre_ucd.c
pcre_valid_utf8.c
pcre_version.c
pcre_xclass.c
)
SET(PCREPOSIX_HEADERS pcreposix.h)
SET(PCREPOSIX_SOURCES pcreposix.c)
# Fix static compilation with MSVC: https://bugs.exim.org/show_bug.cgi?id=1681
# This code was taken from the CMake wiki, not from WebM.
# Build setup
ADD_DEFINITIONS(-DHAVE_CONFIG_H)
IF(MSVC)
ADD_DEFINITIONS(-D_CRT_SECURE_NO_DEPRECATE -D_CRT_SECURE_NO_WARNINGS)
ENDIF(MSVC)
SET(CMAKE_INCLUDE_CURRENT_DIR 1)
SET(targets)
# Libraries
# pcre
INCLUDE_DIRECTORIES(${PROJECT_BINARY_DIR}/src/pcre)
ADD_LIBRARY(pcre OBJECT ${PCRE_HEADERS} ${PCRE_SOURCES} ${PCREPOSIX_SOURCES})
# end CMakeLists.txt
deps/pcre/COPYING 0 → 100644
PCRE LICENCE
Please see the file LICENCE in the PCRE distribution for licensing details.
End
deps/pcre/cmake/COPYING-CMAKE-SCRIPTS 0 → 100644
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:
1. Redistributions of source code must retain the copyright
notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
3. The name of the author may not be used to endorse or promote products
derived from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.
deps/pcre/cmake/FindEditline.cmake 0 → 100644
# Modified from FindReadline.cmake (PH Feb 2012)
if(EDITLINE_INCLUDE_DIR AND EDITLINE_LIBRARY AND NCURSES_LIBRARY)
set(EDITLINE_FOUND TRUE)
else(EDITLINE_INCLUDE_DIR AND EDITLINE_LIBRARY AND NCURSES_LIBRARY)
FIND_PATH(EDITLINE_INCLUDE_DIR readline.h
/usr/include/editline
/usr/include/edit/readline
/usr/include/readline
)
FIND_LIBRARY(EDITLINE_LIBRARY NAMES edit)
include(FindPackageHandleStandardArgs)
FIND_PACKAGE_HANDLE_STANDARD_ARGS(Editline DEFAULT_MSG EDITLINE_INCLUDE_DIR EDITLINE_LIBRARY )
MARK_AS_ADVANCED(EDITLINE_INCLUDE_DIR EDITLINE_LIBRARY)
endif(EDITLINE_INCLUDE_DIR AND EDITLINE_LIBRARY AND NCURSES_LIBRARY)
deps/pcre/cmake/FindPackageHandleStandardArgs.cmake 0 → 100644
# FIND_PACKAGE_HANDLE_STANDARD_ARGS(NAME (DEFAULT_MSG|"Custom failure message") VAR1 ... )
# This macro is intended to be used in FindXXX.cmake modules files.
# It handles the REQUIRED and QUIET argument to FIND_PACKAGE() and
# it also sets the <UPPERCASED_NAME>_FOUND variable.
# The package is found if all variables listed are TRUE.
# Example:
#
# FIND_PACKAGE_HANDLE_STANDARD_ARGS(LibXml2 DEFAULT_MSG LIBXML2_LIBRARIES LIBXML2_INCLUDE_DIR)
#
# LibXml2 is considered to be found, if both LIBXML2_LIBRARIES and
# LIBXML2_INCLUDE_DIR are valid. Then also LIBXML2_FOUND is set to TRUE.
# If it is not found and REQUIRED was used, it fails with FATAL_ERROR,
# independent whether QUIET was used or not.
# If it is found, the location is reported using the VAR1 argument, so
# here a message "Found LibXml2: /usr/lib/libxml2.so" will be printed out.
# If the second argument is DEFAULT_MSG, the message in the failure case will
# be "Could NOT find LibXml2", if you don't like this message you can specify
# your own custom failure message there.
MACRO(FIND_PACKAGE_HANDLE_STANDARD_ARGS _NAME _FAIL_MSG _VAR1 )
IF("${_FAIL_MSG}" STREQUAL "DEFAULT_MSG")
IF (${_NAME}_FIND_REQUIRED)
SET(_FAIL_MESSAGE "Could not find REQUIRED package ${_NAME}")
ELSE (${_NAME}_FIND_REQUIRED)
SET(_FAIL_MESSAGE "Could not find OPTIONAL package ${_NAME}")
ENDIF (${_NAME}_FIND_REQUIRED)
ELSE("${_FAIL_MSG}" STREQUAL "DEFAULT_MSG")
SET(_FAIL_MESSAGE "${_FAIL_MSG}")
ENDIF("${_FAIL_MSG}" STREQUAL "DEFAULT_MSG")
STRING(TOUPPER ${_NAME} _NAME_UPPER)
SET(${_NAME_UPPER}_FOUND TRUE)
IF(NOT ${_VAR1})
SET(${_NAME_UPPER}_FOUND FALSE)
ENDIF(NOT ${_VAR1})
FOREACH(_CURRENT_VAR ${ARGN})
IF(NOT ${_CURRENT_VAR})
SET(${_NAME_UPPER}_FOUND FALSE)
ENDIF(NOT ${_CURRENT_VAR})
ENDFOREACH(_CURRENT_VAR)
IF (${_NAME_UPPER}_FOUND)
IF (NOT ${_NAME}_FIND_QUIETLY)
MESSAGE(STATUS "Found ${_NAME}: ${${_VAR1}}")
ENDIF (NOT ${_NAME}_FIND_QUIETLY)
ELSE (${_NAME_UPPER}_FOUND)
IF (${_NAME}_FIND_REQUIRED)
MESSAGE(FATAL_ERROR "${_FAIL_MESSAGE}")
ELSE (${_NAME}_FIND_REQUIRED)
IF (NOT ${_NAME}_FIND_QUIETLY)
MESSAGE(STATUS "${_FAIL_MESSAGE}")
ENDIF (NOT ${_NAME}_FIND_QUIETLY)
ENDIF (${_NAME}_FIND_REQUIRED)
ENDIF (${_NAME_UPPER}_FOUND)
ENDMACRO(FIND_PACKAGE_HANDLE_STANDARD_ARGS)
deps/pcre/cmake/FindReadline.cmake 0 → 100644
# from http://websvn.kde.org/trunk/KDE/kdeedu/cmake/modules/FindReadline.cmake
# http://websvn.kde.org/trunk/KDE/kdeedu/cmake/modules/COPYING-CMAKE-SCRIPTS
# --> BSD licensed
#
# GNU Readline library finder
if(READLINE_INCLUDE_DIR AND READLINE_LIBRARY AND NCURSES_LIBRARY)
set(READLINE_FOUND TRUE)
else(READLINE_INCLUDE_DIR AND READLINE_LIBRARY AND NCURSES_LIBRARY)
FIND_PATH(READLINE_INCLUDE_DIR readline/readline.h
/usr/include/readline
)
# 2008-04-22 The next clause used to read like this:
#
# FIND_LIBRARY(READLINE_LIBRARY NAMES readline)
# FIND_LIBRARY(NCURSES_LIBRARY NAMES ncurses )
# include(FindPackageHandleStandardArgs)
# FIND_PACKAGE_HANDLE_STANDARD_ARGS(Readline DEFAULT_MSG NCURSES_LIBRARY READLINE_INCLUDE_DIR READLINE_LIBRARY )
#
# I was advised to modify it such that it will find an ncurses library if
# required, but not if one was explicitly given, that is, it allows the
# default to be overridden. PH
FIND_LIBRARY(READLINE_LIBRARY NAMES readline)
include(FindPackageHandleStandardArgs)
FIND_PACKAGE_HANDLE_STANDARD_ARGS(Readline DEFAULT_MSG READLINE_INCLUDE_DIR READLINE_LIBRARY )
MARK_AS_ADVANCED(READLINE_INCLUDE_DIR READLINE_LIBRARY)
endif(READLINE_INCLUDE_DIR AND READLINE_LIBRARY AND NCURSES_LIBRARY)
deps/pcre/config.h.in 0 → 100644
/* config.h for CMake builds */
#cmakedefine HAVE_DIRENT_H 1
#cmakedefine HAVE_SYS_STAT_H 1
#cmakedefine HAVE_SYS_TYPES_H 1
#cmakedefine HAVE_UNISTD_H 1
#cmakedefine HAVE_WINDOWS_H 1
#cmakedefine HAVE_STDINT_H 1
#cmakedefine HAVE_INTTYPES_H 1
#cmakedefine HAVE_TYPE_TRAITS_H 1
#cmakedefine HAVE_BITS_TYPE_TRAITS_H 1
#cmakedefine HAVE_BCOPY 1
#cmakedefine HAVE_MEMMOVE 1
#cmakedefine HAVE_STRERROR 1
#cmakedefine HAVE_STRTOLL 1
#cmakedefine HAVE_STRTOQ 1
#cmakedefine HAVE__STRTOI64 1
#cmakedefine PCRE_STATIC 1
#cmakedefine SUPPORT_PCRE8 1
#cmakedefine SUPPORT_PCRE16 1
#cmakedefine SUPPORT_PCRE32 1
#cmakedefine SUPPORT_JIT 1
#cmakedefine SUPPORT_PCREGREP_JIT 1
#cmakedefine SUPPORT_UTF 1
#cmakedefine SUPPORT_UCP 1
#cmakedefine EBCDIC 1
#cmakedefine EBCDIC_NL25 1
#cmakedefine BSR_ANYCRLF 1
#cmakedefine NO_RECURSE 1
#cmakedefine HAVE_LONG_LONG 1
#cmakedefine HAVE_UNSIGNED_LONG_LONG 1
#cmakedefine SUPPORT_LIBBZ2 1
#cmakedefine SUPPORT_LIBZ 1
#cmakedefine SUPPORT_LIBEDIT 1
#cmakedefine SUPPORT_LIBREADLINE 1
#cmakedefine SUPPORT_VALGRIND 1
#cmakedefine SUPPORT_GCOV 1
#define NEWLINE @NEWLINE@
#define POSIX_MALLOC_THRESHOLD @PCRE_POSIX_MALLOC_THRESHOLD@
#define LINK_SIZE @PCRE_LINK_SIZE@
#define PARENS_NEST_LIMIT @PCRE_PARENS_NEST_LIMIT@
#define MATCH_LIMIT @PCRE_MATCH_LIMIT@
#define MATCH_LIMIT_RECURSION @PCRE_MATCH_LIMIT_RECURSION@
#define PCREGREP_BUFSIZE @PCREGREP_BUFSIZE@
#define MAX_NAME_SIZE 32
#define MAX_NAME_COUNT 10000
/* end config.h for CMake builds */
deps/pcre/pcre.h 0 → 100644
/*************************************************
* Perl-Compatible Regular Expressions *
*************************************************/
/* This is the public header file for the PCRE library, to be #included by
applications that call the PCRE functions.
Copyright (c) 1997-2014 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.
-----------------------------------------------------------------------------
*/
#ifndef _PCRE_H
#define _PCRE_H
/* The current PCRE version information. */
#define PCRE_MAJOR 8
#define PCRE_MINOR 42
#define PCRE_PRERELEASE
#define PCRE_DATE 2018-03-20
#define PCRE_EXP_DECL extern
/* Have to include stdlib.h in order to ensure that size_t is defined;
it is needed here for malloc. */
#include <stdlib.h>
/* Allow for C++ users */
#ifdef __cplusplus
extern "C" {
#endif
/* Public options. Some are compile-time only, some are run-time only, and some
are both. Most of the compile-time options are saved with the compiled regex so
that they can be inspected during studying (and therefore JIT compiling). Note
that pcre_study() has its own set of options. Originally, all the options
defined here used distinct bits. However, almost all the bits in a 32-bit word
are now used, so in order to conserve them, option bits that were previously
only recognized at matching time (i.e. by pcre_exec() or pcre_dfa_exec()) may
also be used for compile-time options that affect only compiling and are not
relevant for studying or JIT compiling.
Some options for pcre_compile() change its behaviour but do not affect the
behaviour of the execution functions. Other options are passed through to the
execution functions and affect their behaviour, with or without affecting the
behaviour of pcre_compile().
Options that can be passed to pcre_compile() are tagged Cx below, with these
variants:
C1 Affects compile only
C2 Does not affect compile; affects exec, dfa_exec
C3 Affects compile, exec, dfa_exec
C4 Affects compile, exec, dfa_exec, study
C5 Affects compile, exec, study
Options that can be set for pcre_exec() and/or pcre_dfa_exec() are flagged with
E and D, respectively. They take precedence over C3, C4, and C5 settings passed
from pcre_compile(). Those that are compatible with JIT execution are flagged
with J. */
#define PCRE_CASELESS 0x00000001 /* C1 */
#define PCRE_MULTILINE 0x00000002 /* C1 */
#define PCRE_DOTALL 0x00000004 /* C1 */
#define PCRE_EXTENDED 0x00000008 /* C1 */
#define PCRE_ANCHORED 0x00000010 /* C4 E D */
#define PCRE_DOLLAR_ENDONLY 0x00000020 /* C2 */
#define PCRE_EXTRA 0x00000040 /* C1 */
#define PCRE_NOTBOL 0x00000080 /* E D J */
#define PCRE_NOTEOL 0x00000100 /* E D J */
#define PCRE_UNGREEDY 0x00000200 /* C1 */
#define PCRE_NOTEMPTY 0x00000400 /* E D J */
#define PCRE_UTF8 0x00000800 /* C4 ) */
#define PCRE_UTF16 0x00000800 /* C4 ) Synonyms */
#define PCRE_UTF32 0x00000800 /* C4 ) */
#define PCRE_NO_AUTO_CAPTURE 0x00001000 /* C1 */
#define PCRE_NO_UTF8_CHECK 0x00002000 /* C1 E D J ) */
#define PCRE_NO_UTF16_CHECK 0x00002000 /* C1 E D J ) Synonyms */
#define PCRE_NO_UTF32_CHECK 0x00002000 /* C1 E D J ) */
#define PCRE_AUTO_CALLOUT 0x00004000 /* C1 */
#define PCRE_PARTIAL_SOFT 0x00008000 /* E D J ) Synonyms */
#define PCRE_PARTIAL 0x00008000 /* E D J ) */
/* This pair use the same bit. */
#define PCRE_NEVER_UTF 0x00010000 /* C1 ) Overlaid */
#define PCRE_DFA_SHORTEST 0x00010000 /* D ) Overlaid */
/* This pair use the same bit. */
#define PCRE_NO_AUTO_POSSESS 0x00020000 /* C1 ) Overlaid */
#define PCRE_DFA_RESTART 0x00020000 /* D ) Overlaid */
#define PCRE_FIRSTLINE 0x00040000 /* C3 */
#define PCRE_DUPNAMES 0x00080000 /* C1 */
#define PCRE_NEWLINE_CR 0x00100000 /* C3 E D */
#define PCRE_NEWLINE_LF 0x00200000 /* C3 E D */
#define PCRE_NEWLINE_CRLF 0x00300000 /* C3 E D */
#define PCRE_NEWLINE_ANY 0x00400000 /* C3 E D */
#define PCRE_NEWLINE_ANYCRLF 0x00500000 /* C3 E D */
#define PCRE_BSR_ANYCRLF 0x00800000 /* C3 E D */
#define PCRE_BSR_UNICODE 0x01000000 /* C3 E D */
#define PCRE_JAVASCRIPT_COMPAT 0x02000000 /* C5 */
#define PCRE_NO_START_OPTIMIZE 0x04000000 /* C2 E D ) Synonyms */
#define PCRE_NO_START_OPTIMISE 0x04000000 /* C2 E D ) */
#define PCRE_PARTIAL_HARD 0x08000000 /* E D J */
#define PCRE_NOTEMPTY_ATSTART 0x10000000 /* E D J */
#define PCRE_UCP 0x20000000 /* C3 */
/* Exec-time and get/set-time error codes */
#define PCRE_ERROR_NOMATCH (-1)
#define PCRE_ERROR_NULL (-2)
#define PCRE_ERROR_BADOPTION (-3)
#define PCRE_ERROR_BADMAGIC (-4)
#define PCRE_ERROR_UNKNOWN_OPCODE (-5)
#define PCRE_ERROR_UNKNOWN_NODE (-5) /* For backward compatibility */
#define PCRE_ERROR_NOMEMORY (-6)
#define PCRE_ERROR_NOSUBSTRING (-7)
#define PCRE_ERROR_MATCHLIMIT (-8)
#define PCRE_ERROR_CALLOUT (-9) /* Never used by PCRE itself */
#define PCRE_ERROR_BADUTF8 (-10) /* Same for 8/16/32 */
#define PCRE_ERROR_BADUTF16 (-10) /* Same for 8/16/32 */
#define PCRE_ERROR_BADUTF32 (-10) /* Same for 8/16/32 */
#define PCRE_ERROR_BADUTF8_OFFSET (-11) /* Same for 8/16 */
#define PCRE_ERROR_BADUTF16_OFFSET (-11) /* Same for 8/16 */
#define PCRE_ERROR_PARTIAL (-12)
#define PCRE_ERROR_BADPARTIAL (-13)
#define PCRE_ERROR_INTERNAL (-14)
#define PCRE_ERROR_BADCOUNT (-15)
#define PCRE_ERROR_DFA_UITEM (-16)
#define PCRE_ERROR_DFA_UCOND (-17)
#define PCRE_ERROR_DFA_UMLIMIT (-18)
#define PCRE_ERROR_DFA_WSSIZE (-19)
#define PCRE_ERROR_DFA_RECURSE (-20)
#define PCRE_ERROR_RECURSIONLIMIT (-21)
#define PCRE_ERROR_NULLWSLIMIT (-22) /* No longer actually used */
#define PCRE_ERROR_BADNEWLINE (-23)
#define PCRE_ERROR_BADOFFSET (-24)
#define PCRE_ERROR_SHORTUTF8 (-25)
#define PCRE_ERROR_SHORTUTF16 (-25) /* Same for 8/16 */
#define PCRE_ERROR_RECURSELOOP (-26)
#define PCRE_ERROR_JIT_STACKLIMIT (-27)
#define PCRE_ERROR_BADMODE (-28)
#define PCRE_ERROR_BADENDIANNESS (-29)
#define PCRE_ERROR_DFA_BADRESTART (-30)
#define PCRE_ERROR_JIT_BADOPTION (-31)
#define PCRE_ERROR_BADLENGTH (-32)
#define PCRE_ERROR_UNSET (-33)
/* Specific error codes for UTF-8 validity checks */
#define PCRE_UTF8_ERR0 0
#define PCRE_UTF8_ERR1 1
#define PCRE_UTF8_ERR2 2
#define PCRE_UTF8_ERR3 3
#define PCRE_UTF8_ERR4 4
#define PCRE_UTF8_ERR5 5
#define PCRE_UTF8_ERR6 6
#define PCRE_UTF8_ERR7 7
#define PCRE_UTF8_ERR8 8
#define PCRE_UTF8_ERR9 9
#define PCRE_UTF8_ERR10 10
#define PCRE_UTF8_ERR11 11
#define PCRE_UTF8_ERR12 12
#define PCRE_UTF8_ERR13 13
#define PCRE_UTF8_ERR14 14
#define PCRE_UTF8_ERR15 15
#define PCRE_UTF8_ERR16 16
#define PCRE_UTF8_ERR17 17
#define PCRE_UTF8_ERR18 18
#define PCRE_UTF8_ERR19 19
#define PCRE_UTF8_ERR20 20
#define PCRE_UTF8_ERR21 21
#define PCRE_UTF8_ERR22 22 /* Unused (was non-character) */
/* Specific error codes for UTF-16 validity checks */
#define PCRE_UTF16_ERR0 0
#define PCRE_UTF16_ERR1 1
#define PCRE_UTF16_ERR2 2
#define PCRE_UTF16_ERR3 3
#define PCRE_UTF16_ERR4 4 /* Unused (was non-character) */
/* Specific error codes for UTF-32 validity checks */
#define PCRE_UTF32_ERR0 0
#define PCRE_UTF32_ERR1 1
#define PCRE_UTF32_ERR2 2 /* Unused (was non-character) */
#define PCRE_UTF32_ERR3 3
/* Request types for pcre_fullinfo() */
#define PCRE_INFO_OPTIONS 0
#define PCRE_INFO_SIZE 1
#define PCRE_INFO_CAPTURECOUNT 2
#define PCRE_INFO_BACKREFMAX 3
#define PCRE_INFO_FIRSTBYTE 4
#define PCRE_INFO_FIRSTCHAR 4 /* For backwards compatibility */
#define PCRE_INFO_FIRSTTABLE 5
#define PCRE_INFO_LASTLITERAL 6
#define PCRE_INFO_NAMEENTRYSIZE 7
#define PCRE_INFO_NAMECOUNT 8
#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
#define PCRE_INFO_HASCRORLF 14
#define PCRE_INFO_MINLENGTH 15
#define PCRE_INFO_JIT 16
#define PCRE_INFO_JITSIZE 17
#define PCRE_INFO_MAXLOOKBEHIND 18
#define PCRE_INFO_FIRSTCHARACTER 19
#define PCRE_INFO_FIRSTCHARACTERFLAGS 20
#define PCRE_INFO_REQUIREDCHAR 21
#define PCRE_INFO_REQUIREDCHARFLAGS 22
#define PCRE_INFO_MATCHLIMIT 23
#define PCRE_INFO_RECURSIONLIMIT 24
#define PCRE_INFO_MATCH_EMPTY 25
/* Request types for pcre_config(). Do not re-arrange, in order to remain
compatible. */
#define PCRE_CONFIG_UTF8 0
#define PCRE_CONFIG_NEWLINE 1
#define PCRE_CONFIG_LINK_SIZE 2
#define PCRE_CONFIG_POSIX_MALLOC_THRESHOLD 3
#define PCRE_CONFIG_MATCH_LIMIT 4
#define PCRE_CONFIG_STACKRECURSE 5
#define PCRE_CONFIG_UNICODE_PROPERTIES 6
#define PCRE_CONFIG_MATCH_LIMIT_RECURSION 7
#define PCRE_CONFIG_BSR 8
#define PCRE_CONFIG_JIT 9
#define PCRE_CONFIG_UTF16 10
#define PCRE_CONFIG_JITTARGET 11
#define PCRE_CONFIG_UTF32 12
#define PCRE_CONFIG_PARENS_LIMIT 13
/* Request types for pcre_study(). Do not re-arrange, in order to remain
compatible. */
#define PCRE_STUDY_JIT_COMPILE 0x0001
#define PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE 0x0002
#define PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE 0x0004
#define PCRE_STUDY_EXTRA_NEEDED 0x0008
/* Bit flags for the pcre[16|32]_extra structure. Do not re-arrange or redefine
these bits, just add new ones on the end, in order to remain compatible. */
#define PCRE_EXTRA_STUDY_DATA 0x0001
#define PCRE_EXTRA_MATCH_LIMIT 0x0002
#define PCRE_EXTRA_CALLOUT_DATA 0x0004
#define PCRE_EXTRA_TABLES 0x0008
#define PCRE_EXTRA_MATCH_LIMIT_RECURSION 0x0010
#define PCRE_EXTRA_MARK 0x0020
#define PCRE_EXTRA_EXECUTABLE_JIT 0x0040
/* Types */
struct real_pcre8_or_16; /* declaration; the definition is private */
typedef struct real_pcre8_or_16 pcre;
struct real_pcre8_or_16; /* declaration; the definition is private */
typedef struct real_pcre8_or_16 pcre16;
struct real_pcre32; /* declaration; the definition is private */
typedef struct real_pcre32 pcre32;
struct real_pcre_jit_stack; /* declaration; the definition is private */
typedef struct real_pcre_jit_stack pcre_jit_stack;
struct real_pcre16_jit_stack; /* declaration; the definition is private */
typedef struct real_pcre16_jit_stack pcre16_jit_stack;
struct real_pcre32_jit_stack; /* declaration; the definition is private */
typedef struct real_pcre32_jit_stack pcre32_jit_stack;
/* If PCRE is compiled with 16 bit character support, PCRE_UCHAR16 must contain
a 16 bit wide signed data type. Otherwise it can be a dummy data type since
pcre16 functions are not implemented. There is a check for this in pcre_internal.h. */
#ifndef PCRE_UCHAR16
#define PCRE_UCHAR16 unsigned short
#endif
#ifndef PCRE_SPTR16
#define PCRE_SPTR16 const PCRE_UCHAR16 *
#endif
/* If PCRE is compiled with 32 bit character support, PCRE_UCHAR32 must contain
a 32 bit wide signed data type. Otherwise it can be a dummy data type since
pcre32 functions are not implemented. There is a check for this in pcre_internal.h. */
#ifndef PCRE_UCHAR32
#define PCRE_UCHAR32 unsigned int
#endif
#ifndef PCRE_SPTR32
#define PCRE_SPTR32 const PCRE_UCHAR32 *
#endif
/* When PCRE is compiled as a C++ library, the subject pointer type can be
replaced with a custom type. For conventional use, the public interface is a
const char *. */
#ifndef PCRE_SPTR
#define PCRE_SPTR const char *
#endif
/* The structure for passing additional data to pcre_exec(). This is defined in
such as way as to be extensible. Always add new fields at the end, in order to
remain compatible. */
typedef struct pcre_extra {
unsigned long int flags; /* Bits for which fields are set */
void *study_data; /* Opaque data from pcre_study() */
unsigned long int match_limit; /* Maximum number of calls to match() */
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 */
void *executable_jit; /* Contains a pointer to a compiled jit code */
} pcre_extra;
/* Same structure as above, but with 16 bit char pointers. */
typedef struct pcre16_extra {
unsigned long int flags; /* Bits for which fields are set */
void *study_data; /* Opaque data from pcre_study() */
unsigned long int match_limit; /* Maximum number of calls to match() */
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() */
PCRE_UCHAR16 **mark; /* For passing back a mark pointer */
void *executable_jit; /* Contains a pointer to a compiled jit code */
} pcre16_extra;
/* Same structure as above, but with 32 bit char pointers. */
typedef struct pcre32_extra {
unsigned long int flags; /* Bits for which fields are set */
void *study_data; /* Opaque data from pcre_study() */
unsigned long int match_limit; /* Maximum number of calls to match() */
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() */
PCRE_UCHAR32 **mark; /* For passing back a mark pointer */
void *executable_jit; /* Contains a pointer to a compiled jit code */
} pcre32_extra;
/* The structure for passing out data via the pcre_callout_function. We use a
structure so that new fields can be added on the end in future versions,
without changing the API of the function, thereby allowing old clients to work
without modification. */
typedef struct pcre_callout_block {
int version; /* Identifies version of block */
/* ------------------------ Version 0 ------------------------------- */
int callout_number; /* Number compiled into pattern */
int *offset_vector; /* The offset vector */
PCRE_SPTR subject; /* The subject being matched */
int subject_length; /* The length of the subject */
int start_match; /* Offset to start of this match attempt */
int current_position; /* Where we currently are in the subject */
int capture_top; /* Max current capture */
int capture_last; /* Most recently closed capture */
void *callout_data; /* Data passed in with the call */
/* ------------------- Added for Version 1 -------------------------- */
int pattern_position; /* Offset to next item in the pattern */
int next_item_length; /* Length of next item in the pattern */
/* ------------------- Added for Version 2 -------------------------- */
const unsigned char *mark; /* Pointer to current mark or NULL */
/* ------------------------------------------------------------------ */
} pcre_callout_block;
/* Same structure as above, but with 16 bit char pointers. */
typedef struct pcre16_callout_block {
int version; /* Identifies version of block */
/* ------------------------ Version 0 ------------------------------- */
int callout_number; /* Number compiled into pattern */
int *offset_vector; /* The offset vector */
PCRE_SPTR16 subject; /* The subject being matched */
int subject_length; /* The length of the subject */
int start_match; /* Offset to start of this match attempt */
int current_position; /* Where we currently are in the subject */
int capture_top; /* Max current capture */
int capture_last; /* Most recently closed capture */
void *callout_data; /* Data passed in with the call */
/* ------------------- Added for Version 1 -------------------------- */
int pattern_position; /* Offset to next item in the pattern */
int next_item_length; /* Length of next item in the pattern */
/* ------------------- Added for Version 2 -------------------------- */
const PCRE_UCHAR16 *mark; /* Pointer to current mark or NULL */
/* ------------------------------------------------------------------ */
} pcre16_callout_block;
/* Same structure as above, but with 32 bit char pointers. */
typedef struct pcre32_callout_block {
int version; /* Identifies version of block */
/* ------------------------ Version 0 ------------------------------- */
int callout_number; /* Number compiled into pattern */
int *offset_vector; /* The offset vector */
PCRE_SPTR32 subject; /* The subject being matched */
int subject_length; /* The length of the subject */
int start_match; /* Offset to start of this match attempt */
int current_position; /* Where we currently are in the subject */
int capture_top; /* Max current capture */
int capture_last; /* Most recently closed capture */
void *callout_data; /* Data passed in with the call */
/* ------------------- Added for Version 1 -------------------------- */
int pattern_position; /* Offset to next item in the pattern */
int next_item_length; /* Length of next item in the pattern */
/* ------------------- Added for Version 2 -------------------------- */
const PCRE_UCHAR32 *mark; /* Pointer to current mark or NULL */
/* ------------------------------------------------------------------ */
} pcre32_callout_block;
/* Indirection for store get and free functions. These can be set to
alternative malloc/free functions if required. Special ones are used in the
non-recursive case for "frames". There is also an optional callout function
that is triggered by the (?) regex item. For Virtual Pascal, these definitions
have to take another form. */
#ifndef VPCOMPAT
PCRE_EXP_DECL void *(*pcre_malloc)(size_t);
PCRE_EXP_DECL void (*pcre_free)(void *);
PCRE_EXP_DECL void *(*pcre_stack_malloc)(size_t);
PCRE_EXP_DECL void (*pcre_stack_free)(void *);
PCRE_EXP_DECL int (*pcre_callout)(pcre_callout_block *);
PCRE_EXP_DECL int (*pcre_stack_guard)(void);
PCRE_EXP_DECL void *(*pcre16_malloc)(size_t);
PCRE_EXP_DECL void (*pcre16_free)(void *);
PCRE_EXP_DECL void *(*pcre16_stack_malloc)(size_t);
PCRE_EXP_DECL void (*pcre16_stack_free)(void *);
PCRE_EXP_DECL int (*pcre16_callout)(pcre16_callout_block *);
PCRE_EXP_DECL int (*pcre16_stack_guard)(void);
PCRE_EXP_DECL void *(*pcre32_malloc)(size_t);
PCRE_EXP_DECL void (*pcre32_free)(void *);
PCRE_EXP_DECL void *(*pcre32_stack_malloc)(size_t);
PCRE_EXP_DECL void (*pcre32_stack_free)(void *);
PCRE_EXP_DECL int (*pcre32_callout)(pcre32_callout_block *);
PCRE_EXP_DECL int (*pcre32_stack_guard)(void);
#else /* VPCOMPAT */
PCRE_EXP_DECL void *pcre_malloc(size_t);
PCRE_EXP_DECL void pcre_free(void *);
PCRE_EXP_DECL void *pcre_stack_malloc(size_t);
PCRE_EXP_DECL void pcre_stack_free(void *);
PCRE_EXP_DECL int pcre_callout(pcre_callout_block *);
PCRE_EXP_DECL int pcre_stack_guard(void);
PCRE_EXP_DECL void *pcre16_malloc(size_t);
PCRE_EXP_DECL void pcre16_free(void *);
PCRE_EXP_DECL void *pcre16_stack_malloc(size_t);
PCRE_EXP_DECL void pcre16_stack_free(void *);
PCRE_EXP_DECL int pcre16_callout(pcre16_callout_block *);
PCRE_EXP_DECL int pcre16_stack_guard(void);
PCRE_EXP_DECL void *pcre32_malloc(size_t);
PCRE_EXP_DECL void pcre32_free(void *);
PCRE_EXP_DECL void *pcre32_stack_malloc(size_t);
PCRE_EXP_DECL void pcre32_stack_free(void *);
PCRE_EXP_DECL int pcre32_callout(pcre32_callout_block *);
PCRE_EXP_DECL int pcre32_stack_guard(void);
#endif /* VPCOMPAT */
/* User defined callback which provides a stack just before the match starts. */
typedef pcre_jit_stack *(*pcre_jit_callback)(void *);
typedef pcre16_jit_stack *(*pcre16_jit_callback)(void *);
typedef pcre32_jit_stack *(*pcre32_jit_callback)(void *);
/* Exported PCRE functions */
PCRE_EXP_DECL pcre *pcre_compile(const char *, int, const char **, int *,
const unsigned char *);
PCRE_EXP_DECL pcre16 *pcre16_compile(PCRE_SPTR16, int, const char **, int *,
const unsigned char *);
PCRE_EXP_DECL pcre32 *pcre32_compile(PCRE_SPTR32, int, const char **, int *,
const unsigned char *);
PCRE_EXP_DECL pcre *pcre_compile2(const char *, int, int *, const char **,
int *, const unsigned char *);
PCRE_EXP_DECL pcre16 *pcre16_compile2(PCRE_SPTR16, int, int *, const char **,
int *, const unsigned char *);
PCRE_EXP_DECL pcre32 *pcre32_compile2(PCRE_SPTR32, int, int *, const char **,
int *, const unsigned char *);
PCRE_EXP_DECL int pcre_config(int, void *);
PCRE_EXP_DECL int pcre16_config(int, void *);
PCRE_EXP_DECL int pcre32_config(int, void *);
PCRE_EXP_DECL int pcre_copy_named_substring(const pcre *, const char *,
int *, int, const char *, char *, int);
PCRE_EXP_DECL int pcre16_copy_named_substring(const pcre16 *, PCRE_SPTR16,
int *, int, PCRE_SPTR16, PCRE_UCHAR16 *, int);
PCRE_EXP_DECL int pcre32_copy_named_substring(const pcre32 *, PCRE_SPTR32,
int *, int, PCRE_SPTR32, PCRE_UCHAR32 *, int);
PCRE_EXP_DECL int pcre_copy_substring(const char *, int *, int, int,
char *, int);
PCRE_EXP_DECL int pcre16_copy_substring(PCRE_SPTR16, int *, int, int,
PCRE_UCHAR16 *, int);
PCRE_EXP_DECL int pcre32_copy_substring(PCRE_SPTR32, int *, int, int,
PCRE_UCHAR32 *, int);
PCRE_EXP_DECL int pcre_dfa_exec(const pcre *, const pcre_extra *,
const char *, int, int, int, int *, int , int *, int);
PCRE_EXP_DECL int pcre16_dfa_exec(const pcre16 *, const pcre16_extra *,
PCRE_SPTR16, int, int, int, int *, int , int *, int);
PCRE_EXP_DECL int pcre32_dfa_exec(const pcre32 *, const pcre32_extra *,
PCRE_SPTR32, int, int, int, int *, int , int *, int);
PCRE_EXP_DECL int pcre_exec(const pcre *, const pcre_extra *, PCRE_SPTR,
int, int, int, int *, int);
PCRE_EXP_DECL int pcre16_exec(const pcre16 *, const pcre16_extra *,
PCRE_SPTR16, int, int, int, int *, int);
PCRE_EXP_DECL int pcre32_exec(const pcre32 *, const pcre32_extra *,
PCRE_SPTR32, int, int, int, int *, int);
PCRE_EXP_DECL int pcre_jit_exec(const pcre *, const pcre_extra *,
PCRE_SPTR, int, int, int, int *, int,
pcre_jit_stack *);
PCRE_EXP_DECL int pcre16_jit_exec(const pcre16 *, const pcre16_extra *,
PCRE_SPTR16, int, int, int, int *, int,
pcre16_jit_stack *);
PCRE_EXP_DECL int pcre32_jit_exec(const pcre32 *, const pcre32_extra *,
PCRE_SPTR32, int, int, int, int *, int,
pcre32_jit_stack *);
PCRE_EXP_DECL void pcre_free_substring(const char *);
PCRE_EXP_DECL void pcre16_free_substring(PCRE_SPTR16);
PCRE_EXP_DECL void pcre32_free_substring(PCRE_SPTR32);
PCRE_EXP_DECL void pcre_free_substring_list(const char **);
PCRE_EXP_DECL void pcre16_free_substring_list(PCRE_SPTR16 *);
PCRE_EXP_DECL void pcre32_free_substring_list(PCRE_SPTR32 *);
PCRE_EXP_DECL int pcre_fullinfo(const pcre *, const pcre_extra *, int,
void *);
PCRE_EXP_DECL int pcre16_fullinfo(const pcre16 *, const pcre16_extra *, int,
void *);
PCRE_EXP_DECL int pcre32_fullinfo(const pcre32 *, const pcre32_extra *, int,
void *);
PCRE_EXP_DECL int pcre_get_named_substring(const pcre *, const char *,
int *, int, const char *, const char **);
PCRE_EXP_DECL int pcre16_get_named_substring(const pcre16 *, PCRE_SPTR16,
int *, int, PCRE_SPTR16, PCRE_SPTR16 *);
PCRE_EXP_DECL int pcre32_get_named_substring(const pcre32 *, PCRE_SPTR32,
int *, int, PCRE_SPTR32, PCRE_SPTR32 *);
PCRE_EXP_DECL int pcre_get_stringnumber(const pcre *, const char *);
PCRE_EXP_DECL int pcre16_get_stringnumber(const pcre16 *, PCRE_SPTR16);
PCRE_EXP_DECL int pcre32_get_stringnumber(const pcre32 *, PCRE_SPTR32);
PCRE_EXP_DECL int pcre_get_stringtable_entries(const pcre *, const char *,
char **, char **);
PCRE_EXP_DECL int pcre16_get_stringtable_entries(const pcre16 *, PCRE_SPTR16,
PCRE_UCHAR16 **, PCRE_UCHAR16 **);
PCRE_EXP_DECL int pcre32_get_stringtable_entries(const pcre32 *, PCRE_SPTR32,
PCRE_UCHAR32 **, PCRE_UCHAR32 **);
PCRE_EXP_DECL int pcre_get_substring(const char *, int *, int, int,
const char **);
PCRE_EXP_DECL int pcre16_get_substring(PCRE_SPTR16, int *, int, int,
PCRE_SPTR16 *);
PCRE_EXP_DECL int pcre32_get_substring(PCRE_SPTR32, int *, int, int,
PCRE_SPTR32 *);
PCRE_EXP_DECL int pcre_get_substring_list(const char *, int *, int,
const char ***);
PCRE_EXP_DECL int pcre16_get_substring_list(PCRE_SPTR16, int *, int,
PCRE_SPTR16 **);
PCRE_EXP_DECL int pcre32_get_substring_list(PCRE_SPTR32, int *, int,
PCRE_SPTR32 **);
PCRE_EXP_DECL const unsigned char *pcre_maketables(void);
PCRE_EXP_DECL const unsigned char *pcre16_maketables(void);
PCRE_EXP_DECL const unsigned char *pcre32_maketables(void);
PCRE_EXP_DECL int pcre_refcount(pcre *, int);
PCRE_EXP_DECL int pcre16_refcount(pcre16 *, int);
PCRE_EXP_DECL int pcre32_refcount(pcre32 *, int);
PCRE_EXP_DECL pcre_extra *pcre_study(const pcre *, int, const char **);
PCRE_EXP_DECL pcre16_extra *pcre16_study(const pcre16 *, int, const char **);
PCRE_EXP_DECL pcre32_extra *pcre32_study(const pcre32 *, int, const char **);
PCRE_EXP_DECL void pcre_free_study(pcre_extra *);
PCRE_EXP_DECL void pcre16_free_study(pcre16_extra *);
PCRE_EXP_DECL void pcre32_free_study(pcre32_extra *);
PCRE_EXP_DECL const char *pcre_version(void);
PCRE_EXP_DECL const char *pcre16_version(void);
PCRE_EXP_DECL const char *pcre32_version(void);
/* Utility functions for byte order swaps. */
PCRE_EXP_DECL int pcre_pattern_to_host_byte_order(pcre *, pcre_extra *,
const unsigned char *);
PCRE_EXP_DECL int pcre16_pattern_to_host_byte_order(pcre16 *, pcre16_extra *,
const unsigned char *);
PCRE_EXP_DECL int pcre32_pattern_to_host_byte_order(pcre32 *, pcre32_extra *,
const unsigned char *);
PCRE_EXP_DECL int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *,
PCRE_SPTR16, int, int *, int);
PCRE_EXP_DECL int pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *,
PCRE_SPTR32, int, int *, int);
/* JIT compiler related functions. */
PCRE_EXP_DECL pcre_jit_stack *pcre_jit_stack_alloc(int, int);
PCRE_EXP_DECL pcre16_jit_stack *pcre16_jit_stack_alloc(int, int);
PCRE_EXP_DECL pcre32_jit_stack *pcre32_jit_stack_alloc(int, int);
PCRE_EXP_DECL void pcre_jit_stack_free(pcre_jit_stack *);
PCRE_EXP_DECL void pcre16_jit_stack_free(pcre16_jit_stack *);
PCRE_EXP_DECL void pcre32_jit_stack_free(pcre32_jit_stack *);
PCRE_EXP_DECL void pcre_assign_jit_stack(pcre_extra *,
pcre_jit_callback, void *);
PCRE_EXP_DECL void pcre16_assign_jit_stack(pcre16_extra *,
pcre16_jit_callback, void *);
PCRE_EXP_DECL void pcre32_assign_jit_stack(pcre32_extra *,
pcre32_jit_callback, void *);
PCRE_EXP_DECL void pcre_jit_free_unused_memory(void);
PCRE_EXP_DECL void pcre16_jit_free_unused_memory(void);
PCRE_EXP_DECL void pcre32_jit_free_unused_memory(void);
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* End of pcre.h */
deps/pcre/pcre_byte_order.c 0 → 100644
/*************************************************
* 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) 1997-2014 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 an internal function that tests a compiled pattern to
see if it was compiled with the opposite endianness. If so, it uses an
auxiliary local function to flip the appropriate bytes. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
/*************************************************
* Swap byte functions *
*************************************************/
/* The following functions swap the bytes of a pcre_uint16
and pcre_uint32 value.
Arguments:
value any number
Returns: the byte swapped value
*/
static pcre_uint32
swap_uint32(pcre_uint32 value)
{
return ((value & 0x000000ff) << 24) |
((value & 0x0000ff00) << 8) |
((value & 0x00ff0000) >> 8) |
(value >> 24);
}
static pcre_uint16
swap_uint16(pcre_uint16 value)
{
return (value >> 8) | (value << 8);
}
/*************************************************
* Test for a byte-flipped compiled regex *
*************************************************/
/* This function swaps the bytes of a compiled pattern usually
loaded form the disk. It also sets the tables pointer, which
is likely an invalid pointer after reload.
Arguments:
argument_re points to the compiled expression
extra_data points to extra data or is NULL
tables points to the character tables or NULL
Returns: 0 if the swap is successful, negative on error
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DECL int pcre_pattern_to_host_byte_order(pcre *argument_re,
pcre_extra *extra_data, const unsigned char *tables)
#elif defined COMPILE_PCRE16
PCRE_EXP_DECL int pcre16_pattern_to_host_byte_order(pcre16 *argument_re,
pcre16_extra *extra_data, const unsigned char *tables)
#elif defined COMPILE_PCRE32
PCRE_EXP_DECL int pcre32_pattern_to_host_byte_order(pcre32 *argument_re,
pcre32_extra *extra_data, const unsigned char *tables)
#endif
{
REAL_PCRE *re = (REAL_PCRE *)argument_re;
pcre_study_data *study;
#ifndef COMPILE_PCRE8
pcre_uchar *ptr;
int length;
#if defined SUPPORT_UTF && defined COMPILE_PCRE16
BOOL utf;
BOOL utf16_char;
#endif /* SUPPORT_UTF && COMPILE_PCRE16 */
#endif /* !COMPILE_PCRE8 */
if (re == NULL) return PCRE_ERROR_NULL;
if (re->magic_number == MAGIC_NUMBER)
{
if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
re->tables = tables;
return 0;
}
if (re->magic_number != REVERSED_MAGIC_NUMBER) return PCRE_ERROR_BADMAGIC;
if ((swap_uint32(re->flags) & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
re->magic_number = MAGIC_NUMBER;
re->size = swap_uint32(re->size);
re->options = swap_uint32(re->options);
re->flags = swap_uint32(re->flags);
re->limit_match = swap_uint32(re->limit_match);
re->limit_recursion = swap_uint32(re->limit_recursion);
#if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
re->first_char = swap_uint16(re->first_char);
re->req_char = swap_uint16(re->req_char);
#elif defined COMPILE_PCRE32
re->first_char = swap_uint32(re->first_char);
re->req_char = swap_uint32(re->req_char);
#endif
re->max_lookbehind = swap_uint16(re->max_lookbehind);
re->top_bracket = swap_uint16(re->top_bracket);
re->top_backref = swap_uint16(re->top_backref);
re->name_table_offset = swap_uint16(re->name_table_offset);
re->name_entry_size = swap_uint16(re->name_entry_size);
re->name_count = swap_uint16(re->name_count);
re->ref_count = swap_uint16(re->ref_count);
re->tables = tables;
if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_STUDY_DATA) != 0)
{
study = (pcre_study_data *)extra_data->study_data;
study->size = swap_uint32(study->size);
study->flags = swap_uint32(study->flags);
study->minlength = swap_uint32(study->minlength);
}
#ifndef COMPILE_PCRE8
ptr = (pcre_uchar *)re + re->name_table_offset;
length = re->name_count * re->name_entry_size;
#if defined SUPPORT_UTF && defined COMPILE_PCRE16
utf = (re->options & PCRE_UTF16) != 0;
utf16_char = FALSE;
#endif /* SUPPORT_UTF && COMPILE_PCRE16 */
while(TRUE)
{
/* Swap previous characters. */
while (length-- > 0)
{
#if defined COMPILE_PCRE16
*ptr = swap_uint16(*ptr);
#elif defined COMPILE_PCRE32
*ptr = swap_uint32(*ptr);
#endif
ptr++;
}
#if defined SUPPORT_UTF && defined COMPILE_PCRE16
if (utf16_char)
{
if (HAS_EXTRALEN(ptr[-1]))
{
/* We know that there is only one extra character in UTF-16. */
*ptr = swap_uint16(*ptr);
ptr++;
}
}
utf16_char = FALSE;
#endif /* SUPPORT_UTF */
/* Get next opcode. */
length = 0;
#if defined COMPILE_PCRE16
*ptr = swap_uint16(*ptr);
#elif defined COMPILE_PCRE32
*ptr = swap_uint32(*ptr);
#endif
switch (*ptr)
{
case OP_END:
return 0;
#if defined SUPPORT_UTF && defined COMPILE_PCRE16
case OP_CHAR:
case OP_CHARI:
case OP_NOT:
case OP_NOTI:
case OP_STAR:
case OP_MINSTAR:
case OP_PLUS:
case OP_MINPLUS:
case OP_QUERY:
case OP_MINQUERY:
case OP_UPTO:
case OP_MINUPTO:
case OP_EXACT:
case OP_POSSTAR:
case OP_POSPLUS:
case OP_POSQUERY:
case OP_POSUPTO:
case OP_STARI:
case OP_MINSTARI:
case OP_PLUSI:
case OP_MINPLUSI:
case OP_QUERYI:
case OP_MINQUERYI:
case OP_UPTOI:
case OP_MINUPTOI:
case OP_EXACTI:
case OP_POSSTARI:
case OP_POSPLUSI:
case OP_POSQUERYI:
case OP_POSUPTOI:
case OP_NOTSTAR:
case OP_NOTMINSTAR:
case OP_NOTPLUS:
case OP_NOTMINPLUS:
case OP_NOTQUERY:
case OP_NOTMINQUERY:
case OP_NOTUPTO:
case OP_NOTMINUPTO:
case OP_NOTEXACT:
case OP_NOTPOSSTAR:
case OP_NOTPOSPLUS:
case OP_NOTPOSQUERY:
case OP_NOTPOSUPTO:
case OP_NOTSTARI:
case OP_NOTMINSTARI:
case OP_NOTPLUSI:
case OP_NOTMINPLUSI:
case OP_NOTQUERYI:
case OP_NOTMINQUERYI:
case OP_NOTUPTOI:
case OP_NOTMINUPTOI:
case OP_NOTEXACTI:
case OP_NOTPOSSTARI:
case OP_NOTPOSPLUSI:
case OP_NOTPOSQUERYI:
case OP_NOTPOSUPTOI:
if (utf) utf16_char = TRUE;
#endif
/* Fall through. */
default:
length = PRIV(OP_lengths)[*ptr] - 1;
break;
case OP_CLASS:
case OP_NCLASS:
/* Skip the character bit map. */
ptr += 32/sizeof(pcre_uchar);
length = 0;
break;
case OP_XCLASS:
/* Reverse the size of the XCLASS instance. */
ptr++;
#if defined COMPILE_PCRE16
*ptr = swap_uint16(*ptr);
#elif defined COMPILE_PCRE32
*ptr = swap_uint32(*ptr);
#endif
#ifndef COMPILE_PCRE32
if (LINK_SIZE > 1)
{
/* LINK_SIZE can be 1 or 2 in 16 bit mode. */
ptr++;
*ptr = swap_uint16(*ptr);
}
#endif
ptr++;
length = (GET(ptr, -LINK_SIZE)) - (1 + LINK_SIZE + 1);
#if defined COMPILE_PCRE16
*ptr = swap_uint16(*ptr);
#elif defined COMPILE_PCRE32
*ptr = swap_uint32(*ptr);
#endif
if ((*ptr & XCL_MAP) != 0)
{
/* Skip the character bit map. */
ptr += 32/sizeof(pcre_uchar);
length -= 32/sizeof(pcre_uchar);
}
break;
}
ptr++;
}
/* Control should never reach here in 16/32 bit mode. */
#else /* In 8-bit mode, the pattern does not need to be processed. */
return 0;
#endif /* !COMPILE_PCRE8 */
}
/* End of pcre_byte_order.c */
deps/pcre/pcre_chartables.c 0 → 100644
/*************************************************
* Perl-Compatible Regular Expressions *
*************************************************/
/* 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 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 #includes are present because without them 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. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
const pcre_uint8 PRIV(default_tables)[] = {
/* This table is a lower casing table. */
0, 1, 2, 3, 4, 5, 6, 7,
8, 9, 10, 11, 12, 13, 14, 15,
16, 17, 18, 19, 20, 21, 22, 23,
24, 25, 26, 27, 28, 29, 30, 31,
32, 33, 34, 35, 36, 37, 38, 39,
40, 41, 42, 43, 44, 45, 46, 47,
48, 49, 50, 51, 52, 53, 54, 55,
56, 57, 58, 59, 60, 61, 62, 63,
64, 97, 98, 99,100,101,102,103,
104,105,106,107,108,109,110,111,
112,113,114,115,116,117,118,119,
120,121,122, 91, 92, 93, 94, 95,
96, 97, 98, 99,100,101,102,103,
104,105,106,107,108,109,110,111,
112,113,114,115,116,117,118,119,
120,121,122,123,124,125,126,127,
128,129,130,131,132,133,134,135,
136,137,138,139,140,141,142,143,
144,145,146,147,148,149,150,151,
152,153,154,155,156,157,158,159,
160,161,162,163,164,165,166,167,
168,169,170,171,172,173,174,175,
176,177,178,179,180,181,182,183,
184,185,186,187,188,189,190,191,
192,193,194,195,196,197,198,199,
200,201,202,203,204,205,206,207,
208,209,210,211,212,213,214,215,
216,217,218,219,220,221,222,223,
224,225,226,227,228,229,230,231,
232,233,234,235,236,237,238,239,
240,241,242,243,244,245,246,247,
248,249,250,251,252,253,254,255,
/* This table is a case flipping table. */
0, 1, 2, 3, 4, 5, 6, 7,
8, 9, 10, 11, 12, 13, 14, 15,
16, 17, 18, 19, 20, 21, 22, 23,
24, 25, 26, 27, 28, 29, 30, 31,
32, 33, 34, 35, 36, 37, 38, 39,
40, 41, 42, 43, 44, 45, 46, 47,
48, 49, 50, 51, 52, 53, 54, 55,
56, 57, 58, 59, 60, 61, 62, 63,
64, 97, 98, 99,100,101,102,103,
104,105,106,107,108,109,110,111,
112,113,114,115,116,117,118,119,
120,121,122, 91, 92, 93, 94, 95,
96, 65, 66, 67, 68, 69, 70, 71,
72, 73, 74, 75, 76, 77, 78, 79,
80, 81, 82, 83, 84, 85, 86, 87,
88, 89, 90,123,124,125,126,127,
128,129,130,131,132,133,134,135,
136,137,138,139,140,141,142,143,
144,145,146,147,148,149,150,151,
152,153,154,155,156,157,158,159,
160,161,162,163,164,165,166,167,
168,169,170,171,172,173,174,175,
176,177,178,179,180,181,182,183,
184,185,186,187,188,189,190,191,
192,193,194,195,196,197,198,199,
200,201,202,203,204,205,206,207,
208,209,210,211,212,213,214,215,
216,217,218,219,220,221,222,223,
224,225,226,227,228,229,230,231,
232,233,234,235,236,237,238,239,
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. */
0x00,0x3e,0x00,0x00,0x01,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
0x7e,0x00,0x00,0x00,0x7e,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0xfe,0xff,0xff,0x07,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0x07,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
0xfe,0xff,0xff,0x87,0xfe,0xff,0xff,0x07,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0xff,
0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0xff,0xff,0xff,0xff,
0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0xfe,0xff,0x00,0xfc,
0x01,0x00,0x00,0xf8,0x01,0x00,0x00,0x78,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0xff,0xff,0xff,0xff,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x80,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
/* This table identifies various classes of character by individual bits:
0x01 white space character
0x02 letter
0x04 decimal digit
0x08 hexadecimal digit
0x10 alphanumeric or '_'
0x80 regular expression metacharacter or binary zero
*/
0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 */
0x00,0x01,0x01,0x01,0x01,0x01,0x00,0x00, /* 8- 15 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */
0x01,0x00,0x00,0x00,0x80,0x00,0x00,0x00, /* - ' */
0x80,0x80,0x80,0x80,0x00,0x00,0x80,0x00, /* ( - / */
0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /* 0 - 7 */
0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x80, /* 8 - ? */
0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* @ - G */
0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* H - O */
0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* P - W */
0x12,0x12,0x12,0x80,0x80,0x00,0x80,0x10, /* X - _ */
0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* ` - g */
0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* h - o */
0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* p - w */
0x12,0x12,0x12,0x80,0x80,0x00,0x00,0x00, /* x -127 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */
/* End of pcre_chartables.c */
deps/pcre/pcre_compile.c 0 → 100644
This source diff could not be displayed because it is too large. You can view the blob instead.
deps/pcre/pcre_config.c 0 → 100644
/*************************************************
* 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) 1997-2012 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 the external function pcre_config(). */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
/* Keep the original link size. */
static int real_link_size = LINK_SIZE;
#include "pcre_internal.h"
/*************************************************
* Return info about what features are configured *
*************************************************/
/* This function has an extensible interface so that additional items can be
added compatibly.
Arguments:
what what information is required
where where to put the information
Returns: 0 if data returned, negative on error
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_config(int what, void *where)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_config(int what, void *where)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_config(int what, void *where)
#endif
{
switch (what)
{
case PCRE_CONFIG_UTF8:
#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
*((int *)where) = 0;
return PCRE_ERROR_BADOPTION;
#else
#if defined SUPPORT_UTF
*((int *)where) = 1;
#else
*((int *)where) = 0;
#endif
break;
#endif
case PCRE_CONFIG_UTF16:
#if defined COMPILE_PCRE8 || defined COMPILE_PCRE32
*((int *)where) = 0;
return PCRE_ERROR_BADOPTION;
#else
#if defined SUPPORT_UTF
*((int *)where) = 1;
#else
*((int *)where) = 0;
#endif
break;
#endif
case PCRE_CONFIG_UTF32:
#if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
*((int *)where) = 0;
return PCRE_ERROR_BADOPTION;
#else
#if defined SUPPORT_UTF
*((int *)where) = 1;
#else
*((int *)where) = 0;
#endif
break;
#endif
case PCRE_CONFIG_UNICODE_PROPERTIES:
#ifdef SUPPORT_UCP
*((int *)where) = 1;
#else
*((int *)where) = 0;
#endif
break;
case PCRE_CONFIG_JIT:
#ifdef SUPPORT_JIT
*((int *)where) = 1;
#else
*((int *)where) = 0;
#endif
break;
case PCRE_CONFIG_JITTARGET:
#ifdef SUPPORT_JIT
*((const char **)where) = PRIV(jit_get_target)();
#else
*((const char **)where) = NULL;
#endif
break;
case PCRE_CONFIG_NEWLINE:
*((int *)where) = NEWLINE;
break;
case PCRE_CONFIG_BSR:
#ifdef BSR_ANYCRLF
*((int *)where) = 1;
#else
*((int *)where) = 0;
#endif
break;
case PCRE_CONFIG_LINK_SIZE:
*((int *)where) = real_link_size;
break;
case PCRE_CONFIG_POSIX_MALLOC_THRESHOLD:
*((int *)where) = POSIX_MALLOC_THRESHOLD;
break;
case PCRE_CONFIG_PARENS_LIMIT:
*((unsigned long int *)where) = PARENS_NEST_LIMIT;
break;
case PCRE_CONFIG_MATCH_LIMIT:
*((unsigned long int *)where) = MATCH_LIMIT;
break;
case PCRE_CONFIG_MATCH_LIMIT_RECURSION:
*((unsigned long int *)where) = MATCH_LIMIT_RECURSION;
break;
case PCRE_CONFIG_STACKRECURSE:
#ifdef NO_RECURSE
*((int *)where) = 0;
#else
*((int *)where) = 1;
#endif
break;
default: return PCRE_ERROR_BADOPTION;
}
return 0;
}
/* End of pcre_config.c */
deps/pcre/pcre_dfa_exec.c 0 → 100644
This source diff could not be displayed because it is too large. You can view the blob instead.
deps/pcre/pcre_exec.c 0 → 100644
This source diff could not be displayed because it is too large. You can view the blob instead.
deps/pcre/pcre_fullinfo.c 0 → 100644
/*************************************************
* 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) 1997-2013 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 the external function pcre_fullinfo(), which returns
information about a compiled pattern. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
/*************************************************
* Return info about compiled pattern *
*************************************************/
/* This is a newer "info" function which has an extensible interface so
that additional items can be added compatibly.
Arguments:
argument_re points to compiled code
extra_data points extra data, or NULL
what what information is required
where where to put the information
Returns: 0 if data returned, negative on error
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_fullinfo(const pcre *argument_re, const pcre_extra *extra_data,
int what, void *where)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_fullinfo(const pcre16 *argument_re, const pcre16_extra *extra_data,
int what, void *where)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_fullinfo(const pcre32 *argument_re, const pcre32_extra *extra_data,
int what, void *where)
#endif
{
const REAL_PCRE *re = (const REAL_PCRE *)argument_re;
const pcre_study_data *study = NULL;
if (re == NULL || where == NULL) return PCRE_ERROR_NULL;
if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_STUDY_DATA) != 0)
study = (const pcre_study_data *)extra_data->study_data;
/* Check that the first field in the block is the magic number. If it is not,
return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to
REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which
means that the pattern is likely compiled with different endianness. */
if (re->magic_number != MAGIC_NUMBER)
return re->magic_number == REVERSED_MAGIC_NUMBER?
PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC;
/* Check that this pattern was compiled in the correct bit mode */
if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
switch (what)
{
case PCRE_INFO_OPTIONS:
*((unsigned long int *)where) = re->options & PUBLIC_COMPILE_OPTIONS;
break;
case PCRE_INFO_SIZE:
*((size_t *)where) = re->size;
break;
case PCRE_INFO_STUDYSIZE:
*((size_t *)where) = (study == NULL)? 0 : study->size;
break;
case PCRE_INFO_JITSIZE:
#ifdef SUPPORT_JIT
*((size_t *)where) =
(extra_data != NULL &&
(extra_data->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 &&
extra_data->executable_jit != NULL)?
PRIV(jit_get_size)(extra_data->executable_jit) : 0;
#else
*((size_t *)where) = 0;
#endif
break;
case PCRE_INFO_CAPTURECOUNT:
*((int *)where) = re->top_bracket;
break;
case PCRE_INFO_BACKREFMAX:
*((int *)where) = re->top_backref;
break;
case PCRE_INFO_FIRSTBYTE:
*((int *)where) =
((re->flags & PCRE_FIRSTSET) != 0)? (int)re->first_char :
((re->flags & PCRE_STARTLINE) != 0)? -1 : -2;
break;
case PCRE_INFO_FIRSTCHARACTER:
*((pcre_uint32 *)where) =
(re->flags & PCRE_FIRSTSET) != 0 ? re->first_char : 0;
break;
case PCRE_INFO_FIRSTCHARACTERFLAGS:
*((int *)where) =
((re->flags & PCRE_FIRSTSET) != 0) ? 1 :
((re->flags & PCRE_STARTLINE) != 0) ? 2 : 0;
break;
/* Make sure we pass back the pointer to the bit vector in the external
block, not the internal copy (with flipped integer fields). */
case PCRE_INFO_FIRSTTABLE:
*((const pcre_uint8 **)where) =
(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)?
(int)(study->minlength) : -1;
break;
case PCRE_INFO_JIT:
*((int *)where) = extra_data != NULL &&
(extra_data->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 &&
extra_data->executable_jit != NULL;
break;
case PCRE_INFO_LASTLITERAL:
*((int *)where) =
((re->flags & PCRE_REQCHSET) != 0)? (int)re->req_char : -1;
break;
case PCRE_INFO_REQUIREDCHAR:
*((pcre_uint32 *)where) =
((re->flags & PCRE_REQCHSET) != 0) ? re->req_char : 0;
break;
case PCRE_INFO_REQUIREDCHARFLAGS:
*((int *)where) =
((re->flags & PCRE_REQCHSET) != 0);
break;
case PCRE_INFO_NAMEENTRYSIZE:
*((int *)where) = re->name_entry_size;
break;
case PCRE_INFO_NAMECOUNT:
*((int *)where) = re->name_count;
break;
case PCRE_INFO_NAMETABLE:
*((const pcre_uchar **)where) = (const pcre_uchar *)re + re->name_table_offset;
break;
case PCRE_INFO_DEFAULT_TABLES:
*((const pcre_uint8 **)where) = (const pcre_uint8 *)(PRIV(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;
case PCRE_INFO_JCHANGED:
*((int *)where) = (re->flags & PCRE_JCHANGED) != 0;
break;
case PCRE_INFO_HASCRORLF:
*((int *)where) = (re->flags & PCRE_HASCRORLF) != 0;
break;
case PCRE_INFO_MAXLOOKBEHIND:
*((int *)where) = re->max_lookbehind;
break;
case PCRE_INFO_MATCHLIMIT:
if ((re->flags & PCRE_MLSET) == 0) return PCRE_ERROR_UNSET;
*((pcre_uint32 *)where) = re->limit_match;
break;
case PCRE_INFO_RECURSIONLIMIT:
if ((re->flags & PCRE_RLSET) == 0) return PCRE_ERROR_UNSET;
*((pcre_uint32 *)where) = re->limit_recursion;
break;
case PCRE_INFO_MATCH_EMPTY:
*((int *)where) = (re->flags & PCRE_MATCH_EMPTY) != 0;
break;
default: return PCRE_ERROR_BADOPTION;
}
return 0;
}
/* End of pcre_fullinfo.c */
deps/pcre/pcre_get.c 0 → 100644
/*************************************************
* 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) 1997-2012 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 some convenience functions for extracting substrings
from the subject string after a regex match has succeeded. The original idea
for these functions came from Scott Wimer. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
/*************************************************
* Find number for named string *
*************************************************/
/* This function is used by the get_first_set() function below, as well
as being generally available. It assumes that names are unique.
Arguments:
code the compiled regex
stringname the name whose number is required
Returns: the number of the named parentheses, or a negative number
(PCRE_ERROR_NOSUBSTRING) if not found
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_get_stringnumber(const pcre *code, const char *stringname)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_get_stringnumber(const pcre16 *code, PCRE_SPTR16 stringname)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_get_stringnumber(const pcre32 *code, PCRE_SPTR32 stringname)
#endif
{
int rc;
int entrysize;
int top, bot;
pcre_uchar *nametable;
#ifdef COMPILE_PCRE8
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
return rc;
if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
return rc;
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
return rc;
#endif
#ifdef COMPILE_PCRE16
if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
return rc;
if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
return rc;
if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
return rc;
#endif
#ifdef COMPILE_PCRE32
if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
return rc;
if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
return rc;
if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
return rc;
#endif
bot = 0;
while (top > bot)
{
int mid = (top + bot) / 2;
pcre_uchar *entry = nametable + entrysize*mid;
int c = STRCMP_UC_UC((pcre_uchar *)stringname,
(pcre_uchar *)(entry + IMM2_SIZE));
if (c == 0) return GET2(entry, 0);
if (c > 0) bot = mid + 1; else top = mid;
}
return PCRE_ERROR_NOSUBSTRING;
}
/*************************************************
* Find (multiple) entries for named string *
*************************************************/
/* This is used by the get_first_set() function below, as well as being
generally available. It is used when duplicated names are permitted.
Arguments:
code the compiled regex
stringname the name whose entries required
firstptr where to put the pointer to the first entry
lastptr where to put the pointer to the last entry
Returns: the length of each entry, or a negative number
(PCRE_ERROR_NOSUBSTRING) if not found
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_get_stringtable_entries(const pcre *code, const char *stringname,
char **firstptr, char **lastptr)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_get_stringtable_entries(const pcre16 *code, PCRE_SPTR16 stringname,
PCRE_UCHAR16 **firstptr, PCRE_UCHAR16 **lastptr)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_get_stringtable_entries(const pcre32 *code, PCRE_SPTR32 stringname,
PCRE_UCHAR32 **firstptr, PCRE_UCHAR32 **lastptr)
#endif
{
int rc;
int entrysize;
int top, bot;
pcre_uchar *nametable, *lastentry;
#ifdef COMPILE_PCRE8
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
return rc;
if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
return rc;
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
return rc;
#endif
#ifdef COMPILE_PCRE16
if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
return rc;
if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
return rc;
if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
return rc;
#endif
#ifdef COMPILE_PCRE32
if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
return rc;
if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
return rc;
if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
return rc;
#endif
lastentry = nametable + entrysize * (top - 1);
bot = 0;
while (top > bot)
{
int mid = (top + bot) / 2;
pcre_uchar *entry = nametable + entrysize*mid;
int c = STRCMP_UC_UC((pcre_uchar *)stringname,
(pcre_uchar *)(entry + IMM2_SIZE));
if (c == 0)
{
pcre_uchar *first = entry;
pcre_uchar *last = entry;
while (first > nametable)
{
if (STRCMP_UC_UC((pcre_uchar *)stringname,
(pcre_uchar *)(first - entrysize + IMM2_SIZE)) != 0) break;
first -= entrysize;
}
while (last < lastentry)
{
if (STRCMP_UC_UC((pcre_uchar *)stringname,
(pcre_uchar *)(last + entrysize + IMM2_SIZE)) != 0) break;
last += entrysize;
}
#if defined COMPILE_PCRE8
*firstptr = (char *)first;
*lastptr = (char *)last;
#elif defined COMPILE_PCRE16
*firstptr = (PCRE_UCHAR16 *)first;
*lastptr = (PCRE_UCHAR16 *)last;
#elif defined COMPILE_PCRE32
*firstptr = (PCRE_UCHAR32 *)first;
*lastptr = (PCRE_UCHAR32 *)last;
#endif
return entrysize;
}
if (c > 0) bot = mid + 1; else top = mid;
}
return PCRE_ERROR_NOSUBSTRING;
}
/*************************************************
* Find first set of multiple named strings *
*************************************************/
/* This function allows for duplicate names in the table of named substrings.
It returns the number of the first one that was set in a pattern match.
Arguments:
code the compiled regex
stringname the name of the capturing substring
ovector the vector of matched substrings
stringcount number of captured substrings
Returns: the number of the first that is set,
or the number of the last one if none are set,
or a negative number on error
*/
#if defined COMPILE_PCRE8
static int
get_first_set(const pcre *code, const char *stringname, int *ovector,
int stringcount)
#elif defined COMPILE_PCRE16
static int
get_first_set(const pcre16 *code, PCRE_SPTR16 stringname, int *ovector,
int stringcount)
#elif defined COMPILE_PCRE32
static int
get_first_set(const pcre32 *code, PCRE_SPTR32 stringname, int *ovector,
int stringcount)
#endif
{
const REAL_PCRE *re = (const REAL_PCRE *)code;
int entrysize;
pcre_uchar *entry;
#if defined COMPILE_PCRE8
char *first, *last;
#elif defined COMPILE_PCRE16
PCRE_UCHAR16 *first, *last;
#elif defined COMPILE_PCRE32
PCRE_UCHAR32 *first, *last;
#endif
#if defined COMPILE_PCRE8
if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0)
return pcre_get_stringnumber(code, stringname);
entrysize = pcre_get_stringtable_entries(code, stringname, &first, &last);
#elif defined COMPILE_PCRE16
if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0)
return pcre16_get_stringnumber(code, stringname);
entrysize = pcre16_get_stringtable_entries(code, stringname, &first, &last);
#elif defined COMPILE_PCRE32
if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0)
return pcre32_get_stringnumber(code, stringname);
entrysize = pcre32_get_stringtable_entries(code, stringname, &first, &last);
#endif
if (entrysize <= 0) return entrysize;
for (entry = (pcre_uchar *)first; entry <= (pcre_uchar *)last; entry += entrysize)
{
int n = GET2(entry, 0);
if (n < stringcount && ovector[n*2] >= 0) return n;
}
return GET2(entry, 0);
}
/*************************************************
* Copy captured string to given buffer *
*************************************************/
/* This function copies a single captured substring into a given buffer.
Note that we use memcpy() rather than strncpy() in case there are binary zeros
in the string.
Arguments:
subject the subject string that was matched
ovector pointer to the offsets table
stringcount the number of substrings that were captured
(i.e. the yield of the pcre_exec call, unless
that was zero, in which case it should be 1/3
of the offset table size)
stringnumber the number of the required substring
buffer where to put the substring
size the size of the buffer
Returns: if successful:
the length of the copied string, not including the zero
that is put on the end; can be zero
if not successful:
PCRE_ERROR_NOMEMORY (-6) buffer too small
PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_copy_substring(const char *subject, int *ovector, int stringcount,
int stringnumber, char *buffer, int size)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_copy_substring(PCRE_SPTR16 subject, int *ovector, int stringcount,
int stringnumber, PCRE_UCHAR16 *buffer, int size)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_copy_substring(PCRE_SPTR32 subject, int *ovector, int stringcount,
int stringnumber, PCRE_UCHAR32 *buffer, int size)
#endif
{
int yield;
if (stringnumber < 0 || stringnumber >= stringcount)
return PCRE_ERROR_NOSUBSTRING;
stringnumber *= 2;
yield = ovector[stringnumber+1] - ovector[stringnumber];
if (size < yield + 1) return PCRE_ERROR_NOMEMORY;
memcpy(buffer, subject + ovector[stringnumber], IN_UCHARS(yield));
buffer[yield] = 0;
return yield;
}
/*************************************************
* Copy named captured string to given buffer *
*************************************************/
/* This function copies a single captured substring into a given buffer,
identifying it by name. If the regex permits duplicate names, the first
substring that is set is chosen.
Arguments:
code the compiled regex
subject the subject string that was matched
ovector pointer to the offsets table
stringcount the number of substrings that were captured
(i.e. the yield of the pcre_exec call, unless
that was zero, in which case it should be 1/3
of the offset table size)
stringname the name of the required substring
buffer where to put the substring
size the size of the buffer
Returns: if successful:
the length of the copied string, not including the zero
that is put on the end; can be zero
if not successful:
PCRE_ERROR_NOMEMORY (-6) buffer too small
PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
*/
#if defined COMPILE_PCRE8
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)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_copy_named_substring(const pcre16 *code, PCRE_SPTR16 subject,
int *ovector, int stringcount, PCRE_SPTR16 stringname,
PCRE_UCHAR16 *buffer, int size)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_copy_named_substring(const pcre32 *code, PCRE_SPTR32 subject,
int *ovector, int stringcount, PCRE_SPTR32 stringname,
PCRE_UCHAR32 *buffer, int size)
#endif
{
int n = get_first_set(code, stringname, ovector, stringcount);
if (n <= 0) return n;
#if defined COMPILE_PCRE8
return pcre_copy_substring(subject, ovector, stringcount, n, buffer, size);
#elif defined COMPILE_PCRE16
return pcre16_copy_substring(subject, ovector, stringcount, n, buffer, size);
#elif defined COMPILE_PCRE32
return pcre32_copy_substring(subject, ovector, stringcount, n, buffer, size);
#endif
}
/*************************************************
* Copy all captured strings to new store *
*************************************************/
/* This function gets one chunk of store and builds a list of pointers and all
of the captured substrings in it. A NULL pointer is put on the end of the list.
Arguments:
subject the subject string that was matched
ovector pointer to the offsets table
stringcount the number of substrings that were captured
(i.e. the yield of the pcre_exec call, unless
that was zero, in which case it should be 1/3
of the offset table size)
listptr set to point to the list of pointers
Returns: if successful: 0
if not successful:
PCRE_ERROR_NOMEMORY (-6) failed to get store
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_get_substring_list(const char *subject, int *ovector, int stringcount,
const char ***listptr)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_get_substring_list(PCRE_SPTR16 subject, int *ovector, int stringcount,
PCRE_SPTR16 **listptr)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_get_substring_list(PCRE_SPTR32 subject, int *ovector, int stringcount,
PCRE_SPTR32 **listptr)
#endif
{
int i;
int size = sizeof(pcre_uchar *);
int double_count = stringcount * 2;
pcre_uchar **stringlist;
pcre_uchar *p;
for (i = 0; i < double_count; i += 2)
{
size += sizeof(pcre_uchar *) + IN_UCHARS(1);
if (ovector[i+1] > ovector[i]) size += IN_UCHARS(ovector[i+1] - ovector[i]);
}
stringlist = (pcre_uchar **)(PUBL(malloc))(size);
if (stringlist == NULL) return PCRE_ERROR_NOMEMORY;
#if defined COMPILE_PCRE8
*listptr = (const char **)stringlist;
#elif defined COMPILE_PCRE16
*listptr = (PCRE_SPTR16 *)stringlist;
#elif defined COMPILE_PCRE32
*listptr = (PCRE_SPTR32 *)stringlist;
#endif
p = (pcre_uchar *)(stringlist + stringcount + 1);
for (i = 0; i < double_count; i += 2)
{
int len = (ovector[i+1] > ovector[i])? (ovector[i+1] - ovector[i]) : 0;
memcpy(p, subject + ovector[i], IN_UCHARS(len));
*stringlist++ = p;
p += len;
*p++ = 0;
}
*stringlist = NULL;
return 0;
}
/*************************************************
* Free store obtained by get_substring_list *
*************************************************/
/* This function exists for the benefit of people calling PCRE from non-C
programs that can call its functions, but not free() or (PUBL(free))()
directly.
Argument: the result of a previous pcre_get_substring_list()
Returns: nothing
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
pcre_free_substring_list(const char **pointer)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
pcre16_free_substring_list(PCRE_SPTR16 *pointer)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
pcre32_free_substring_list(PCRE_SPTR32 *pointer)
#endif
{
(PUBL(free))((void *)pointer);
}
/*************************************************
* Copy captured string to new store *
*************************************************/
/* This function copies a single captured substring into a piece of new
store
Arguments:
subject the subject string that was matched
ovector pointer to the offsets table
stringcount the number of substrings that were captured
(i.e. the yield of the pcre_exec call, unless
that was zero, in which case it should be 1/3
of the offset table size)
stringnumber the number of the required substring
stringptr where to put a pointer to the substring
Returns: if successful:
the length of the string, not including the zero that
is put on the end; can be zero
if not successful:
PCRE_ERROR_NOMEMORY (-6) failed to get store
PCRE_ERROR_NOSUBSTRING (-7) substring not present
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_get_substring(const char *subject, int *ovector, int stringcount,
int stringnumber, const char **stringptr)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_get_substring(PCRE_SPTR16 subject, int *ovector, int stringcount,
int stringnumber, PCRE_SPTR16 *stringptr)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_get_substring(PCRE_SPTR32 subject, int *ovector, int stringcount,
int stringnumber, PCRE_SPTR32 *stringptr)
#endif
{
int yield;
pcre_uchar *substring;
if (stringnumber < 0 || stringnumber >= stringcount)
return PCRE_ERROR_NOSUBSTRING;
stringnumber *= 2;
yield = ovector[stringnumber+1] - ovector[stringnumber];
substring = (pcre_uchar *)(PUBL(malloc))(IN_UCHARS(yield + 1));
if (substring == NULL) return PCRE_ERROR_NOMEMORY;
memcpy(substring, subject + ovector[stringnumber], IN_UCHARS(yield));
substring[yield] = 0;
#if defined COMPILE_PCRE8
*stringptr = (const char *)substring;
#elif defined COMPILE_PCRE16
*stringptr = (PCRE_SPTR16)substring;
#elif defined COMPILE_PCRE32
*stringptr = (PCRE_SPTR32)substring;
#endif
return yield;
}
/*************************************************
* Copy named captured string to new store *
*************************************************/
/* This function copies a single captured substring, identified by name, into
new store. If the regex permits duplicate names, the first substring that is
set is chosen.
Arguments:
code the compiled regex
subject the subject string that was matched
ovector pointer to the offsets table
stringcount the number of substrings that were captured
(i.e. the yield of the pcre_exec call, unless
that was zero, in which case it should be 1/3
of the offset table size)
stringname the name of the required substring
stringptr where to put the pointer
Returns: if successful:
the length of the copied string, not including the zero
that is put on the end; can be zero
if not successful:
PCRE_ERROR_NOMEMORY (-6) couldn't get memory
PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
*/
#if defined COMPILE_PCRE8
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)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_get_named_substring(const pcre16 *code, PCRE_SPTR16 subject,
int *ovector, int stringcount, PCRE_SPTR16 stringname,
PCRE_SPTR16 *stringptr)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_get_named_substring(const pcre32 *code, PCRE_SPTR32 subject,
int *ovector, int stringcount, PCRE_SPTR32 stringname,
PCRE_SPTR32 *stringptr)
#endif
{
int n = get_first_set(code, stringname, ovector, stringcount);
if (n <= 0) return n;
#if defined COMPILE_PCRE8
return pcre_get_substring(subject, ovector, stringcount, n, stringptr);
#elif defined COMPILE_PCRE16
return pcre16_get_substring(subject, ovector, stringcount, n, stringptr);
#elif defined COMPILE_PCRE32
return pcre32_get_substring(subject, ovector, stringcount, n, stringptr);
#endif
}
/*************************************************
* Free store obtained by get_substring *
*************************************************/
/* This function exists for the benefit of people calling PCRE from non-C
programs that can call its functions, but not free() or (PUBL(free))()
directly.
Argument: the result of a previous pcre_get_substring()
Returns: nothing
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
pcre_free_substring(const char *pointer)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
pcre16_free_substring(PCRE_SPTR16 pointer)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
pcre32_free_substring(PCRE_SPTR32 pointer)
#endif
{
(PUBL(free))((void *)pointer);
}
/* End of pcre_get.c */
deps/pcre/pcre_globals.c 0 → 100644
/*************************************************
* 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) 1997-2014 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 global variables that are exported by the PCRE library.
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.
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"
#endif
#include "pcre_internal.h"
#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 *(*PUBL(malloc))(size_t) = LocalPcreMalloc;
PCRE_EXP_DATA_DEFN void (*PUBL(free))(void *) = LocalPcreFree;
PCRE_EXP_DATA_DEFN void *(*PUBL(stack_malloc))(size_t) = LocalPcreMalloc;
PCRE_EXP_DATA_DEFN void (*PUBL(stack_free))(void *) = LocalPcreFree;
PCRE_EXP_DATA_DEFN int (*PUBL(callout))(PUBL(callout_block) *) = NULL;
PCRE_EXP_DATA_DEFN int (*PUBL(stack_guard))(void) = NULL;
#elif !defined VPCOMPAT
PCRE_EXP_DATA_DEFN void *(*PUBL(malloc))(size_t) = malloc;
PCRE_EXP_DATA_DEFN void (*PUBL(free))(void *) = free;
PCRE_EXP_DATA_DEFN void *(*PUBL(stack_malloc))(size_t) = malloc;
PCRE_EXP_DATA_DEFN void (*PUBL(stack_free))(void *) = free;
PCRE_EXP_DATA_DEFN int (*PUBL(callout))(PUBL(callout_block) *) = NULL;
PCRE_EXP_DATA_DEFN int (*PUBL(stack_guard))(void) = NULL;
#endif
/* End of pcre_globals.c */
deps/pcre/pcre_internal.h 0 → 100644
This source diff could not be displayed because it is too large. You can view the blob instead.
deps/pcre/pcre_jit_compile.c 0 → 100644
This source diff could not be displayed because it is too large. You can view the blob instead.
deps/pcre/pcre_maketables.c 0 → 100644
/*************************************************
* 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) 1997-2012 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 the external function pcre_maketables(), which builds
character tables for PCRE in the current locale. The file is compiled on its
own as part of the PCRE library. However, it is also included in the
compilation of dftables.c, in which case the macro DFTABLES is defined. */
#ifndef DFTABLES
# ifdef HAVE_CONFIG_H
# include "config.h"
# endif
# include "pcre_internal.h"
#endif
/*************************************************
* Create PCRE character tables *
*************************************************/
/* This function builds a set of character tables for use by PCRE and returns
a pointer to them. They are build using the ctype functions, and consequently
their contents will depend upon the current locale setting. When compiled as
part of the library, the store is obtained via PUBL(malloc)(), but when
compiled inside dftables, use malloc().
Arguments: none
Returns: pointer to the contiguous block of data
*/
#if defined COMPILE_PCRE8
const unsigned char *
pcre_maketables(void)
#elif defined COMPILE_PCRE16
const unsigned char *
pcre16_maketables(void)
#elif defined COMPILE_PCRE32
const unsigned char *
pcre32_maketables(void)
#endif
{
unsigned char *yield, *p;
int i;
#ifndef DFTABLES
yield = (unsigned char*)(PUBL(malloc))(tables_length);
#else
yield = (unsigned char*)malloc(tables_length);
#endif
if (yield == NULL) return NULL;
p = yield;
/* First comes the lower casing table */
for (i = 0; i < 256; i++) *p++ = tolower(i);
/* Next the case-flipping table */
for (i = 0; i < 256; i++) *p++ = islower(i)? toupper(i) : tolower(i);
/* Then the character class tables. Don't try to be clever and save effort on
exclusive ones - in some locales things may be different.
Note that the table for "space" includes everything "isspace" gives, including
VT in the default locale. This makes it work for the POSIX class [:space:].
From release 8.34 is is also correct for Perl space, because Perl added VT at
release 5.18.
Note also that it is possible for a character to be alnum or alpha without
being lower or upper, such as "male and female ordinals" (\xAA and \xBA) in the
fr_FR locale (at least under Debian Linux's locales as of 12/2005). So we must
test for alnum specially. */
memset(p, 0, cbit_length);
for (i = 0; i < 256; i++)
{
if (isdigit(i)) p[cbit_digit + i/8] |= 1 << (i&7);
if (isupper(i)) p[cbit_upper + i/8] |= 1 << (i&7);
if (islower(i)) p[cbit_lower + i/8] |= 1 << (i&7);
if (isalnum(i)) p[cbit_word + i/8] |= 1 << (i&7);
if (i == '_') p[cbit_word + i/8] |= 1 << (i&7);
if (isspace(i)) p[cbit_space + i/8] |= 1 << (i&7);
if (isxdigit(i))p[cbit_xdigit + i/8] |= 1 << (i&7);
if (isgraph(i)) p[cbit_graph + i/8] |= 1 << (i&7);
if (isprint(i)) p[cbit_print + i/8] |= 1 << (i&7);
if (ispunct(i)) p[cbit_punct + i/8] |= 1 << (i&7);
if (iscntrl(i)) p[cbit_cntrl + i/8] |= 1 << (i&7);
}
p += cbit_length;
/* Finally, the character type table. In this, we used to exclude VT from the
white space chars, because Perl didn't recognize it as such for \s and for
comments within regexes. However, Perl changed at release 5.18, so PCRE changed
at release 8.34. */
for (i = 0; i < 256; i++)
{
int x = 0;
if (isspace(i)) x += ctype_space;
if (isalpha(i)) x += ctype_letter;
if (isdigit(i)) x += ctype_digit;
if (isxdigit(i)) x += ctype_xdigit;
if (isalnum(i) || i == '_') x += ctype_word;
/* Note: strchr includes the terminating zero in the characters it considers.
In this instance, that is ok because we want binary zero to be flagged as a
meta-character, which in this sense is any character that terminates a run
of data characters. */
if (strchr("\\*+?{^.$|()[", i) != 0) x += ctype_meta;
*p++ = x;
}
return yield;
}
/* End of pcre_maketables.c */
deps/pcre/pcre_newline.c 0 → 100644
/*************************************************
* 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) 1997-2012 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 internal functions for testing newlines when more than
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, NLTYPE_ANYCRLF,
and NLTYPE_ANY. The full list of Unicode newline characters is taken from
http://unicode.org/unicode/reports/tr18/. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
/*************************************************
* Check for newline at given position *
*************************************************/
/* It is guaranteed that the initial value of ptr is less than the end of the
string that is being processed.
Arguments:
ptr pointer to possible newline
type the newline type
endptr pointer to the end of the string
lenptr where to return the length
utf TRUE if in utf mode
Returns: TRUE or FALSE
*/
BOOL
PRIV(is_newline)(PCRE_PUCHAR ptr, int type, PCRE_PUCHAR endptr, int *lenptr,
BOOL utf)
{
pcre_uint32 c;
(void)utf;
#ifdef SUPPORT_UTF
if (utf)
{
GETCHAR(c, ptr);
}
else
#endif /* SUPPORT_UTF */
c = *ptr;
/* Note that this function is called only for ANY or ANYCRLF. */
if (type == NLTYPE_ANYCRLF) switch(c)
{
case CHAR_LF: *lenptr = 1; return TRUE;
case CHAR_CR: *lenptr = (ptr < endptr - 1 && ptr[1] == CHAR_LF)? 2 : 1;
return TRUE;
default: return FALSE;
}
/* NLTYPE_ANY */
else switch(c)
{
#ifdef EBCDIC
case CHAR_NEL:
#endif
case CHAR_LF:
case CHAR_VT:
case CHAR_FF: *lenptr = 1; return TRUE;
case CHAR_CR:
*lenptr = (ptr < endptr - 1 && ptr[1] == CHAR_LF)? 2 : 1;
return TRUE;
#ifndef EBCDIC
#ifdef COMPILE_PCRE8
case CHAR_NEL: *lenptr = utf? 2 : 1; return TRUE;
case 0x2028: /* LS */
case 0x2029: *lenptr = 3; return TRUE; /* PS */
#else /* COMPILE_PCRE16 || COMPILE_PCRE32 */
case CHAR_NEL:
case 0x2028: /* LS */
case 0x2029: *lenptr = 1; return TRUE; /* PS */
#endif /* COMPILE_PCRE8 */
#endif /* Not EBCDIC */
default: return FALSE;
}
}
/*************************************************
* Check for newline at previous position *
*************************************************/
/* It is guaranteed that the initial value of ptr is greater than the start of
the string that is being processed.
Arguments:
ptr pointer to possible newline
type the newline type
startptr pointer to the start of the string
lenptr where to return the length
utf TRUE if in utf mode
Returns: TRUE or FALSE
*/
BOOL
PRIV(was_newline)(PCRE_PUCHAR ptr, int type, PCRE_PUCHAR startptr, int *lenptr,
BOOL utf)
{
pcre_uint32 c;
(void)utf;
ptr--;
#ifdef SUPPORT_UTF
if (utf)
{
BACKCHAR(ptr);
GETCHAR(c, ptr);
}
else
#endif /* SUPPORT_UTF */
c = *ptr;
/* Note that this function is called only for ANY or ANYCRLF. */
if (type == NLTYPE_ANYCRLF) switch(c)
{
case CHAR_LF:
*lenptr = (ptr > startptr && ptr[-1] == CHAR_CR)? 2 : 1;
return TRUE;
case CHAR_CR: *lenptr = 1; return TRUE;
default: return FALSE;
}
/* NLTYPE_ANY */
else switch(c)
{
case CHAR_LF:
*lenptr = (ptr > startptr && ptr[-1] == CHAR_CR)? 2 : 1;
return TRUE;
#ifdef EBCDIC
case CHAR_NEL:
#endif
case CHAR_VT:
case CHAR_FF:
case CHAR_CR: *lenptr = 1; return TRUE;
#ifndef EBCDIC
#ifdef COMPILE_PCRE8
case CHAR_NEL: *lenptr = utf? 2 : 1; return TRUE;
case 0x2028: /* LS */
case 0x2029: *lenptr = 3; return TRUE; /* PS */
#else /* COMPILE_PCRE16 || COMPILE_PCRE32 */
case CHAR_NEL:
case 0x2028: /* LS */
case 0x2029: *lenptr = 1; return TRUE; /* PS */
#endif /* COMPILE_PCRE8 */
#endif /* NotEBCDIC */
default: return FALSE;
}
}
/* End of pcre_newline.c */
deps/pcre/pcre_ord2utf8.c 0 → 100644
/*************************************************
* 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) 1997-2012 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 file contains a private PCRE function that converts an ordinal
character value into a UTF8 string. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#define COMPILE_PCRE8
#include "pcre_internal.h"
/*************************************************
* Convert character value to UTF-8 *
*************************************************/
/* This function takes an integer value in the range 0 - 0x10ffff
and encodes it as a UTF-8 character in 1 to 4 pcre_uchars.
Arguments:
cvalue the character value
buffer pointer to buffer for result - at least 6 pcre_uchars long
Returns: number of characters placed in the buffer
*/
unsigned
int
PRIV(ord2utf)(pcre_uint32 cvalue, pcre_uchar *buffer)
{
#ifdef SUPPORT_UTF
register int i, j;
for (i = 0; i < PRIV(utf8_table1_size); i++)
if ((int)cvalue <= PRIV(utf8_table1)[i]) break;
buffer += i;
for (j = i; j > 0; j--)
{
*buffer-- = 0x80 | (cvalue & 0x3f);
cvalue >>= 6;
}
*buffer = PRIV(utf8_table2)[i] | cvalue;
return i + 1;
#else
(void)(cvalue); /* Keep compiler happy; this function won't ever be */
(void)(buffer); /* called when SUPPORT_UTF is not defined. */
return 0;
#endif
}
/* End of pcre_ord2utf8.c */
deps/pcre/pcre_printint.c 0 → 100644
/*************************************************
* 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) 1997-2012 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 a PCRE private debugging function for printing out the
internal form of a compiled regular expression, along with some supporting
local functions. This source file is used in two places:
(1) It is #included by pcre_compile.c when it is compiled in debugging mode
(PCRE_DEBUG defined in pcre_internal.h). It is not included in production
compiles. In this case PCRE_INCLUDED is defined.
(2) It is also compiled separately and linked with pcretest.c, which can be
asked to print out a compiled regex for debugging purposes. */
#ifndef PCRE_INCLUDED
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
/* For pcretest program. */
#define PRIV(name) name
/* We have to include pcre_internal.h because we need the internal info for
displaying the results of pcre_study() and we also need to know about the
internal macros, structures, and other internal data values; pcretest has
"inside information" compared to a program that strictly follows the PCRE API.
Although pcre_internal.h does itself include pcre.h, we explicitly include it
here before pcre_internal.h so that the PCRE_EXP_xxx macros get set
appropriately for an application, not for building PCRE. */
#include "pcre.h"
#include "pcre_internal.h"
/* These are the funtions that are contained within. It doesn't seem worth
having a separate .h file just for this. */
#endif /* PCRE_INCLUDED */
#ifdef PCRE_INCLUDED
static /* Keep the following function as private. */
#endif
#if defined COMPILE_PCRE8
void pcre_printint(pcre *external_re, FILE *f, BOOL print_lengths);
#elif defined COMPILE_PCRE16
void pcre16_printint(pcre *external_re, FILE *f, BOOL print_lengths);
#elif defined COMPILE_PCRE32
void pcre32_printint(pcre *external_re, FILE *f, BOOL print_lengths);
#endif
/* Macro that decides whether a character should be output as a literal or in
hexadecimal. We don't use isprint() because that can vary from system to system
(even without the use of locales) and we want the output always to be the same,
for testing purposes. */
#ifdef EBCDIC
#define PRINTABLE(c) ((c) >= 64 && (c) < 255)
#else
#define PRINTABLE(c) ((c) >= 32 && (c) < 127)
#endif
/* The table of operator names. */
static const char *priv_OP_names[] = { OP_NAME_LIST };
/* This table of operator lengths is not actually used by the working code,
but its size is needed for a check that ensures it is the correct size for the
number of opcodes (thus catching update omissions). */
static const pcre_uint8 priv_OP_lengths[] = { OP_LENGTHS };
/*************************************************
* Print single- or multi-byte character *
*************************************************/
static unsigned int
print_char(FILE *f, pcre_uchar *ptr, BOOL utf)
{
pcre_uint32 c = *ptr;
#ifndef SUPPORT_UTF
(void)utf; /* Avoid compiler warning */
if (PRINTABLE(c)) fprintf(f, "%c", (char)c);
else if (c <= 0x80) fprintf(f, "\\x%02x", c);
else fprintf(f, "\\x{%x}", c);
return 0;
#else
#if defined COMPILE_PCRE8
if (!utf || (c & 0xc0) != 0xc0)
{
if (PRINTABLE(c)) fprintf(f, "%c", (char)c);
else if (c < 0x80) fprintf(f, "\\x%02x", c);
else fprintf(f, "\\x{%02x}", c);
return 0;
}
else
{
int i;
int a = PRIV(utf8_table4)[c & 0x3f]; /* Number of additional bytes */
int s = 6*a;
c = (c & PRIV(utf8_table3)[a]) << s;
for (i = 1; i <= a; i++)
{
/* This is a check for malformed UTF-8; it should only occur if the sanity
check has been turned off. Rather than swallow random bytes, just stop if
we hit a bad one. Print it with \X instead of \x as an indication. */
if ((ptr[i] & 0xc0) != 0x80)
{
fprintf(f, "\\X{%x}", c);
return i - 1;
}
/* The byte is OK */
s -= 6;
c |= (ptr[i] & 0x3f) << s;
}
fprintf(f, "\\x{%x}", c);
return a;
}
#elif defined COMPILE_PCRE16
if (!utf || (c & 0xfc00) != 0xd800)
{
if (PRINTABLE(c)) fprintf(f, "%c", (char)c);
else if (c <= 0x80) fprintf(f, "\\x%02x", c);
else fprintf(f, "\\x{%02x}", c);
return 0;
}
else
{
/* This is a check for malformed UTF-16; it should only occur if the sanity
check has been turned off. Rather than swallow a low surrogate, just stop if
we hit a bad one. Print it with \X instead of \x as an indication. */
if ((ptr[1] & 0xfc00) != 0xdc00)
{
fprintf(f, "\\X{%x}", c);
return 0;
}
c = (((c & 0x3ff) << 10) | (ptr[1] & 0x3ff)) + 0x10000;
fprintf(f, "\\x{%x}", c);
return 1;
}
#elif defined COMPILE_PCRE32
if (!utf || (c & 0xfffff800u) != 0xd800u)
{
if (PRINTABLE(c)) fprintf(f, "%c", (char)c);
else if (c <= 0x80) fprintf(f, "\\x%02x", c);
else fprintf(f, "\\x{%x}", c);
return 0;
}
else
{
/* This is a check for malformed UTF-32; it should only occur if the sanity
check has been turned off. Rather than swallow a surrogate, just stop if
we hit one. Print it with \X instead of \x as an indication. */
fprintf(f, "\\X{%x}", c);
return 0;
}
#endif /* COMPILE_PCRE[8|16|32] */
#endif /* SUPPORT_UTF */
}
/*************************************************
* Print uchar string (regardless of utf) *
*************************************************/
static void
print_puchar(FILE *f, PCRE_PUCHAR ptr)
{
while (*ptr != '\0')
{
register pcre_uint32 c = *ptr++;
if (PRINTABLE(c)) fprintf(f, "%c", c); else fprintf(f, "\\x{%x}", c);
}
}
/*************************************************
* Find Unicode property name *
*************************************************/
static const char *
get_ucpname(unsigned int ptype, unsigned int pvalue)
{
#ifdef SUPPORT_UCP
int i;
for (i = PRIV(utt_size) - 1; i >= 0; i--)
{
if (ptype == PRIV(utt)[i].type && pvalue == PRIV(utt)[i].value) break;
}
return (i >= 0)? PRIV(utt_names) + PRIV(utt)[i].name_offset : "??";
#else
/* It gets harder and harder to shut off unwanted compiler warnings. */
ptype = ptype * pvalue;
return (ptype == pvalue)? "??" : "??";
#endif
}
/*************************************************
* Print Unicode property value *
*************************************************/
/* "Normal" properties can be printed from tables. The PT_CLIST property is a
pseudo-property that contains a pointer to a list of case-equivalent
characters. This is used only when UCP support is available and UTF mode is
selected. It should never occur otherwise, but just in case it does, have
something ready to print. */
static void
print_prop(FILE *f, pcre_uchar *code, const char *before, const char *after)
{
if (code[1] != PT_CLIST)
{
fprintf(f, "%s%s %s%s", before, priv_OP_names[*code], get_ucpname(code[1],
code[2]), after);
}
else
{
const char *not = (*code == OP_PROP)? "" : "not ";
#ifndef SUPPORT_UCP
fprintf(f, "%s%sclist %d%s", before, not, code[2], after);
#else
const pcre_uint32 *p = PRIV(ucd_caseless_sets) + code[2];
fprintf (f, "%s%sclist", before, not);
while (*p < NOTACHAR) fprintf(f, " %04x", *p++);
fprintf(f, "%s", after);
#endif
}
}
/*************************************************
* Print compiled regex *
*************************************************/
/* Make this function work for a regex with integers either byte order.
However, we assume that what we are passed is a compiled regex. The
print_lengths flag controls whether offsets and lengths of items are printed.
They can be turned off from pcretest so that automatic tests on bytecode can be
written that do not depend on the value of LINK_SIZE. */
#ifdef PCRE_INCLUDED
static /* Keep the following function as private. */
#endif
#if defined COMPILE_PCRE8
void
pcre_printint(pcre *external_re, FILE *f, BOOL print_lengths)
#elif defined COMPILE_PCRE16
void
pcre16_printint(pcre *external_re, FILE *f, BOOL print_lengths)
#elif defined COMPILE_PCRE32
void
pcre32_printint(pcre *external_re, FILE *f, BOOL print_lengths)
#endif
{
REAL_PCRE *re = (REAL_PCRE *)external_re;
pcre_uchar *codestart, *code;
BOOL utf;
unsigned int options = re->options;
int offset = re->name_table_offset;
int count = re->name_count;
int size = re->name_entry_size;
if (re->magic_number != MAGIC_NUMBER)
{
offset = ((offset << 8) & 0xff00) | ((offset >> 8) & 0xff);
count = ((count << 8) & 0xff00) | ((count >> 8) & 0xff);
size = ((size << 8) & 0xff00) | ((size >> 8) & 0xff);
options = ((options << 24) & 0xff000000) |
((options << 8) & 0x00ff0000) |
((options >> 8) & 0x0000ff00) |
((options >> 24) & 0x000000ff);
}
code = codestart = (pcre_uchar *)re + offset + count * size;
/* PCRE_UTF(16|32) have the same value as PCRE_UTF8. */
utf = (options & PCRE_UTF8) != 0;
for(;;)
{
pcre_uchar *ccode;
const char *flag = " ";
pcre_uint32 c;
unsigned int extra = 0;
if (print_lengths)
fprintf(f, "%3d ", (int)(code - codestart));
else
fprintf(f, " ");
switch(*code)
{
/* ========================================================================== */
/* These cases are never obeyed. This is a fudge that causes a compile-
time error if the vectors OP_names or OP_lengths, which are indexed
by opcode, are not the correct length. It seems to be the only way to do
such a check at compile time, as the sizeof() operator does not work in
the C preprocessor. */
case OP_TABLE_LENGTH:
case OP_TABLE_LENGTH +
((sizeof(priv_OP_names)/sizeof(const char *) == OP_TABLE_LENGTH) &&
(sizeof(priv_OP_lengths) == OP_TABLE_LENGTH)):
break;
/* ========================================================================== */
case OP_END:
fprintf(f, " %s\n", priv_OP_names[*code]);
fprintf(f, "------------------------------------------------------------------\n");
return;
case OP_CHAR:
fprintf(f, " ");
do
{
code++;
code += 1 + print_char(f, code, utf);
}
while (*code == OP_CHAR);
fprintf(f, "\n");
continue;
case OP_CHARI:
fprintf(f, " /i ");
do
{
code++;
code += 1 + print_char(f, code, utf);
}
while (*code == OP_CHARI);
fprintf(f, "\n");
continue;
case OP_CBRA:
case OP_CBRAPOS:
case OP_SCBRA:
case OP_SCBRAPOS:
if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
else fprintf(f, " ");
fprintf(f, "%s %d", priv_OP_names[*code], GET2(code, 1+LINK_SIZE));
break;
case OP_BRA:
case OP_BRAPOS:
case OP_SBRA:
case OP_SBRAPOS:
case OP_KETRMAX:
case OP_KETRMIN:
case OP_KETRPOS:
case OP_ALT:
case OP_KET:
case OP_ASSERT:
case OP_ASSERT_NOT:
case OP_ASSERTBACK:
case OP_ASSERTBACK_NOT:
case OP_ONCE:
case OP_ONCE_NC:
case OP_COND:
case OP_SCOND:
case OP_REVERSE:
if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
else fprintf(f, " ");
fprintf(f, "%s", priv_OP_names[*code]);
break;
case OP_CLOSE:
fprintf(f, " %s %d", priv_OP_names[*code], GET2(code, 1));
break;
case OP_CREF:
fprintf(f, "%3d %s", GET2(code,1), priv_OP_names[*code]);
break;
case OP_DNCREF:
{
pcre_uchar *entry = (pcre_uchar *)re + offset + (GET2(code, 1) * size) +
IMM2_SIZE;
fprintf(f, " %s Cond ref <", flag);
print_puchar(f, entry);
fprintf(f, ">%d", GET2(code, 1 + IMM2_SIZE));
}
break;
case OP_RREF:
c = GET2(code, 1);
if (c == RREF_ANY)
fprintf(f, " Cond recurse any");
else
fprintf(f, " Cond recurse %d", c);
break;
case OP_DNRREF:
{
pcre_uchar *entry = (pcre_uchar *)re + offset + (GET2(code, 1) * size) +
IMM2_SIZE;
fprintf(f, " %s Cond recurse <", flag);
print_puchar(f, entry);
fprintf(f, ">%d", GET2(code, 1 + IMM2_SIZE));
}
break;
case OP_DEF:
fprintf(f, " Cond def");
break;
case OP_STARI:
case OP_MINSTARI:
case OP_POSSTARI:
case OP_PLUSI:
case OP_MINPLUSI:
case OP_POSPLUSI:
case OP_QUERYI:
case OP_MINQUERYI:
case OP_POSQUERYI:
flag = "/i";
/* Fall through */
case OP_STAR:
case OP_MINSTAR:
case OP_POSSTAR:
case OP_PLUS:
case OP_MINPLUS:
case OP_POSPLUS:
case OP_QUERY:
case OP_MINQUERY:
case OP_POSQUERY:
case OP_TYPESTAR:
case OP_TYPEMINSTAR:
case OP_TYPEPOSSTAR:
case OP_TYPEPLUS:
case OP_TYPEMINPLUS:
case OP_TYPEPOSPLUS:
case OP_TYPEQUERY:
case OP_TYPEMINQUERY:
case OP_TYPEPOSQUERY:
fprintf(f, " %s ", flag);
if (*code >= OP_TYPESTAR)
{
if (code[1] == OP_PROP || code[1] == OP_NOTPROP)
{
print_prop(f, code + 1, "", " ");
extra = 2;
}
else fprintf(f, "%s", priv_OP_names[code[1]]);
}
else extra = print_char(f, code+1, utf);
fprintf(f, "%s", priv_OP_names[*code]);
break;
case OP_EXACTI:
case OP_UPTOI:
case OP_MINUPTOI:
case OP_POSUPTOI:
flag = "/i";
/* Fall through */
case OP_EXACT:
case OP_UPTO:
case OP_MINUPTO:
case OP_POSUPTO:
fprintf(f, " %s ", flag);
extra = print_char(f, code + 1 + IMM2_SIZE, utf);
fprintf(f, "{");
if (*code != OP_EXACT && *code != OP_EXACTI) fprintf(f, "0,");
fprintf(f, "%d}", GET2(code,1));
if (*code == OP_MINUPTO || *code == OP_MINUPTOI) fprintf(f, "?");
else if (*code == OP_POSUPTO || *code == OP_POSUPTOI) fprintf(f, "+");
break;
case OP_TYPEEXACT:
case OP_TYPEUPTO:
case OP_TYPEMINUPTO:
case OP_TYPEPOSUPTO:
if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
{
print_prop(f, code + IMM2_SIZE + 1, " ", " ");
extra = 2;
}
else fprintf(f, " %s", priv_OP_names[code[1 + IMM2_SIZE]]);
fprintf(f, "{");
if (*code != OP_TYPEEXACT) fprintf(f, "0,");
fprintf(f, "%d}", GET2(code,1));
if (*code == OP_TYPEMINUPTO) fprintf(f, "?");
else if (*code == OP_TYPEPOSUPTO) fprintf(f, "+");
break;
case OP_NOTI:
flag = "/i";
/* Fall through */
case OP_NOT:
fprintf(f, " %s [^", flag);
extra = print_char(f, code + 1, utf);
fprintf(f, "]");
break;
case OP_NOTSTARI:
case OP_NOTMINSTARI:
case OP_NOTPOSSTARI:
case OP_NOTPLUSI:
case OP_NOTMINPLUSI:
case OP_NOTPOSPLUSI:
case OP_NOTQUERYI:
case OP_NOTMINQUERYI:
case OP_NOTPOSQUERYI:
flag = "/i";
/* Fall through */
case OP_NOTSTAR:
case OP_NOTMINSTAR:
case OP_NOTPOSSTAR:
case OP_NOTPLUS:
case OP_NOTMINPLUS:
case OP_NOTPOSPLUS:
case OP_NOTQUERY:
case OP_NOTMINQUERY:
case OP_NOTPOSQUERY:
fprintf(f, " %s [^", flag);
extra = print_char(f, code + 1, utf);
fprintf(f, "]%s", priv_OP_names[*code]);
break;
case OP_NOTEXACTI:
case OP_NOTUPTOI:
case OP_NOTMINUPTOI:
case OP_NOTPOSUPTOI:
flag = "/i";
/* Fall through */
case OP_NOTEXACT:
case OP_NOTUPTO:
case OP_NOTMINUPTO:
case OP_NOTPOSUPTO:
fprintf(f, " %s [^", flag);
extra = print_char(f, code + 1 + IMM2_SIZE, utf);
fprintf(f, "]{");
if (*code != OP_NOTEXACT && *code != OP_NOTEXACTI) fprintf(f, "0,");
fprintf(f, "%d}", GET2(code,1));
if (*code == OP_NOTMINUPTO || *code == OP_NOTMINUPTOI) fprintf(f, "?");
else
if (*code == OP_NOTPOSUPTO || *code == OP_NOTPOSUPTOI) fprintf(f, "+");
break;
case OP_RECURSE:
if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
else fprintf(f, " ");
fprintf(f, "%s", priv_OP_names[*code]);
break;
case OP_REFI:
flag = "/i";
/* Fall through */
case OP_REF:
fprintf(f, " %s \\%d", flag, GET2(code,1));
ccode = code + priv_OP_lengths[*code];
goto CLASS_REF_REPEAT;
case OP_DNREFI:
flag = "/i";
/* Fall through */
case OP_DNREF:
{
pcre_uchar *entry = (pcre_uchar *)re + offset + (GET2(code, 1) * size) +
IMM2_SIZE;
fprintf(f, " %s \\k<", flag);
print_puchar(f, entry);
fprintf(f, ">%d", GET2(code, 1 + IMM2_SIZE));
}
ccode = code + priv_OP_lengths[*code];
goto CLASS_REF_REPEAT;
case OP_CALLOUT:
fprintf(f, " %s %d %d %d", priv_OP_names[*code], code[1], GET(code,2),
GET(code, 2 + LINK_SIZE));
break;
case OP_PROP:
case OP_NOTPROP:
print_prop(f, code, " ", "");
break;
/* OP_XCLASS cannot occur in 8-bit, non-UTF mode. However, there's no harm
in having this code always here, and it makes it less messy without all
those #ifdefs. */
case OP_CLASS:
case OP_NCLASS:
case OP_XCLASS:
{
int i;
unsigned int min, max;
BOOL printmap;
BOOL invertmap = FALSE;
pcre_uint8 *map;
pcre_uint8 inverted_map[32];
fprintf(f, " [");
if (*code == OP_XCLASS)
{
extra = GET(code, 1);
ccode = code + LINK_SIZE + 1;
printmap = (*ccode & XCL_MAP) != 0;
if ((*ccode & XCL_NOT) != 0)
{
invertmap = (*ccode & XCL_HASPROP) == 0;
fprintf(f, "^");
}
ccode++;
}
else
{
printmap = TRUE;
ccode = code + 1;
}
/* Print a bit map */
if (printmap)
{
map = (pcre_uint8 *)ccode;
if (invertmap)
{
for (i = 0; i < 32; i++) inverted_map[i] = ~map[i];
map = inverted_map;
}
for (i = 0; i < 256; i++)
{
if ((map[i/8] & (1 << (i&7))) != 0)
{
int j;
for (j = i+1; j < 256; j++)
if ((map[j/8] & (1 << (j&7))) == 0) break;
if (i == '-' || i == ']') fprintf(f, "\\");
if (PRINTABLE(i)) fprintf(f, "%c", i);
else fprintf(f, "\\x%02x", i);
if (--j > i)
{
if (j != i + 1) fprintf(f, "-");
if (j == '-' || j == ']') fprintf(f, "\\");
if (PRINTABLE(j)) fprintf(f, "%c", j);
else fprintf(f, "\\x%02x", j);
}
i = j;
}
}
ccode += 32 / sizeof(pcre_uchar);
}
/* For an XCLASS there is always some additional data */
if (*code == OP_XCLASS)
{
pcre_uchar ch;
while ((ch = *ccode++) != XCL_END)
{
BOOL not = FALSE;
const char *notch = "";
switch(ch)
{
case XCL_NOTPROP:
not = TRUE;
notch = "^";
/* Fall through */
case XCL_PROP:
{
unsigned int ptype = *ccode++;
unsigned int pvalue = *ccode++;
switch(ptype)
{
case PT_PXGRAPH:
fprintf(f, "[:%sgraph:]", notch);
break;
case PT_PXPRINT:
fprintf(f, "[:%sprint:]", notch);
break;
case PT_PXPUNCT:
fprintf(f, "[:%spunct:]", notch);
break;
default:
fprintf(f, "\\%c{%s}", (not? 'P':'p'),
get_ucpname(ptype, pvalue));
break;
}
}
break;
default:
ccode += 1 + print_char(f, ccode, utf);
if (ch == XCL_RANGE)
{
fprintf(f, "-");
ccode += 1 + print_char(f, ccode, utf);
}
break;
}
}
}
/* Indicate a non-UTF class which was created by negation */
fprintf(f, "]%s", (*code == OP_NCLASS)? " (neg)" : "");
/* Handle repeats after a class or a back reference */
CLASS_REF_REPEAT:
switch(*ccode)
{
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRPLUS:
case OP_CRMINPLUS:
case OP_CRQUERY:
case OP_CRMINQUERY:
case OP_CRPOSSTAR:
case OP_CRPOSPLUS:
case OP_CRPOSQUERY:
fprintf(f, "%s", priv_OP_names[*ccode]);
extra += priv_OP_lengths[*ccode];
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
case OP_CRPOSRANGE:
min = GET2(ccode,1);
max = GET2(ccode,1 + IMM2_SIZE);
if (max == 0) fprintf(f, "{%u,}", min);
else fprintf(f, "{%u,%u}", min, max);
if (*ccode == OP_CRMINRANGE) fprintf(f, "?");
else if (*ccode == OP_CRPOSRANGE) fprintf(f, "+");
extra += priv_OP_lengths[*ccode];
break;
/* Do nothing if it's not a repeat; this code stops picky compilers
warning about the lack of a default code path. */
default:
break;
}
}
break;
case OP_MARK:
case OP_PRUNE_ARG:
case OP_SKIP_ARG:
case OP_THEN_ARG:
fprintf(f, " %s ", priv_OP_names[*code]);
print_puchar(f, code + 2);
extra += code[1];
break;
case OP_THEN:
fprintf(f, " %s", priv_OP_names[*code]);
break;
case OP_CIRCM:
case OP_DOLLM:
flag = "/m";
/* Fall through */
/* Anything else is just an item with no data, but possibly a flag. */
default:
fprintf(f, " %s %s", flag, priv_OP_names[*code]);
break;
}
code += priv_OP_lengths[*code] + extra;
fprintf(f, "\n");
}
}
/* End of pcre_printint.src */
deps/pcre/pcre_refcount.c 0 → 100644
/*************************************************
* 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) 1997-2012 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 the external function pcre_refcount(), which is an
auxiliary function that can be used to maintain a reference count in a compiled
pattern data block. This might be helpful in applications where the block is
shared by different users. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
/*************************************************
* Maintain reference count *
*************************************************/
/* The reference count is a 16-bit field, initialized to zero. It is not
possible to transfer a non-zero count from one host to a different host that
has a different byte order - though I can't see why anyone in their right mind
would ever want to do that!
Arguments:
argument_re points to compiled code
adjust value to add to the count
Returns: the (possibly updated) count value (a non-negative number), or
a negative error number
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_refcount(pcre *argument_re, int adjust)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre16_refcount(pcre16 *argument_re, int adjust)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
pcre32_refcount(pcre32 *argument_re, int adjust)
#endif
{
REAL_PCRE *re = (REAL_PCRE *)argument_re;
if (re == NULL) return PCRE_ERROR_NULL;
if (re->magic_number != MAGIC_NUMBER) return PCRE_ERROR_BADMAGIC;
if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
re->ref_count = (-adjust > re->ref_count)? 0 :
(adjust + re->ref_count > 65535)? 65535 :
re->ref_count + adjust;
return re->ref_count;
}
/* End of pcre_refcount.c */
deps/pcre/pcre_string_utils.c 0 → 100644
/*************************************************
* 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) 1997-2014 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 internal functions for comparing and finding the length
of strings for different data item sizes. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
#ifndef COMPILE_PCRE8
/*************************************************
* Compare string utilities *
*************************************************/
/* The following two functions compares two strings. Basically a strcmp
for non 8 bit characters.
Arguments:
str1 first string
str2 second string
Returns: 0 if both string are equal (like strcmp), 1 otherwise
*/
int
PRIV(strcmp_uc_uc)(const pcre_uchar *str1, const pcre_uchar *str2)
{
pcre_uchar c1;
pcre_uchar c2;
while (*str1 != '\0' || *str2 != '\0')
{
c1 = *str1++;
c2 = *str2++;
if (c1 != c2)
return ((c1 > c2) << 1) - 1;
}
/* Both length and characters must be equal. */
return 0;
}
#ifdef COMPILE_PCRE32
int
PRIV(strcmp_uc_uc_utf)(const pcre_uchar *str1, const pcre_uchar *str2)
{
pcre_uchar c1;
pcre_uchar c2;
while (*str1 != '\0' || *str2 != '\0')
{
c1 = UCHAR21INC(str1);
c2 = UCHAR21INC(str2);
if (c1 != c2)
return ((c1 > c2) << 1) - 1;
}
/* Both length and characters must be equal. */
return 0;
}
#endif /* COMPILE_PCRE32 */
int
PRIV(strcmp_uc_c8)(const pcre_uchar *str1, const char *str2)
{
const pcre_uint8 *ustr2 = (pcre_uint8 *)str2;
pcre_uchar c1;
pcre_uchar c2;
while (*str1 != '\0' || *ustr2 != '\0')
{
c1 = *str1++;
c2 = (pcre_uchar)*ustr2++;
if (c1 != c2)
return ((c1 > c2) << 1) - 1;
}
/* Both length and characters must be equal. */
return 0;
}
#ifdef COMPILE_PCRE32
int
PRIV(strcmp_uc_c8_utf)(const pcre_uchar *str1, const char *str2)
{
const pcre_uint8 *ustr2 = (pcre_uint8 *)str2;
pcre_uchar c1;
pcre_uchar c2;
while (*str1 != '\0' || *ustr2 != '\0')
{
c1 = UCHAR21INC(str1);
c2 = (pcre_uchar)*ustr2++;
if (c1 != c2)
return ((c1 > c2) << 1) - 1;
}
/* Both length and characters must be equal. */
return 0;
}
#endif /* COMPILE_PCRE32 */
/* The following two functions compares two, fixed length
strings. Basically an strncmp for non 8 bit characters.
Arguments:
str1 first string
str2 second string
num size of the string
Returns: 0 if both string are equal (like strcmp), 1 otherwise
*/
int
PRIV(strncmp_uc_uc)(const pcre_uchar *str1, const pcre_uchar *str2, unsigned int num)
{
pcre_uchar c1;
pcre_uchar c2;
while (num-- > 0)
{
c1 = *str1++;
c2 = *str2++;
if (c1 != c2)
return ((c1 > c2) << 1) - 1;
}
/* Both length and characters must be equal. */
return 0;
}
int
PRIV(strncmp_uc_c8)(const pcre_uchar *str1, const char *str2, unsigned int num)
{
const pcre_uint8 *ustr2 = (pcre_uint8 *)str2;
pcre_uchar c1;
pcre_uchar c2;
while (num-- > 0)
{
c1 = *str1++;
c2 = (pcre_uchar)*ustr2++;
if (c1 != c2)
return ((c1 > c2) << 1) - 1;
}
/* Both length and characters must be equal. */
return 0;
}
/* The following function returns with the length of
a zero terminated string. Basically an strlen for non 8 bit characters.
Arguments:
str string
Returns: length of the string
*/
unsigned int
PRIV(strlen_uc)(const pcre_uchar *str)
{
unsigned int len = 0;
while (*str++ != 0)
len++;
return len;
}
#endif /* !COMPILE_PCRE8 */
/* End of pcre_string_utils.c */
deps/pcre/pcre_study.c 0 → 100644
/*************************************************
* 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) 1997-2012 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 the external function pcre_study(), along with local
supporting functions. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#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, SSB_UNKNOWN };
/*************************************************
* 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:
re compiled pattern block
code pointer to start of group (the bracket)
startcode pointer to start of the whole pattern's code
options the compiling options
recurses chain of recurse_check to catch mutual recursion
countptr pointer to call count (to catch over complexity)
Returns: the minimum length
-1 if \C in UTF-8 mode or (*ACCEPT) was encountered
-2 internal error (missing capturing bracket)
-3 internal error (opcode not listed)
*/
static int
find_minlength(const REAL_PCRE *re, const pcre_uchar *code,
const pcre_uchar *startcode, int options, recurse_check *recurses,
int *countptr)
{
int length = -1;
/* PCRE_UTF16 has the same value as PCRE_UTF8. */
BOOL utf = (options & PCRE_UTF8) != 0;
BOOL had_recurse = FALSE;
recurse_check this_recurse;
register int branchlength = 0;
register pcre_uchar *cc = (pcre_uchar *)code + 1 + LINK_SIZE;
if ((*countptr)++ > 1000) return -1; /* too complex */
if (*code == OP_CBRA || *code == OP_SCBRA ||
*code == OP_CBRAPOS || *code == OP_SCBRAPOS) cc += IMM2_SIZE;
/* 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;
pcre_uchar *cs, *ce;
register pcre_uchar 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_CBRAPOS:
case OP_SCBRAPOS:
case OP_BRAPOS:
case OP_SBRAPOS:
case OP_ONCE:
case OP_ONCE_NC:
d = find_minlength(re, cc, startcode, options, recurses, countptr);
if (d < 0) return d;
branchlength += d;
do cc += GET(cc, 1); while (*cc == OP_ALT);
cc += 1 + LINK_SIZE;
break;
/* ACCEPT makes things far too complicated; we have to give up. */
case OP_ACCEPT:
case OP_ASSERT_ACCEPT:
return -1;
/* 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. If an
ACCEPT was previously encountered, use the length that was in force at that
time, and pass back the shortest ACCEPT length. */
case OP_ALT:
case OP_KET:
case OP_KETRMAX:
case OP_KETRMIN:
case OP_KETRPOS:
case OP_END:
if (length < 0 || (!had_recurse && branchlength < length))
length = branchlength;
if (op != 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_DNCREF:
case OP_RREF:
case OP_DNRREF:
case OP_DEF:
case OP_CALLOUT:
case OP_SOD:
case OP_SOM:
case OP_EOD:
case OP_EODN:
case OP_CIRC:
case OP_CIRCM:
case OP_DOLL:
case OP_DOLLM:
case OP_NOT_WORD_BOUNDARY:
case OP_WORD_BOUNDARY:
cc += PRIV(OP_lengths)[*cc];
break;
/* Skip over a subpattern that has a {0} or {0,x} quantifier */
case OP_BRAZERO:
case OP_BRAMINZERO:
case OP_BRAPOSZERO:
case OP_SKIPZERO:
cc += PRIV(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_CHARI:
case OP_NOT:
case OP_NOTI:
case OP_PLUS:
case OP_PLUSI:
case OP_MINPLUS:
case OP_MINPLUSI:
case OP_POSPLUS:
case OP_POSPLUSI:
case OP_NOTPLUS:
case OP_NOTPLUSI:
case OP_NOTMINPLUS:
case OP_NOTMINPLUSI:
case OP_NOTPOSPLUS:
case OP_NOTPOSPLUSI:
branchlength++;
cc += 2;
#ifdef SUPPORT_UTF
if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
#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_EXACTI:
case OP_NOTEXACT:
case OP_NOTEXACTI:
branchlength += GET2(cc,1);
cc += 2 + IMM2_SIZE;
#ifdef SUPPORT_UTF
if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
#endif
break;
case OP_TYPEEXACT:
branchlength += GET2(cc,1);
cc += 2 + IMM2_SIZE + ((cc[1 + IMM2_SIZE] == OP_PROP
|| cc[1 + IMM2_SIZE] == OP_NOTPROP)? 2 : 0);
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, but it also might match just
one. */
case OP_ANYNL:
branchlength += 1;
cc++;
break;
/* The single-byte matcher means we can't proceed in UTF-8 mode. (In
non-UTF-8 mode \C will actually be turned into OP_ALLANY, so won't ever
appear, but leave the code, just in case.) */
case OP_ANYBYTE:
#ifdef SUPPORT_UTF
if (utf) 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 += PRIV(OP_lengths)[op];
break;
case OP_TYPEUPTO:
case OP_TYPEMINUPTO:
case OP_TYPEPOSUPTO:
if (cc[1 + IMM2_SIZE] == OP_PROP
|| cc[1 + IMM2_SIZE] == OP_NOTPROP) cc += 2;
cc += PRIV(OP_lengths)[op];
break;
/* Check a class for variable quantification */
case OP_CLASS:
case OP_NCLASS:
#if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
case OP_XCLASS:
/* The original code caused an unsigned overflow in 64 bit systems,
so now we use a conditional statement. */
if (op == OP_XCLASS)
cc += GET(cc, 1);
else
cc += PRIV(OP_lengths)[OP_CLASS];
#else
cc += PRIV(OP_lengths)[OP_CLASS];
#endif
switch (*cc)
{
case OP_CRPLUS:
case OP_CRMINPLUS:
case OP_CRPOSPLUS:
branchlength++;
/* Fall through */
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRQUERY:
case OP_CRMINQUERY:
case OP_CRPOSSTAR:
case OP_CRPOSQUERY:
cc++;
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
case OP_CRPOSRANGE:
branchlength += GET2(cc,1);
cc += 1 + 2 * IMM2_SIZE;
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_DNREF: /* Duplicate named pattern back reference */
case OP_DNREFI:
if ((options & PCRE_JAVASCRIPT_COMPAT) == 0)
{
int count = GET2(cc, 1+IMM2_SIZE);
pcre_uchar *slot = (pcre_uchar *)re +
re->name_table_offset + GET2(cc, 1) * re->name_entry_size;
d = INT_MAX;
while (count-- > 0)
{
ce = cs = (pcre_uchar *)PRIV(find_bracket)(startcode, utf, GET2(slot, 0));
if (cs == NULL) return -2;
do ce += GET(ce, 1); while (*ce == OP_ALT);
if (cc > cs && cc < ce) /* Simple recursion */
{
d = 0;
had_recurse = TRUE;
break;
}
else
{
recurse_check *r = recurses;
for (r = recurses; r != NULL; r = r->prev) if (r->group == cs) break;
if (r != NULL) /* Mutual recursion */
{
d = 0;
had_recurse = TRUE;
break;
}
else
{
int dd;
this_recurse.prev = recurses;
this_recurse.group = cs;
dd = find_minlength(re, cs, startcode, options, &this_recurse,
countptr);
if (dd < d) d = dd;
}
}
slot += re->name_entry_size;
}
}
else d = 0;
cc += 1 + 2*IMM2_SIZE;
goto REPEAT_BACK_REFERENCE;
case OP_REF: /* Single back reference */
case OP_REFI:
if ((options & PCRE_JAVASCRIPT_COMPAT) == 0)
{
ce = cs = (pcre_uchar *)PRIV(find_bracket)(startcode, utf, GET2(cc, 1));
if (cs == NULL) return -2;
do ce += GET(ce, 1); while (*ce == OP_ALT);
if (cc > cs && cc < ce) /* Simple recursion */
{
d = 0;
had_recurse = TRUE;
}
else
{
recurse_check *r = recurses;
for (r = recurses; r != NULL; r = r->prev) if (r->group == cs) break;
if (r != NULL) /* Mutual recursion */
{
d = 0;
had_recurse = TRUE;
}
else
{
this_recurse.prev = recurses;
this_recurse.group = cs;
d = find_minlength(re, cs, startcode, options, &this_recurse,
countptr);
}
}
}
else d = 0;
cc += 1 + IMM2_SIZE;
/* Handle repeated back references */
REPEAT_BACK_REFERENCE:
switch (*cc)
{
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRQUERY:
case OP_CRMINQUERY:
case OP_CRPOSSTAR:
case OP_CRPOSQUERY:
min = 0;
cc++;
break;
case OP_CRPLUS:
case OP_CRMINPLUS:
case OP_CRPOSPLUS:
min = 1;
cc++;
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
case OP_CRPOSRANGE:
min = GET2(cc, 1);
cc += 1 + 2 * IMM2_SIZE;
break;
default:
min = 1;
break;
}
branchlength += min * d;
break;
/* We can easily detect direct recursion, but not mutual recursion. This is
caught by a recursion depth count. */
case OP_RECURSE:
cs = ce = (pcre_uchar *)startcode + GET(cc, 1);
do ce += GET(ce, 1); while (*ce == OP_ALT);
if (cc > cs && cc < ce) /* Simple recursion */
had_recurse = TRUE;
else
{
recurse_check *r = recurses;
for (r = recurses; r != NULL; r = r->prev) if (r->group == cs) break;
if (r != NULL) /* Mutual recursion */
had_recurse = TRUE;
else
{
this_recurse.prev = recurses;
this_recurse.group = cs;
branchlength += find_minlength(re, cs, startcode, options,
&this_recurse, countptr);
}
}
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. As it
happens, the "NOT" versions of these opcodes are used at present only for
ASCII characters, so they could be omitted from this list. However, in
future that may change, so we include them here so as not to leave a
gotcha for a future maintainer. */
case OP_UPTO:
case OP_UPTOI:
case OP_NOTUPTO:
case OP_NOTUPTOI:
case OP_MINUPTO:
case OP_MINUPTOI:
case OP_NOTMINUPTO:
case OP_NOTMINUPTOI:
case OP_POSUPTO:
case OP_POSUPTOI:
case OP_NOTPOSUPTO:
case OP_NOTPOSUPTOI:
case OP_STAR:
case OP_STARI:
case OP_NOTSTAR:
case OP_NOTSTARI:
case OP_MINSTAR:
case OP_MINSTARI:
case OP_NOTMINSTAR:
case OP_NOTMINSTARI:
case OP_POSSTAR:
case OP_POSSTARI:
case OP_NOTPOSSTAR:
case OP_NOTPOSSTARI:
case OP_QUERY:
case OP_QUERYI:
case OP_NOTQUERY:
case OP_NOTQUERYI:
case OP_MINQUERY:
case OP_MINQUERYI:
case OP_NOTMINQUERY:
case OP_NOTMINQUERYI:
case OP_POSQUERY:
case OP_POSQUERYI:
case OP_NOTPOSQUERY:
case OP_NOTPOSQUERYI:
cc += PRIV(OP_lengths)[op];
#ifdef SUPPORT_UTF
if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
#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 += PRIV(OP_lengths)[op] + cc[1];
break;
/* The remaining opcodes are just skipped over. */
case OP_CLOSE:
case OP_COMMIT:
case OP_FAIL:
case OP_PRUNE:
case OP_SET_SOM:
case OP_SKIP:
case OP_THEN:
cc += PRIV(OP_lengths)[op];
break;
/* This should not occur: we list all opcodes explicitly so that when
new ones get added they are properly considered. */
default:
return -3;
}
}
/* Control never gets here */
}
/*************************************************
* Set a bit and maybe its alternate case *
*************************************************/
/* 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
p points to the character
caseless the caseless flag
cd the block with char table pointers
utf TRUE for UTF-8 / UTF-16 / UTF-32 mode
Returns: pointer after the character
*/
static const pcre_uchar *
set_table_bit(pcre_uint8 *start_bits, const pcre_uchar *p, BOOL caseless,
compile_data *cd, BOOL utf)
{
pcre_uint32 c = *p;
#ifdef COMPILE_PCRE8
SET_BIT(c);
#ifdef SUPPORT_UTF
if (utf && c > 127)
{
GETCHARINC(c, p);
#ifdef SUPPORT_UCP
if (caseless)
{
pcre_uchar buff[6];
c = UCD_OTHERCASE(c);
(void)PRIV(ord2utf)(c, buff);
SET_BIT(buff[0]);
}
#endif /* Not SUPPORT_UCP */
return p;
}
#else /* Not SUPPORT_UTF */
(void)(utf); /* Stops warning for unused parameter */
#endif /* SUPPORT_UTF */
/* 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;
#endif /* COMPILE_PCRE8 */
#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
if (c > 0xff)
{
c = 0xff;
caseless = FALSE;
}
SET_BIT(c);
#ifdef SUPPORT_UTF
if (utf && c > 127)
{
GETCHARINC(c, p);
#ifdef SUPPORT_UCP
if (caseless)
{
c = UCD_OTHERCASE(c);
if (c > 0xff)
c = 0xff;
SET_BIT(c);
}
#endif /* SUPPORT_UCP */
return p;
}
#else /* Not SUPPORT_UTF */
(void)(utf); /* Stops warning for unused parameter */
#endif /* SUPPORT_UTF */
if (caseless && (cd->ctypes[c] & ctype_letter) != 0) SET_BIT(cd->fcc[c]);
return p + 1;
#endif
}
/*************************************************
* 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_type_bits(pcre_uint8 *start_bits, int cbit_type, unsigned int table_limit,
compile_data *cd)
{
register pcre_uint32 c;
for (c = 0; c < table_limit; c++) start_bits[c] |= cd->cbits[c+cbit_type];
#if defined SUPPORT_UTF && defined COMPILE_PCRE8
if (table_limit == 32) return;
for (c = 128; c < 256; c++)
{
if ((cd->cbits[c/8] & (1 << (c&7))) != 0)
{
pcre_uchar buff[6];
(void)PRIV(ord2utf)(c, buff);
SET_BIT(buff[0]);
}
}
#endif
}
/*************************************************
* 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(pcre_uint8 *start_bits, int cbit_type, unsigned int table_limit,
compile_data *cd)
{
register pcre_uint32 c;
for (c = 0; c < table_limit; c++) start_bits[c] |= ~cd->cbits[c+cbit_type];
#if defined SUPPORT_UTF && defined COMPILE_PCRE8
if (table_limit != 32) for (c = 24; c < 32; c++) start_bits[c] = 0xff;
#endif
}
/*************************************************
* Create bitmap of starting bytes *
*************************************************/
/* This function scans a compiled unanchored expression recursively and
attempts to build a bitmap of the set of possible starting bytes. As time goes
by, we may be able to get more clever at doing this. The SSB_CONTINUE return is
useful for parenthesized groups in patterns such as (a*)b where the group
provides some optional starting bytes but scanning must continue at the outer
level to find at least one mandatory byte. At the outermost level, this
function fails unless the result is SSB_DONE.
Arguments:
code points to an expression
start_bits points to a 32-byte table, initialized to 0
utf TRUE if in UTF-8 / UTF-16 / UTF-32 mode
cd the block with char table pointers
Returns: SSB_FAIL => Failed to find any starting bytes
SSB_DONE => Found mandatory starting bytes
SSB_CONTINUE => Found optional starting bytes
SSB_UNKNOWN => Hit an unrecognized opcode
*/
static int
set_start_bits(const pcre_uchar *code, pcre_uint8 *start_bits, BOOL utf,
compile_data *cd)
{
register pcre_uint32 c;
int yield = SSB_DONE;
#if defined SUPPORT_UTF && defined COMPILE_PCRE8
int table_limit = utf? 16:32;
#else
int table_limit = 32;
#endif
#if 0
/* ========================================================================= */
/* The following comment and code was inserted in January 1999. In May 2006,
when it was observed to cause compiler warnings about unused values, I took it
out again. If anybody is still using OS/2, they will have to put it back
manually. */
/* This next statement and the later reference to dummy are here in order to
trick the optimizer of the IBM C compiler for OS/2 into generating correct
code. Apparently IBM isn't going to fix the problem, and we would rather not
disable optimization (in this module it actually makes a big difference, and
the pcre module can use all the optimization it can get). */
volatile int dummy;
/* ========================================================================= */
#endif
do
{
BOOL try_next = TRUE;
const pcre_uchar *tcode = code + 1 + LINK_SIZE;
if (*code == OP_CBRA || *code == OP_SCBRA ||
*code == OP_CBRAPOS || *code == OP_SCBRAPOS) tcode += IMM2_SIZE;
while (try_next) /* Loop for items in this branch */
{
int rc;
switch(*tcode)
{
/* If we reach something we don't understand, it means a new opcode has
been created that hasn't been added to this code. Hopefully this problem
will be discovered during testing. */
default:
return SSB_UNKNOWN;
/* Fail for a valid opcode that implies no starting bits. */
case OP_ACCEPT:
case OP_ASSERT_ACCEPT:
case OP_ALLANY:
case OP_ANY:
case OP_ANYBYTE:
case OP_CIRC:
case OP_CIRCM:
case OP_CLOSE:
case OP_COMMIT:
case OP_COND:
case OP_CREF:
case OP_DEF:
case OP_DNCREF:
case OP_DNREF:
case OP_DNREFI:
case OP_DNRREF:
case OP_DOLL:
case OP_DOLLM:
case OP_END:
case OP_EOD:
case OP_EODN:
case OP_EXTUNI:
case OP_FAIL:
case OP_MARK:
case OP_NOT:
case OP_NOTEXACT:
case OP_NOTEXACTI:
case OP_NOTI:
case OP_NOTMINPLUS:
case OP_NOTMINPLUSI:
case OP_NOTMINQUERY:
case OP_NOTMINQUERYI:
case OP_NOTMINSTAR:
case OP_NOTMINSTARI:
case OP_NOTMINUPTO:
case OP_NOTMINUPTOI:
case OP_NOTPLUS:
case OP_NOTPLUSI:
case OP_NOTPOSPLUS:
case OP_NOTPOSPLUSI:
case OP_NOTPOSQUERY:
case OP_NOTPOSQUERYI:
case OP_NOTPOSSTAR:
case OP_NOTPOSSTARI:
case OP_NOTPOSUPTO:
case OP_NOTPOSUPTOI:
case OP_NOTPROP:
case OP_NOTQUERY:
case OP_NOTQUERYI:
case OP_NOTSTAR:
case OP_NOTSTARI:
case OP_NOTUPTO:
case OP_NOTUPTOI:
case OP_NOT_HSPACE:
case OP_NOT_VSPACE:
case OP_PRUNE:
case OP_PRUNE_ARG:
case OP_RECURSE:
case OP_REF:
case OP_REFI:
case OP_REVERSE:
case OP_RREF:
case OP_SCOND:
case OP_SET_SOM:
case OP_SKIP:
case OP_SKIP_ARG:
case OP_SOD:
case OP_SOM:
case OP_THEN:
case OP_THEN_ARG:
return SSB_FAIL;
/* A "real" property test implies no starting bits, but the fake property
PT_CLIST identifies a list of characters. These lists are short, as they
are used for characters with more than one "other case", so there is no
point in recognizing them for OP_NOTPROP. */
case OP_PROP:
if (tcode[1] != PT_CLIST) return SSB_FAIL;
{
const pcre_uint32 *p = PRIV(ucd_caseless_sets) + tcode[2];
while ((c = *p++) < NOTACHAR)
{
#if defined SUPPORT_UTF && defined COMPILE_PCRE8
if (utf)
{
pcre_uchar buff[6];
(void)PRIV(ord2utf)(c, buff);
c = buff[0];
}
#endif
if (c > 0xff) SET_BIT(0xff); else SET_BIT(c);
}
}
try_next = FALSE;
break;
/* We can ignore word boundary tests. */
case OP_WORD_BOUNDARY:
case OP_NOT_WORD_BOUNDARY:
tcode++;
break;
/* If we hit a bracket or a positive lookahead assertion, recurse to set
bits from within the subpattern. If it can't find anything, we have to
give up. If it finds some mandatory character(s), we are done for this
branch. Otherwise, carry on scanning after the subpattern. */
case OP_BRA:
case OP_SBRA:
case OP_CBRA:
case OP_SCBRA:
case OP_BRAPOS:
case OP_SBRAPOS:
case OP_CBRAPOS:
case OP_SCBRAPOS:
case OP_ONCE:
case OP_ONCE_NC:
case OP_ASSERT:
rc = set_start_bits(tcode, start_bits, utf, cd);
if (rc == SSB_FAIL || rc == SSB_UNKNOWN) return rc;
if (rc == SSB_DONE) try_next = FALSE; else
{
do tcode += GET(tcode, 1); while (*tcode == OP_ALT);
tcode += 1 + LINK_SIZE;
}
break;
/* If we hit ALT or KET, it means we haven't found anything mandatory in
this branch, though we might have found something optional. For ALT, we
continue with the next alternative, but we have to arrange that the final
result from subpattern is SSB_CONTINUE rather than SSB_DONE. For KET,
return SSB_CONTINUE: if this is the top level, that indicates failure,
but after a nested subpattern, it causes scanning to continue. */
case OP_ALT:
yield = SSB_CONTINUE;
try_next = FALSE;
break;
case OP_KET:
case OP_KETRMAX:
case OP_KETRMIN:
case OP_KETRPOS:
return SSB_CONTINUE;
/* Skip over callout */
case OP_CALLOUT:
tcode += 2 + 2*LINK_SIZE;
break;
/* Skip over lookbehind and negative lookahead assertions */
case OP_ASSERT_NOT:
case OP_ASSERTBACK:
case OP_ASSERTBACK_NOT:
do tcode += GET(tcode, 1); while (*tcode == OP_ALT);
tcode += 1 + LINK_SIZE;
break;
/* BRAZERO does the bracket, but carries on. */
case OP_BRAZERO:
case OP_BRAMINZERO:
case OP_BRAPOSZERO:
rc = set_start_bits(++tcode, start_bits, utf, cd);
if (rc == SSB_FAIL || rc == SSB_UNKNOWN) return rc;
/* =========================================================================
See the comment at the head of this function concerning the next line,
which was an old fudge for the benefit of OS/2.
dummy = 1;
========================================================================= */
do tcode += GET(tcode,1); while (*tcode == OP_ALT);
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:
case OP_MINSTAR:
case OP_POSSTAR:
case OP_QUERY:
case OP_MINQUERY:
case OP_POSQUERY:
tcode = set_table_bit(start_bits, tcode + 1, FALSE, cd, utf);
break;
case OP_STARI:
case OP_MINSTARI:
case OP_POSSTARI:
case OP_QUERYI:
case OP_MINQUERYI:
case OP_POSQUERYI:
tcode = set_table_bit(start_bits, tcode + 1, TRUE, cd, utf);
break;
/* Single-char upto sets the bit and tries the next */
case OP_UPTO:
case OP_MINUPTO:
case OP_POSUPTO:
tcode = set_table_bit(start_bits, tcode + 1 + IMM2_SIZE, FALSE, cd, utf);
break;
case OP_UPTOI:
case OP_MINUPTOI:
case OP_POSUPTOI:
tcode = set_table_bit(start_bits, tcode + 1 + IMM2_SIZE, TRUE, cd, utf);
break;
/* At least one single char sets the bit and stops */
case OP_EXACT:
tcode += IMM2_SIZE;
/* Fall through */
case OP_CHAR:
case OP_PLUS:
case OP_MINPLUS:
case OP_POSPLUS:
(void)set_table_bit(start_bits, tcode + 1, FALSE, cd, utf);
try_next = FALSE;
break;
case OP_EXACTI:
tcode += IMM2_SIZE;
/* Fall through */
case OP_CHARI:
case OP_PLUSI:
case OP_MINPLUSI:
case OP_POSPLUSI:
(void)set_table_bit(start_bits, tcode + 1, TRUE, cd, utf);
try_next = FALSE;
break;
/* 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(CHAR_HT);
SET_BIT(CHAR_SPACE);
#ifdef SUPPORT_UTF
if (utf)
{
#ifdef COMPILE_PCRE8
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 */
#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
SET_BIT(0xA0);
SET_BIT(0xFF); /* For characters > 255 */
#endif /* COMPILE_PCRE[8|16|32] */
}
else
#endif /* SUPPORT_UTF */
{
#ifndef EBCDIC
SET_BIT(0xA0);
#endif /* Not EBCDIC */
#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
SET_BIT(0xFF); /* For characters > 255 */
#endif /* COMPILE_PCRE[16|32] */
}
try_next = FALSE;
break;
case OP_ANYNL:
case OP_VSPACE:
SET_BIT(CHAR_LF);
SET_BIT(CHAR_VT);
SET_BIT(CHAR_FF);
SET_BIT(CHAR_CR);
#ifdef SUPPORT_UTF
if (utf)
{
#ifdef COMPILE_PCRE8
SET_BIT(0xC2); /* For U+0085 */
SET_BIT(0xE2); /* For U+2028, U+2029 */
#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
SET_BIT(CHAR_NEL);
SET_BIT(0xFF); /* For characters > 255 */
#endif /* COMPILE_PCRE[8|16|32] */
}
else
#endif /* SUPPORT_UTF */
{
SET_BIT(CHAR_NEL);
#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
SET_BIT(0xFF); /* For characters > 255 */
#endif
}
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:
set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
try_next = FALSE;
break;
case OP_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 no longer
have to play fancy tricks because Perl added VT to its whitespace at
release 5.18. PCRE added it at release 8.34. */
case OP_NOT_WHITESPACE:
set_nottype_bits(start_bits, cbit_space, table_limit, cd);
try_next = FALSE;
break;
case OP_WHITESPACE:
set_type_bits(start_bits, cbit_space, table_limit, cd);
try_next = FALSE;
break;
case OP_NOT_WORDCHAR:
set_nottype_bits(start_bits, cbit_word, table_limit, cd);
try_next = FALSE;
break;
case OP_WORDCHAR:
set_type_bits(start_bits, cbit_word, table_limit, cd);
try_next = FALSE;
break;
/* One or more character type fudges the pointer and restarts, knowing
it will hit a single character type and stop there. */
case OP_TYPEPLUS:
case OP_TYPEMINPLUS:
case OP_TYPEPOSPLUS:
tcode++;
break;
case OP_TYPEEXACT:
tcode += 1 + IMM2_SIZE;
break;
/* Zero or more repeats of character types set the bits and then
try again. */
case OP_TYPEUPTO:
case OP_TYPEMINUPTO:
case OP_TYPEPOSUPTO:
tcode += IMM2_SIZE; /* Fall through */
case OP_TYPESTAR:
case OP_TYPEMINSTAR:
case OP_TYPEPOSSTAR:
case OP_TYPEQUERY:
case OP_TYPEMINQUERY:
case OP_TYPEPOSQUERY:
switch(tcode[1])
{
default:
case OP_ANY:
case OP_ALLANY:
return SSB_FAIL;
case OP_HSPACE:
SET_BIT(CHAR_HT);
SET_BIT(CHAR_SPACE);
#ifdef SUPPORT_UTF
if (utf)
{
#ifdef COMPILE_PCRE8
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 */
#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
SET_BIT(0xA0);
SET_BIT(0xFF); /* For characters > 255 */
#endif /* COMPILE_PCRE[8|16|32] */
}
else
#endif /* SUPPORT_UTF */
#ifndef EBCDIC
SET_BIT(0xA0);
#endif /* Not EBCDIC */
break;
case OP_ANYNL:
case OP_VSPACE:
SET_BIT(CHAR_LF);
SET_BIT(CHAR_VT);
SET_BIT(CHAR_FF);
SET_BIT(CHAR_CR);
#ifdef SUPPORT_UTF
if (utf)
{
#ifdef COMPILE_PCRE8
SET_BIT(0xC2); /* For U+0085 */
SET_BIT(0xE2); /* For U+2028, U+2029 */
#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
SET_BIT(CHAR_NEL);
SET_BIT(0xFF); /* For characters > 255 */
#endif /* COMPILE_PCRE16 */
}
else
#endif /* SUPPORT_UTF */
SET_BIT(CHAR_NEL);
break;
case OP_NOT_DIGIT:
set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
break;
case OP_DIGIT:
set_type_bits(start_bits, cbit_digit, table_limit, cd);
break;
/* The cbit_space table has vertical tab as whitespace; we no longer
have to play fancy tricks because Perl added VT to its whitespace at
release 5.18. PCRE added it at release 8.34. */
case OP_NOT_WHITESPACE:
set_nottype_bits(start_bits, cbit_space, table_limit, cd);
break;
case OP_WHITESPACE:
set_type_bits(start_bits, cbit_space, table_limit, cd);
break;
case OP_NOT_WORDCHAR:
set_nottype_bits(start_bits, cbit_word, table_limit, cd);
break;
case OP_WORDCHAR:
set_type_bits(start_bits, cbit_word, table_limit, cd);
break;
}
tcode += 2;
break;
/* Character class where all the information is in a bit map: set the
bits and either carry on or not, according to the repeat count. If it was
a negative class, and we are operating with UTF-8 characters, any byte
with a value >= 0xc4 is a potentially valid starter because it starts a
character with a value > 255. */
#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
case OP_XCLASS:
if ((tcode[1 + LINK_SIZE] & XCL_HASPROP) != 0)
return SSB_FAIL;
/* All bits are set. */
if ((tcode[1 + LINK_SIZE] & XCL_MAP) == 0 && (tcode[1 + LINK_SIZE] & XCL_NOT) != 0)
return SSB_FAIL;
#endif
/* Fall through */
case OP_NCLASS:
#if defined SUPPORT_UTF && defined COMPILE_PCRE8
if (utf)
{
start_bits[24] |= 0xf0; /* Bits for 0xc4 - 0xc8 */
memset(start_bits+25, 0xff, 7); /* Bits for 0xc9 - 0xff */
}
#endif
#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
SET_BIT(0xFF); /* For characters > 255 */
#endif
/* Fall through */
case OP_CLASS:
{
pcre_uint8 *map;
#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
map = NULL;
if (*tcode == OP_XCLASS)
{
if ((tcode[1 + LINK_SIZE] & XCL_MAP) != 0)
map = (pcre_uint8 *)(tcode + 1 + LINK_SIZE + 1);
tcode += GET(tcode, 1);
}
else
#endif
{
tcode++;
map = (pcre_uint8 *)tcode;
tcode += 32 / sizeof(pcre_uchar);
}
/* In UTF-8 mode, the bits in a bit map correspond to character
values, not to byte values. However, the bit map we are constructing is
for byte values. So we have to do a conversion for characters whose
value is > 127. In fact, there are only two possible starting bytes for
characters in the range 128 - 255. */
#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
if (map != NULL)
#endif
{
#if defined SUPPORT_UTF && defined COMPILE_PCRE8
if (utf)
{
for (c = 0; c < 16; c++) start_bits[c] |= map[c];
for (c = 128; c < 256; c++)
{
if ((map[c/8] & (1 << (c&7))) != 0)
{
int d = (c >> 6) | 0xc0; /* Set bit for this starter */
start_bits[d/8] |= (1 << (d&7)); /* and then skip on to the */
c = (c & 0xc0) + 0x40 - 1; /* next relevant character. */
}
}
}
else
#endif
{
/* In non-UTF-8 mode, the two bit maps are completely compatible. */
for (c = 0; c < 32; c++) start_bits[c] |= map[c];
}
}
/* Advance past the bit map, and act on what follows. For a zero
minimum repeat, continue; otherwise stop processing. */
switch (*tcode)
{
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRQUERY:
case OP_CRMINQUERY:
case OP_CRPOSSTAR:
case OP_CRPOSQUERY:
tcode++;
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
case OP_CRPOSRANGE:
if (GET2(tcode, 1) == 0) tcode += 1 + 2 * IMM2_SIZE;
else try_next = FALSE;
break;
default:
try_next = FALSE;
break;
}
}
break; /* End of bitmap class handling */
} /* End of switch */
} /* End of try_next loop */
code += GET(code, 1); /* Advance to next branch */
}
while (*code == OP_ALT);
return yield;
}
/*************************************************
* Study a compiled expression *
*************************************************/
/* This function is handed a compiled expression that it must study to produce
information that will speed up the matching. It returns a pcre[16]_extra block
which then gets handed back to pcre_exec().
Arguments:
re points to the compiled expression
options contains option bits
errorptr points to where to place error messages;
set NULL unless error
Returns: pointer to a pcre[16]_extra block, with study_data filled in and
the appropriate flags set;
NULL on error or if no optimization possible
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN pcre_extra * PCRE_CALL_CONVENTION
pcre_study(const pcre *external_re, int options, const char **errorptr)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN pcre16_extra * PCRE_CALL_CONVENTION
pcre16_study(const pcre16 *external_re, int options, const char **errorptr)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN pcre32_extra * PCRE_CALL_CONVENTION
pcre32_study(const pcre32 *external_re, int options, const char **errorptr)
#endif
{
int min;
int count = 0;
BOOL bits_set = FALSE;
pcre_uint8 start_bits[32];
PUBL(extra) *extra = NULL;
pcre_study_data *study;
const pcre_uint8 *tables;
pcre_uchar *code;
compile_data compile_block;
const REAL_PCRE *re = (const REAL_PCRE *)external_re;
*errorptr = NULL;
if (re == NULL || re->magic_number != MAGIC_NUMBER)
{
*errorptr = "argument is not a compiled regular expression";
return NULL;
}
if ((re->flags & PCRE_MODE) == 0)
{
#if defined COMPILE_PCRE8
*errorptr = "argument not compiled in 8 bit mode";
#elif defined COMPILE_PCRE16
*errorptr = "argument not compiled in 16 bit mode";
#elif defined COMPILE_PCRE32
*errorptr = "argument not compiled in 32 bit mode";
#endif
return NULL;
}
if ((options & ~PUBLIC_STUDY_OPTIONS) != 0)
{
*errorptr = "unknown or incorrect option bit(s) set";
return NULL;
}
code = (pcre_uchar *)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", there is no point in
seeking a list of starting bytes. */
if ((re->options & PCRE_ANCHORED) == 0 &&
(re->flags & (PCRE_FIRSTSET|PCRE_STARTLINE)) == 0)
{
int rc;
/* Set the character tables in the block that is passed around */
tables = re->tables;
#if defined COMPILE_PCRE8
if (tables == NULL)
(void)pcre_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,
(void *)(&tables));
#elif defined COMPILE_PCRE16
if (tables == NULL)
(void)pcre16_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,
(void *)(&tables));
#elif defined COMPILE_PCRE32
if (tables == NULL)
(void)pcre32_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,
(void *)(&tables));
#endif
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. */
memset(start_bits, 0, 32 * sizeof(pcre_uint8));
rc = set_start_bits(code, start_bits, (re->options & PCRE_UTF8) != 0,
&compile_block);
bits_set = rc == SSB_DONE;
if (rc == SSB_UNKNOWN)
{
*errorptr = "internal error: opcode not recognized";
return NULL;
}
}
/* Find the minimum length of subject string. */
switch(min = find_minlength(re, code, code, re->options, NULL, &count))
{
case -2: *errorptr = "internal error: missing capturing bracket"; return NULL;
case -3: *errorptr = "internal error: opcode not recognized"; return NULL;
default: break;
}
/* If a set of starting bytes has been identified, or if the minimum length is
greater than zero, or if JIT optimization has been requested, or if
PCRE_STUDY_EXTRA_NEEDED is set, get a pcre[16]_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 data set later by the calling
program. At the moment, the size of pcre_study_data is fixed. We nevertheless
save it in a field for returning via the pcre_fullinfo() function so that if it
becomes variable in the future, we don't have to change that code. */
if (bits_set || min > 0 || (options & (
#ifdef SUPPORT_JIT
PCRE_STUDY_JIT_COMPILE | PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE |
PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE |
#endif
PCRE_STUDY_EXTRA_NEEDED)) != 0)
{
extra = (PUBL(extra) *)(PUBL(malloc))
(sizeof(PUBL(extra)) + sizeof(pcre_study_data));
if (extra == NULL)
{
*errorptr = "failed to get memory";
return NULL;
}
study = (pcre_study_data *)((char *)extra + sizeof(PUBL(extra)));
extra->flags = PCRE_EXTRA_STUDY_DATA;
extra->study_data = study;
study->size = sizeof(pcre_study_data);
study->flags = 0;
/* Set the start bits always, to avoid unset memory errors if the
study data is written to a file, but set the flag only if any of the bits
are set, to save time looking when none are. */
if (bits_set)
{
study->flags |= PCRE_STUDY_MAPPED;
memcpy(study->start_bits, start_bits, sizeof(start_bits));
}
else memset(study->start_bits, 0, 32 * sizeof(pcre_uint8));
#ifdef PCRE_DEBUG
if (bits_set)
{
pcre_uint8 *ptr = start_bits;
int i;
printf("Start bits:\n");
for (i = 0; i < 32; i++)
printf("%3d: %02x%s", i * 8, *ptr++, ((i + 1) & 0x7) != 0? " " : "\n");
}
#endif
/* Always set the minlength value in the block, because the JIT compiler
makes use of it. However, don't set the bit unless the length is greater than
zero - the interpretive pcre_exec() and pcre_dfa_exec() needn't waste time
checking the zero case. */
if (min > 0)
{
study->flags |= PCRE_STUDY_MINLEN;
study->minlength = min;
}
else study->minlength = 0;
/* If JIT support was compiled and requested, attempt the JIT compilation.
If no starting bytes were found, and the minimum length is zero, and JIT
compilation fails, abandon the extra block and return NULL, unless
PCRE_STUDY_EXTRA_NEEDED is set. */
#ifdef SUPPORT_JIT
extra->executable_jit = NULL;
if ((options & PCRE_STUDY_JIT_COMPILE) != 0)
PRIV(jit_compile)(re, extra, JIT_COMPILE);
if ((options & PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE) != 0)
PRIV(jit_compile)(re, extra, JIT_PARTIAL_SOFT_COMPILE);
if ((options & PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE) != 0)
PRIV(jit_compile)(re, extra, JIT_PARTIAL_HARD_COMPILE);
if (study->flags == 0 && (extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) == 0 &&
(options & PCRE_STUDY_EXTRA_NEEDED) == 0)
{
#if defined COMPILE_PCRE8
pcre_free_study(extra);
#elif defined COMPILE_PCRE16
pcre16_free_study(extra);
#elif defined COMPILE_PCRE32
pcre32_free_study(extra);
#endif
extra = NULL;
}
#endif
}
return extra;
}
/*************************************************
* Free the study data *
*************************************************/
/* This function frees the memory that was obtained by pcre_study().
Argument: a pointer to the pcre[16]_extra block
Returns: nothing
*/
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN void
pcre_free_study(pcre_extra *extra)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN void
pcre16_free_study(pcre16_extra *extra)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN void
pcre32_free_study(pcre32_extra *extra)
#endif
{
if (extra == NULL)
return;
#ifdef SUPPORT_JIT
if ((extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 &&
extra->executable_jit != NULL)
PRIV(jit_free)(extra->executable_jit);
#endif
PUBL(free)(extra);
}
/* End of pcre_study.c */
deps/pcre/pcre_tables.c 0 → 100644
/*************************************************
* 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) 1997-2017 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.
-----------------------------------------------------------------------------
*/
#ifndef PCRE_INCLUDED
/* This module contains some fixed tables that are used by more than one of the
PCRE code modules. The tables are also #included by the pcretest program, which
uses macros to change their names from _pcre_xxx to xxxx, thereby avoiding name
clashes with the library. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
#endif /* PCRE_INCLUDED */
/* Table of sizes for the fixed-length opcodes. It's defined in a macro so that
the definition is next to the definition of the opcodes in pcre_internal.h. */
const pcre_uint8 PRIV(OP_lengths)[] = { OP_LENGTHS };
/* Tables of horizontal and vertical whitespace characters, suitable for
adding to classes. */
const pcre_uint32 PRIV(hspace_list)[] = { HSPACE_LIST };
const pcre_uint32 PRIV(vspace_list)[] = { VSPACE_LIST };
/*************************************************
* Tables for UTF-8 support *
*************************************************/
/* These are the breakpoints for different numbers of bytes in a UTF-8
character. */
#if (defined SUPPORT_UTF && defined COMPILE_PCRE8) \
|| (defined PCRE_INCLUDED && (defined SUPPORT_PCRE16 || defined SUPPORT_PCRE32))
/* These tables are also required by pcretest in 16- or 32-bit mode. */
const int PRIV(utf8_table1)[] =
{ 0x7f, 0x7ff, 0xffff, 0x1fffff, 0x3ffffff, 0x7fffffff};
const int PRIV(utf8_table1_size) = sizeof(PRIV(utf8_table1)) / sizeof(int);
/* These are the indicator bits and the mask for the data bits to set in the
first byte of a character, indexed by the number of additional bytes. */
const int PRIV(utf8_table2)[] = { 0, 0xc0, 0xe0, 0xf0, 0xf8, 0xfc};
const int PRIV(utf8_table3)[] = { 0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01};
/* Table of the number of extra bytes, indexed by the first byte masked with
0x3f. The highest number for a valid UTF-8 first byte is in fact 0x3d. */
const pcre_uint8 PRIV(utf8_table4)[] = {
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
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 };
#endif /* (SUPPORT_UTF && COMPILE_PCRE8) || (PCRE_INCLUDED && SUPPORT_PCRE[16|32])*/
#ifdef SUPPORT_UTF
/* Table to translate from particular type value to the general value. */
const pcre_uint32 PRIV(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 */
};
/* This table encodes the rules for finding the end of an extended grapheme
cluster. Every code point has a grapheme break property which is one of the
ucp_gbXX values defined in ucp.h. The 2-dimensional table is indexed by the
properties of two adjacent code points. The left property selects a word from
the table, and the right property selects a bit from that word like this:
ucp_gbtable[left-property] & (1 << right-property)
The value is non-zero if a grapheme break is NOT permitted between the relevant
two code points. The breaking rules are as follows:
1. Break at the start and end of text (pretty obviously).
2. Do not break between a CR and LF; otherwise, break before and after
controls.
3. Do not break Hangul syllable sequences, the rules for which are:
L may be followed by L, V, LV or LVT
LV or V may be followed by V or T
LVT or T may be followed by T
4. Do not break before extending characters.
The next two rules are only for extended grapheme clusters (but that's what we
are implementing).
5. Do not break before SpacingMarks.
6. Do not break after Prepend characters.
7. Otherwise, break everywhere.
*/
const pcre_uint32 PRIV(ucp_gbtable[]) = {
(1<<ucp_gbLF), /* 0 CR */
0, /* 1 LF */
0, /* 2 Control */
(1<<ucp_gbExtend)|(1<<ucp_gbSpacingMark), /* 3 Extend */
(1<<ucp_gbExtend)|(1<<ucp_gbPrepend)| /* 4 Prepend */
(1<<ucp_gbSpacingMark)|(1<<ucp_gbL)|
(1<<ucp_gbV)|(1<<ucp_gbT)|(1<<ucp_gbLV)|
(1<<ucp_gbLVT)|(1<<ucp_gbOther),
(1<<ucp_gbExtend)|(1<<ucp_gbSpacingMark), /* 5 SpacingMark */
(1<<ucp_gbExtend)|(1<<ucp_gbSpacingMark)|(1<<ucp_gbL)| /* 6 L */
(1<<ucp_gbV)|(1<<ucp_gbLV)|(1<<ucp_gbLVT),
(1<<ucp_gbExtend)|(1<<ucp_gbSpacingMark)|(1<<ucp_gbV)| /* 7 V */
(1<<ucp_gbT),
(1<<ucp_gbExtend)|(1<<ucp_gbSpacingMark)|(1<<ucp_gbT), /* 8 T */
(1<<ucp_gbExtend)|(1<<ucp_gbSpacingMark)|(1<<ucp_gbV)| /* 9 LV */
(1<<ucp_gbT),
(1<<ucp_gbExtend)|(1<<ucp_gbSpacingMark)|(1<<ucp_gbT), /* 10 LVT */
(1<<ucp_gbRegionalIndicator), /* 11 RegionalIndicator */
(1<<ucp_gbExtend)|(1<<ucp_gbSpacingMark) /* 12 Other */
};
#ifdef SUPPORT_JIT
/* This table reverses PRIV(ucp_gentype). We can save the cost
of a memory load. */
const int PRIV(ucp_typerange)[] = {
ucp_Cc, ucp_Cs,
ucp_Ll, ucp_Lu,
ucp_Mc, ucp_Mn,
ucp_Nd, ucp_No,
ucp_Pc, ucp_Ps,
ucp_Sc, ucp_So,
ucp_Zl, ucp_Zs,
};
#endif /* SUPPORT_JIT */
/* 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
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 are unlikely.
July 2008: There is now a script called maint/GenerateUtt.py that can be used
to generate this data automatically instead of maintaining it 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_Bassa_Vah0 STR_B STR_a STR_s STR_s STR_a STR_UNDERSCORE STR_V STR_a STR_h "\0"
#define STRING_Batak0 STR_B STR_a STR_t STR_a STR_k "\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_Brahmi0 STR_B STR_r STR_a STR_h STR_m STR_i "\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_Caucasian_Albanian0 STR_C STR_a STR_u STR_c STR_a STR_s STR_i STR_a STR_n STR_UNDERSCORE STR_A STR_l STR_b STR_a STR_n 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_Chakma0 STR_C STR_h STR_a STR_k STR_m STR_a "\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_Duployan0 STR_D STR_u STR_p STR_l STR_o STR_y STR_a STR_n "\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_Elbasan0 STR_E STR_l STR_b STR_a STR_s STR_a STR_n "\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_Grantha0 STR_G STR_r STR_a STR_n STR_t STR_h STR_a "\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_Khojki0 STR_K STR_h STR_o STR_j STR_k STR_i "\0"
#define STRING_Khudawadi0 STR_K STR_h STR_u STR_d STR_a STR_w STR_a STR_d STR_i "\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_A0 STR_L STR_i STR_n STR_e STR_a STR_r STR_UNDERSCORE STR_A "\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_Mahajani0 STR_M STR_a STR_h STR_a STR_j STR_a STR_n STR_i "\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_Mandaic0 STR_M STR_a STR_n STR_d STR_a STR_i STR_c "\0"
#define STRING_Manichaean0 STR_M STR_a STR_n STR_i STR_c STR_h STR_a STR_e STR_a STR_n "\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_Mende_Kikakui0 STR_M STR_e STR_n STR_d STR_e STR_UNDERSCORE STR_K STR_i STR_k STR_a STR_k STR_u STR_i "\0"
#define STRING_Meroitic_Cursive0 STR_M STR_e STR_r STR_o STR_i STR_t STR_i STR_c STR_UNDERSCORE STR_C STR_u STR_r STR_s STR_i STR_v STR_e "\0"
#define STRING_Meroitic_Hieroglyphs0 STR_M STR_e STR_r STR_o STR_i STR_t STR_i STR_c 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_Miao0 STR_M STR_i STR_a STR_o "\0"
#define STRING_Mn0 STR_M STR_n "\0"
#define STRING_Modi0 STR_M STR_o STR_d STR_i "\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_Mro0 STR_M STR_r STR_o "\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_Nabataean0 STR_N STR_a STR_b STR_a STR_t STR_a STR_e STR_a 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_North_Arabian0 STR_O STR_l STR_d STR_UNDERSCORE STR_N STR_o STR_r STR_t STR_h STR_UNDERSCORE STR_A STR_r STR_a STR_b STR_i STR_a STR_n "\0"
#define STRING_Old_Permic0 STR_O STR_l STR_d STR_UNDERSCORE STR_P STR_e STR_r STR_m 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_Pahawh_Hmong0 STR_P STR_a STR_h STR_a STR_w STR_h STR_UNDERSCORE STR_H STR_m STR_o STR_n STR_g "\0"
#define STRING_Palmyrene0 STR_P STR_a STR_l STR_m STR_y STR_r STR_e STR_n STR_e "\0"
#define STRING_Pau_Cin_Hau0 STR_P STR_a STR_u STR_UNDERSCORE STR_C STR_i STR_n STR_UNDERSCORE STR_H STR_a STR_u "\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_Psalter_Pahlavi0 STR_P STR_s STR_a STR_l STR_t STR_e STR_r STR_UNDERSCORE STR_P STR_a STR_h STR_l STR_a STR_v STR_i "\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_Sharada0 STR_S STR_h STR_a STR_r STR_a STR_d STR_a "\0"
#define STRING_Shavian0 STR_S STR_h STR_a STR_v STR_i STR_a STR_n "\0"
#define STRING_Siddham0 STR_S STR_i STR_d STR_d STR_h STR_a STR_m "\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_Sora_Sompeng0 STR_S STR_o STR_r STR_a STR_UNDERSCORE STR_S STR_o STR_m STR_p STR_e STR_n STR_g "\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_Takri0 STR_T STR_a STR_k STR_r STR_i "\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_Tirhuta0 STR_T STR_i STR_r STR_h STR_u STR_t STR_a "\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_Warang_Citi0 STR_W STR_a STR_r STR_a STR_n STR_g STR_UNDERSCORE STR_C STR_i STR_t 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_Xuc0 STR_X STR_u STR_c "\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 PRIV(utt_names)[] =
STRING_Any0
STRING_Arabic0
STRING_Armenian0
STRING_Avestan0
STRING_Balinese0
STRING_Bamum0
STRING_Bassa_Vah0
STRING_Batak0
STRING_Bengali0
STRING_Bopomofo0
STRING_Brahmi0
STRING_Braille0
STRING_Buginese0
STRING_Buhid0
STRING_C0
STRING_Canadian_Aboriginal0
STRING_Carian0
STRING_Caucasian_Albanian0
STRING_Cc0
STRING_Cf0
STRING_Chakma0
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_Duployan0
STRING_Egyptian_Hieroglyphs0
STRING_Elbasan0
STRING_Ethiopic0
STRING_Georgian0
STRING_Glagolitic0
STRING_Gothic0
STRING_Grantha0
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_Khojki0
STRING_Khudawadi0
STRING_L0
STRING_L_AMPERSAND0
STRING_Lao0
STRING_Latin0
STRING_Lepcha0
STRING_Limbu0
STRING_Linear_A0
STRING_Linear_B0
STRING_Lisu0
STRING_Ll0
STRING_Lm0
STRING_Lo0
STRING_Lt0
STRING_Lu0
STRING_Lycian0
STRING_Lydian0
STRING_M0
STRING_Mahajani0
STRING_Malayalam0
STRING_Mandaic0
STRING_Manichaean0
STRING_Mc0
STRING_Me0
STRING_Meetei_Mayek0
STRING_Mende_Kikakui0
STRING_Meroitic_Cursive0
STRING_Meroitic_Hieroglyphs0
STRING_Miao0
STRING_Mn0
STRING_Modi0
STRING_Mongolian0
STRING_Mro0
STRING_Myanmar0
STRING_N0
STRING_Nabataean0
STRING_Nd0
STRING_New_Tai_Lue0
STRING_Nko0
STRING_Nl0
STRING_No0
STRING_Ogham0
STRING_Ol_Chiki0
STRING_Old_Italic0
STRING_Old_North_Arabian0
STRING_Old_Permic0
STRING_Old_Persian0
STRING_Old_South_Arabian0
STRING_Old_Turkic0
STRING_Oriya0
STRING_Osmanya0
STRING_P0
STRING_Pahawh_Hmong0
STRING_Palmyrene0
STRING_Pau_Cin_Hau0
STRING_Pc0
STRING_Pd0
STRING_Pe0
STRING_Pf0
STRING_Phags_Pa0
STRING_Phoenician0
STRING_Pi0
STRING_Po0
STRING_Ps0
STRING_Psalter_Pahlavi0
STRING_Rejang0
STRING_Runic0
STRING_S0
STRING_Samaritan0
STRING_Saurashtra0
STRING_Sc0
STRING_Sharada0
STRING_Shavian0
STRING_Siddham0
STRING_Sinhala0
STRING_Sk0
STRING_Sm0
STRING_So0
STRING_Sora_Sompeng0
STRING_Sundanese0
STRING_Syloti_Nagri0
STRING_Syriac0
STRING_Tagalog0
STRING_Tagbanwa0
STRING_Tai_Le0
STRING_Tai_Tham0
STRING_Tai_Viet0
STRING_Takri0
STRING_Tamil0
STRING_Telugu0
STRING_Thaana0
STRING_Thai0
STRING_Tibetan0
STRING_Tifinagh0
STRING_Tirhuta0
STRING_Ugaritic0
STRING_Vai0
STRING_Warang_Citi0
STRING_Xan0
STRING_Xps0
STRING_Xsp0
STRING_Xuc0
STRING_Xwd0
STRING_Yi0
STRING_Z0
STRING_Zl0
STRING_Zp0
STRING_Zs0;
const ucp_type_table PRIV(utt)[] = {
{ 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_Bassa_Vah },
{ 53, PT_SC, ucp_Batak },
{ 59, PT_SC, ucp_Bengali },
{ 67, PT_SC, ucp_Bopomofo },
{ 76, PT_SC, ucp_Brahmi },
{ 83, PT_SC, ucp_Braille },
{ 91, PT_SC, ucp_Buginese },
{ 100, PT_SC, ucp_Buhid },
{ 106, PT_GC, ucp_C },
{ 108, PT_SC, ucp_Canadian_Aboriginal },
{ 128, PT_SC, ucp_Carian },
{ 135, PT_SC, ucp_Caucasian_Albanian },
{ 154, PT_PC, ucp_Cc },
{ 157, PT_PC, ucp_Cf },
{ 160, PT_SC, ucp_Chakma },
{ 167, PT_SC, ucp_Cham },
{ 172, PT_SC, ucp_Cherokee },
{ 181, PT_PC, ucp_Cn },
{ 184, PT_PC, ucp_Co },
{ 187, PT_SC, ucp_Common },
{ 194, PT_SC, ucp_Coptic },
{ 201, PT_PC, ucp_Cs },
{ 204, PT_SC, ucp_Cuneiform },
{ 214, PT_SC, ucp_Cypriot },
{ 222, PT_SC, ucp_Cyrillic },
{ 231, PT_SC, ucp_Deseret },
{ 239, PT_SC, ucp_Devanagari },
{ 250, PT_SC, ucp_Duployan },
{ 259, PT_SC, ucp_Egyptian_Hieroglyphs },
{ 280, PT_SC, ucp_Elbasan },
{ 288, PT_SC, ucp_Ethiopic },
{ 297, PT_SC, ucp_Georgian },
{ 306, PT_SC, ucp_Glagolitic },
{ 317, PT_SC, ucp_Gothic },
{ 324, PT_SC, ucp_Grantha },
{ 332, PT_SC, ucp_Greek },
{ 338, PT_SC, ucp_Gujarati },
{ 347, PT_SC, ucp_Gurmukhi },
{ 356, PT_SC, ucp_Han },
{ 360, PT_SC, ucp_Hangul },
{ 367, PT_SC, ucp_Hanunoo },
{ 375, PT_SC, ucp_Hebrew },
{ 382, PT_SC, ucp_Hiragana },
{ 391, PT_SC, ucp_Imperial_Aramaic },
{ 408, PT_SC, ucp_Inherited },
{ 418, PT_SC, ucp_Inscriptional_Pahlavi },
{ 440, PT_SC, ucp_Inscriptional_Parthian },
{ 463, PT_SC, ucp_Javanese },
{ 472, PT_SC, ucp_Kaithi },
{ 479, PT_SC, ucp_Kannada },
{ 487, PT_SC, ucp_Katakana },
{ 496, PT_SC, ucp_Kayah_Li },
{ 505, PT_SC, ucp_Kharoshthi },
{ 516, PT_SC, ucp_Khmer },
{ 522, PT_SC, ucp_Khojki },
{ 529, PT_SC, ucp_Khudawadi },
{ 539, PT_GC, ucp_L },
{ 541, PT_LAMP, 0 },
{ 544, PT_SC, ucp_Lao },
{ 548, PT_SC, ucp_Latin },
{ 554, PT_SC, ucp_Lepcha },
{ 561, PT_SC, ucp_Limbu },
{ 567, PT_SC, ucp_Linear_A },
{ 576, PT_SC, ucp_Linear_B },
{ 585, PT_SC, ucp_Lisu },
{ 590, PT_PC, ucp_Ll },
{ 593, PT_PC, ucp_Lm },
{ 596, PT_PC, ucp_Lo },
{ 599, PT_PC, ucp_Lt },
{ 602, PT_PC, ucp_Lu },
{ 605, PT_SC, ucp_Lycian },
{ 612, PT_SC, ucp_Lydian },
{ 619, PT_GC, ucp_M },
{ 621, PT_SC, ucp_Mahajani },
{ 630, PT_SC, ucp_Malayalam },
{ 640, PT_SC, ucp_Mandaic },
{ 648, PT_SC, ucp_Manichaean },
{ 659, PT_PC, ucp_Mc },
{ 662, PT_PC, ucp_Me },
{ 665, PT_SC, ucp_Meetei_Mayek },
{ 678, PT_SC, ucp_Mende_Kikakui },
{ 692, PT_SC, ucp_Meroitic_Cursive },
{ 709, PT_SC, ucp_Meroitic_Hieroglyphs },
{ 730, PT_SC, ucp_Miao },
{ 735, PT_PC, ucp_Mn },
{ 738, PT_SC, ucp_Modi },
{ 743, PT_SC, ucp_Mongolian },
{ 753, PT_SC, ucp_Mro },
{ 757, PT_SC, ucp_Myanmar },
{ 765, PT_GC, ucp_N },
{ 767, PT_SC, ucp_Nabataean },
{ 777, PT_PC, ucp_Nd },
{ 780, PT_SC, ucp_New_Tai_Lue },
{ 792, PT_SC, ucp_Nko },
{ 796, PT_PC, ucp_Nl },
{ 799, PT_PC, ucp_No },
{ 802, PT_SC, ucp_Ogham },
{ 808, PT_SC, ucp_Ol_Chiki },
{ 817, PT_SC, ucp_Old_Italic },
{ 828, PT_SC, ucp_Old_North_Arabian },
{ 846, PT_SC, ucp_Old_Permic },
{ 857, PT_SC, ucp_Old_Persian },
{ 869, PT_SC, ucp_Old_South_Arabian },
{ 887, PT_SC, ucp_Old_Turkic },
{ 898, PT_SC, ucp_Oriya },
{ 904, PT_SC, ucp_Osmanya },
{ 912, PT_GC, ucp_P },
{ 914, PT_SC, ucp_Pahawh_Hmong },
{ 927, PT_SC, ucp_Palmyrene },
{ 937, PT_SC, ucp_Pau_Cin_Hau },
{ 949, PT_PC, ucp_Pc },
{ 952, PT_PC, ucp_Pd },
{ 955, PT_PC, ucp_Pe },
{ 958, PT_PC, ucp_Pf },
{ 961, PT_SC, ucp_Phags_Pa },
{ 970, PT_SC, ucp_Phoenician },
{ 981, PT_PC, ucp_Pi },
{ 984, PT_PC, ucp_Po },
{ 987, PT_PC, ucp_Ps },
{ 990, PT_SC, ucp_Psalter_Pahlavi },
{ 1006, PT_SC, ucp_Rejang },
{ 1013, PT_SC, ucp_Runic },
{ 1019, PT_GC, ucp_S },
{ 1021, PT_SC, ucp_Samaritan },
{ 1031, PT_SC, ucp_Saurashtra },
{ 1042, PT_PC, ucp_Sc },
{ 1045, PT_SC, ucp_Sharada },
{ 1053, PT_SC, ucp_Shavian },
{ 1061, PT_SC, ucp_Siddham },
{ 1069, PT_SC, ucp_Sinhala },
{ 1077, PT_PC, ucp_Sk },
{ 1080, PT_PC, ucp_Sm },
{ 1083, PT_PC, ucp_So },
{ 1086, PT_SC, ucp_Sora_Sompeng },
{ 1099, PT_SC, ucp_Sundanese },
{ 1109, PT_SC, ucp_Syloti_Nagri },
{ 1122, PT_SC, ucp_Syriac },
{ 1129, PT_SC, ucp_Tagalog },
{ 1137, PT_SC, ucp_Tagbanwa },
{ 1146, PT_SC, ucp_Tai_Le },
{ 1153, PT_SC, ucp_Tai_Tham },
{ 1162, PT_SC, ucp_Tai_Viet },
{ 1171, PT_SC, ucp_Takri },
{ 1177, PT_SC, ucp_Tamil },
{ 1183, PT_SC, ucp_Telugu },
{ 1190, PT_SC, ucp_Thaana },
{ 1197, PT_SC, ucp_Thai },
{ 1202, PT_SC, ucp_Tibetan },
{ 1210, PT_SC, ucp_Tifinagh },
{ 1219, PT_SC, ucp_Tirhuta },
{ 1227, PT_SC, ucp_Ugaritic },
{ 1236, PT_SC, ucp_Vai },
{ 1240, PT_SC, ucp_Warang_Citi },
{ 1252, PT_ALNUM, 0 },
{ 1256, PT_PXSPACE, 0 },
{ 1260, PT_SPACE, 0 },
{ 1264, PT_UCNC, 0 },
{ 1268, PT_WORD, 0 },
{ 1272, PT_SC, ucp_Yi },
{ 1275, PT_GC, ucp_Z },
{ 1277, PT_PC, ucp_Zl },
{ 1280, PT_PC, ucp_Zp },
{ 1283, PT_PC, ucp_Zs }
};
const int PRIV(utt_size) = sizeof(PRIV(utt)) / sizeof(ucp_type_table);
#endif /* SUPPORT_UTF */
/* End of pcre_tables.c */
deps/pcre/pcre_ucd.c 0 → 100644
This source diff could not be displayed because it is too large. You can view the blob instead.
deps/pcre/pcre_valid_utf8.c 0 → 100644
/*************************************************
* 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) 1997-2013 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 an internal function for validating UTF-8 character
strings. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
/*************************************************
* Validate a UTF-8 string *
*************************************************/
/* This function is called (optionally) at the start of compile or match, to
check that a supposed UTF-8 string is actually valid. The early check means
that subsequent code can assume it is dealing with a valid string. The check
can be turned off for maximum performance, but the consequences of supplying an
invalid string are then undefined.
Originally, this function checked according to RFC 2279, allowing for values in
the range 0 to 0x7fffffff, up to 6 bytes long, but ensuring that they were in
the canonical format. Once somebody had pointed out RFC 3629 to me (it
obsoletes 2279), additional restrictions were applied. The values are now
limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the
subrange 0xd000 to 0xdfff is excluded. However, the format of 5-byte and 6-byte
characters is still checked.
From release 8.13 more information about the details of the error are passed
back in the returned value:
PCRE_UTF8_ERR0 No error
PCRE_UTF8_ERR1 Missing 1 byte at the end of the string
PCRE_UTF8_ERR2 Missing 2 bytes at the end of the string
PCRE_UTF8_ERR3 Missing 3 bytes at the end of the string
PCRE_UTF8_ERR4 Missing 4 bytes at the end of the string
PCRE_UTF8_ERR5 Missing 5 bytes at the end of the string
PCRE_UTF8_ERR6 2nd-byte's two top bits are not 0x80
PCRE_UTF8_ERR7 3rd-byte's two top bits are not 0x80
PCRE_UTF8_ERR8 4th-byte's two top bits are not 0x80
PCRE_UTF8_ERR9 5th-byte's two top bits are not 0x80
PCRE_UTF8_ERR10 6th-byte's two top bits are not 0x80
PCRE_UTF8_ERR11 5-byte character is not permitted by RFC 3629
PCRE_UTF8_ERR12 6-byte character is not permitted by RFC 3629
PCRE_UTF8_ERR13 4-byte character with value > 0x10ffff is not permitted
PCRE_UTF8_ERR14 3-byte character with value 0xd000-0xdfff is not permitted
PCRE_UTF8_ERR15 Overlong 2-byte sequence
PCRE_UTF8_ERR16 Overlong 3-byte sequence
PCRE_UTF8_ERR17 Overlong 4-byte sequence
PCRE_UTF8_ERR18 Overlong 5-byte sequence (won't ever occur)
PCRE_UTF8_ERR19 Overlong 6-byte sequence (won't ever occur)
PCRE_UTF8_ERR20 Isolated 0x80 byte (not within UTF-8 character)
PCRE_UTF8_ERR21 Byte with the illegal value 0xfe or 0xff
PCRE_UTF8_ERR22 Unused (was non-character)
Arguments:
string points to the string
length length of string, or -1 if the string is zero-terminated
errp pointer to an error position offset variable
Returns: = 0 if the string is a valid UTF-8 string
> 0 otherwise, setting the offset of the bad character
*/
int
PRIV(valid_utf)(PCRE_PUCHAR string, int length, int *erroroffset)
{
#ifdef SUPPORT_UTF
register PCRE_PUCHAR p;
if (length < 0)
{
for (p = string; *p != 0; p++);
length = (int)(p - string);
}
for (p = string; length-- > 0; p++)
{
register pcre_uchar ab, c, d;
c = *p;
if (c < 128) continue; /* ASCII character */
if (c < 0xc0) /* Isolated 10xx xxxx byte */
{
*erroroffset = (int)(p - string);
return PCRE_UTF8_ERR20;
}
if (c >= 0xfe) /* Invalid 0xfe or 0xff bytes */
{
*erroroffset = (int)(p - string);
return PCRE_UTF8_ERR21;
}
ab = PRIV(utf8_table4)[c & 0x3f]; /* Number of additional bytes */
if (length < ab)
{
*erroroffset = (int)(p - string); /* Missing bytes */
return ab - length; /* Codes ERR1 to ERR5 */
}
length -= ab; /* Length remaining */
/* Check top bits in the second byte */
if (((d = *(++p)) & 0xc0) != 0x80)
{
*erroroffset = (int)(p - string) - 1;
return PCRE_UTF8_ERR6;
}
/* For each length, check that the remaining bytes start with the 0x80 bit
set and not the 0x40 bit. Then check for an overlong sequence, and for the
excluded range 0xd800 to 0xdfff. */
switch (ab)
{
/* 2-byte character. No further bytes to check for 0x80. Check first byte
for for xx00 000x (overlong sequence). */
case 1: if ((c & 0x3e) == 0)
{
*erroroffset = (int)(p - string) - 1;
return PCRE_UTF8_ERR15;
}
break;
/* 3-byte character. Check third byte for 0x80. Then check first 2 bytes
for 1110 0000, xx0x xxxx (overlong sequence) or
1110 1101, 1010 xxxx (0xd800 - 0xdfff) */
case 2:
if ((*(++p) & 0xc0) != 0x80) /* Third byte */
{
*erroroffset = (int)(p - string) - 2;
return PCRE_UTF8_ERR7;
}
if (c == 0xe0 && (d & 0x20) == 0)
{
*erroroffset = (int)(p - string) - 2;
return PCRE_UTF8_ERR16;
}
if (c == 0xed && d >= 0xa0)
{
*erroroffset = (int)(p - string) - 2;
return PCRE_UTF8_ERR14;
}
break;
/* 4-byte character. Check 3rd and 4th bytes for 0x80. Then check first 2
bytes for for 1111 0000, xx00 xxxx (overlong sequence), then check for a
character greater than 0x0010ffff (f4 8f bf bf) */
case 3:
if ((*(++p) & 0xc0) != 0x80) /* Third byte */
{
*erroroffset = (int)(p - string) - 2;
return PCRE_UTF8_ERR7;
}
if ((*(++p) & 0xc0) != 0x80) /* Fourth byte */
{
*erroroffset = (int)(p - string) - 3;
return PCRE_UTF8_ERR8;
}
if (c == 0xf0 && (d & 0x30) == 0)
{
*erroroffset = (int)(p - string) - 3;
return PCRE_UTF8_ERR17;
}
if (c > 0xf4 || (c == 0xf4 && d > 0x8f))
{
*erroroffset = (int)(p - string) - 3;
return PCRE_UTF8_ERR13;
}
break;
/* 5-byte and 6-byte characters are not allowed by RFC 3629, and will be
rejected by the length test below. However, we do the appropriate tests
here so that overlong sequences get diagnosed, and also in case there is
ever an option for handling these larger code points. */
/* 5-byte character. Check 3rd, 4th, and 5th bytes for 0x80. Then check for
1111 1000, xx00 0xxx */
case 4:
if ((*(++p) & 0xc0) != 0x80) /* Third byte */
{
*erroroffset = (int)(p - string) - 2;
return PCRE_UTF8_ERR7;
}
if ((*(++p) & 0xc0) != 0x80) /* Fourth byte */
{
*erroroffset = (int)(p - string) - 3;
return PCRE_UTF8_ERR8;
}
if ((*(++p) & 0xc0) != 0x80) /* Fifth byte */
{
*erroroffset = (int)(p - string) - 4;
return PCRE_UTF8_ERR9;
}
if (c == 0xf8 && (d & 0x38) == 0)
{
*erroroffset = (int)(p - string) - 4;
return PCRE_UTF8_ERR18;
}
break;
/* 6-byte character. Check 3rd-6th bytes for 0x80. Then check for
1111 1100, xx00 00xx. */
case 5:
if ((*(++p) & 0xc0) != 0x80) /* Third byte */
{
*erroroffset = (int)(p - string) - 2;
return PCRE_UTF8_ERR7;
}
if ((*(++p) & 0xc0) != 0x80) /* Fourth byte */
{
*erroroffset = (int)(p - string) - 3;
return PCRE_UTF8_ERR8;
}
if ((*(++p) & 0xc0) != 0x80) /* Fifth byte */
{
*erroroffset = (int)(p - string) - 4;
return PCRE_UTF8_ERR9;
}
if ((*(++p) & 0xc0) != 0x80) /* Sixth byte */
{
*erroroffset = (int)(p - string) - 5;
return PCRE_UTF8_ERR10;
}
if (c == 0xfc && (d & 0x3c) == 0)
{
*erroroffset = (int)(p - string) - 5;
return PCRE_UTF8_ERR19;
}
break;
}
/* Character is valid under RFC 2279, but 4-byte and 5-byte characters are
excluded by RFC 3629. The pointer p is currently at the last byte of the
character. */
if (ab > 3)
{
*erroroffset = (int)(p - string) - ab;
return (ab == 4)? PCRE_UTF8_ERR11 : PCRE_UTF8_ERR12;
}
}
#else /* Not SUPPORT_UTF */
(void)(string); /* Keep picky compilers happy */
(void)(length);
(void)(erroroffset);
#endif
return PCRE_UTF8_ERR0; /* This indicates success */
}
/* End of pcre_valid_utf8.c */
deps/pcre/pcre_version.c 0 → 100644
/*************************************************
* 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) 1997-2012 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 the external function pcre_version(), which returns a
string that identifies the PCRE version that is in use. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
/*************************************************
* Return version string *
*************************************************/
/* These macros are the standard way of turning unquoted text into C strings.
They allow macros like PCRE_MAJOR to be defined without quotes, which is
convenient for user programs that want to test its value. */
#define STRING(a) # a
#define XSTRING(s) STRING(s)
/* A problem turned up with PCRE_PRERELEASE, which is defined empty for
production releases. Originally, it was used naively in this code:
return XSTRING(PCRE_MAJOR)
"." XSTRING(PCRE_MINOR)
XSTRING(PCRE_PRERELEASE)
" " XSTRING(PCRE_DATE);
However, when PCRE_PRERELEASE is empty, this leads to an attempted expansion of
STRING(). The C standard states: "If (before argument substitution) any
argument consists of no preprocessing tokens, the behavior is undefined." It
turns out the gcc treats this case as a single empty string - which is what we
really want - but Visual C grumbles about the lack of an argument for the
macro. Unfortunately, both are within their rights. To cope with both ways of
handling this, I had resort to some messy hackery that does a test at run time.
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. */
#if defined COMPILE_PCRE8
PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION
pcre_version(void)
#elif defined COMPILE_PCRE16
PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION
pcre16_version(void)
#elif defined COMPILE_PCRE32
PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION
pcre32_version(void)
#endif
{
return (XSTRING(Z PCRE_PRERELEASE)[1] == 0)?
XSTRING(PCRE_MAJOR.PCRE_MINOR PCRE_DATE) :
XSTRING(PCRE_MAJOR.PCRE_MINOR) XSTRING(PCRE_PRERELEASE PCRE_DATE);
}
/* End of pcre_version.c */
deps/pcre/pcre_xclass.c 0 → 100644
/*************************************************
* 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) 1997-2013 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 an internal function that is used to match an extended
class. It is used by both pcre_exec() and pcre_def_exec(). */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre_internal.h"
/*************************************************
* Match character against an XCLASS *
*************************************************/
/* This function is called to match a character against an extended class that
might contain values > 255 and/or Unicode properties.
Arguments:
c the character
data points to the flag byte of the XCLASS data
Returns: TRUE if character matches, else FALSE
*/
BOOL
PRIV(xclass)(pcre_uint32 c, const pcre_uchar *data, BOOL utf)
{
pcre_uchar t;
BOOL negated = (*data & XCL_NOT) != 0;
(void)utf;
#ifdef COMPILE_PCRE8
/* In 8 bit mode, this must always be TRUE. Help the compiler to know that. */
utf = TRUE;
#endif
/* Character values < 256 are matched against a bitmap, if one is present. If
not, we still carry on, because there may be ranges that start below 256 in the
additional data. */
if (c < 256)
{
if ((*data & XCL_HASPROP) == 0)
{
if ((*data & XCL_MAP) == 0) return negated;
return (((pcre_uint8 *)(data + 1))[c/8] & (1 << (c&7))) != 0;
}
if ((*data & XCL_MAP) != 0 &&
(((pcre_uint8 *)(data + 1))[c/8] & (1 << (c&7))) != 0)
return !negated; /* char found */
}
/* First skip the bit map if present. Then match against the list of Unicode
properties or large chars or ranges that end with a large char. We won't ever
encounter XCL_PROP or XCL_NOTPROP when UCP support is not compiled. */
if ((*data++ & XCL_MAP) != 0) data += 32 / sizeof(pcre_uchar);
while ((t = *data++) != XCL_END)
{
pcre_uint32 x, y;
if (t == XCL_SINGLE)
{
#ifdef SUPPORT_UTF
if (utf)
{
GETCHARINC(x, data); /* macro generates multiple statements */
}
else
#endif
x = *data++;
if (c == x) return !negated;
}
else if (t == XCL_RANGE)
{
#ifdef SUPPORT_UTF
if (utf)
{
GETCHARINC(x, data); /* macro generates multiple statements */
GETCHARINC(y, data); /* macro generates multiple statements */
}
else
#endif
{
x = *data++;
y = *data++;
}
if (c >= x && c <= y) return !negated;
}
#ifdef SUPPORT_UCP
else /* XCL_PROP & XCL_NOTPROP */
{
const ucd_record *prop = GET_UCD(c);
BOOL isprop = t == XCL_PROP;
switch(*data)
{
case PT_ANY:
if (isprop) return !negated;
break;
case PT_LAMP:
if ((prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
prop->chartype == ucp_Lt) == isprop) return !negated;
break;
case PT_GC:
if ((data[1] == PRIV(ucp_gentype)[prop->chartype]) == isprop)
return !negated;
break;
case PT_PC:
if ((data[1] == prop->chartype) == isprop) return !negated;
break;
case PT_SC:
if ((data[1] == prop->script) == isprop) return !negated;
break;
case PT_ALNUM:
if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
PRIV(ucp_gentype)[prop->chartype] == ucp_N) == isprop)
return !negated;
break;
/* Perl space used to exclude VT, but from Perl 5.18 it is included,
which means that Perl space and POSIX space are now identical. PCRE
was changed at release 8.34. */
case PT_SPACE: /* Perl space */
case PT_PXSPACE: /* POSIX space */
switch(c)
{
HSPACE_CASES:
VSPACE_CASES:
if (isprop) return !negated;
break;
default:
if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z) == isprop)
return !negated;
break;
}
break;
case PT_WORD:
if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
PRIV(ucp_gentype)[prop->chartype] == ucp_N || c == CHAR_UNDERSCORE)
== isprop)
return !negated;
break;
case PT_UCNC:
if (c < 0xa0)
{
if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
c == CHAR_GRAVE_ACCENT) == isprop)
return !negated;
}
else
{
if ((c < 0xd800 || c > 0xdfff) == isprop)
return !negated;
}
break;
/* The following three properties can occur only in an XCLASS, as there
is no \p or \P coding for them. */
/* Graphic character. Implement this as not Z (space or separator) and
not C (other), except for Cf (format) with a few exceptions. This seems
to be what Perl does. The exceptional characters are:
U+061C Arabic Letter Mark
U+180E Mongolian Vowel Separator
U+2066 - U+2069 Various "isolate"s
*/
case PT_PXGRAPH:
if ((PRIV(ucp_gentype)[prop->chartype] != ucp_Z &&
(PRIV(ucp_gentype)[prop->chartype] != ucp_C ||
(prop->chartype == ucp_Cf &&
c != 0x061c && c != 0x180e && (c < 0x2066 || c > 0x2069))
)) == isprop)
return !negated;
break;
/* Printable character: same as graphic, with the addition of Zs, i.e.
not Zl and not Zp, and U+180E. */
case PT_PXPRINT:
if ((prop->chartype != ucp_Zl &&
prop->chartype != ucp_Zp &&
(PRIV(ucp_gentype)[prop->chartype] != ucp_C ||
(prop->chartype == ucp_Cf &&
c != 0x061c && (c < 0x2066 || c > 0x2069))
)) == isprop)
return !negated;
break;
/* Punctuation: all Unicode punctuation, plus ASCII characters that
Unicode treats as symbols rather than punctuation, for Perl
compatibility (these are $+<=>^`|~). */
case PT_PXPUNCT:
if ((PRIV(ucp_gentype)[prop->chartype] == ucp_P ||
(c < 128 && PRIV(ucp_gentype)[prop->chartype] == ucp_S)) == isprop)
return !negated;
break;
/* This should never occur, but compilers may mutter if there is no
default. */
default:
return FALSE;
}
data += 2;
}
#endif /* SUPPORT_UCP */
}
return negated; /* char did not match */
}
/* End of pcre_xclass.c */
deps/pcre/pcreposix.c 0 → 100644
/*************************************************
* 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) 1997-2018 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 is a wrapper that provides a POSIX API to the underlying PCRE
functions. */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
/* We include pcre.h before pcre_internal.h so that the PCRE library functions
are declared as "import" for Windows by defining PCRE_EXP_DECL as "import".
This is needed even though pcre_internal.h itself includes pcre.h, because it
does so after it has set PCRE_EXP_DECL to "export" if it is not already set. */
#include "pcre.h"
#include "pcre_internal.h"
#include "pcreposix.h"
/* Table to translate PCRE compile time error codes into POSIX error codes. */
static const int eint[] = {
0, /* no error */
PCRE_REG_EESCAPE, /* \ at end of pattern */
PCRE_REG_EESCAPE, /* \c at end of pattern */
PCRE_REG_EESCAPE, /* unrecognized character follows \ */
PCRE_REG_BADBR, /* numbers out of order in {} quantifier */
/* 5 */
PCRE_REG_BADBR, /* number too big in {} quantifier */
PCRE_REG_EBRACK, /* missing terminating ] for character class */
PCRE_REG_ECTYPE, /* invalid escape sequence in character class */
PCRE_REG_ERANGE, /* range out of order in character class */
PCRE_REG_BADRPT, /* nothing to repeat */
/* 10 */
PCRE_REG_BADRPT, /* operand of unlimited repeat could match the empty string */
PCRE_REG_ASSERT, /* internal error: unexpected repeat */
PCRE_REG_BADPAT, /* unrecognized character after (? */
PCRE_REG_BADPAT, /* POSIX named classes are supported only within a class */
PCRE_REG_EPAREN, /* missing ) */
/* 15 */
PCRE_REG_ESUBREG, /* reference to non-existent subpattern */
PCRE_REG_INVARG, /* erroffset passed as NULL */
PCRE_REG_INVARG, /* unknown option bit(s) set */
PCRE_REG_EPAREN, /* missing ) after comment */
PCRE_REG_ESIZE, /* parentheses nested too deeply */
/* 20 */
PCRE_REG_ESIZE, /* regular expression too large */
PCRE_REG_ESPACE, /* failed to get memory */
PCRE_REG_EPAREN, /* unmatched parentheses */
PCRE_REG_ASSERT, /* internal error: code overflow */
PCRE_REG_BADPAT, /* unrecognized character after (?< */
/* 25 */
PCRE_REG_BADPAT, /* lookbehind assertion is not fixed length */
PCRE_REG_BADPAT, /* malformed number or name after (?( */
PCRE_REG_BADPAT, /* conditional group contains more than two branches */
PCRE_REG_BADPAT, /* assertion expected after (?( */
PCRE_REG_BADPAT, /* (?R or (?[+-]digits must be followed by ) */
/* 30 */
PCRE_REG_ECTYPE, /* unknown POSIX class name */
PCRE_REG_BADPAT, /* POSIX collating elements are not supported */
PCRE_REG_INVARG, /* this version of PCRE is not compiled with PCRE_UTF8 support */
PCRE_REG_BADPAT, /* spare error */
PCRE_REG_BADPAT, /* character value in \x{} or \o{} is too large */
/* 35 */
PCRE_REG_BADPAT, /* invalid condition (?(0) */
PCRE_REG_BADPAT, /* \C not allowed in lookbehind assertion */
PCRE_REG_EESCAPE, /* PCRE does not support \L, \l, \N, \U, or \u */
PCRE_REG_BADPAT, /* number after (?C is > 255 */
PCRE_REG_BADPAT, /* closing ) for (?C expected */
/* 40 */
PCRE_REG_BADPAT, /* recursive call could loop indefinitely */
PCRE_REG_BADPAT, /* unrecognized character after (?P */
PCRE_REG_BADPAT, /* syntax error in subpattern name (missing terminator) */
PCRE_REG_BADPAT, /* two named subpatterns have the same name */
PCRE_REG_BADPAT, /* invalid UTF-8 string */
/* 45 */
PCRE_REG_BADPAT, /* support for \P, \p, and \X has not been compiled */
PCRE_REG_BADPAT, /* malformed \P or \p sequence */
PCRE_REG_BADPAT, /* unknown property name after \P or \p */
PCRE_REG_BADPAT, /* subpattern name is too long (maximum 32 characters) */
PCRE_REG_BADPAT, /* too many named subpatterns (maximum 10,000) */
/* 50 */
PCRE_REG_BADPAT, /* repeated subpattern is too long */
PCRE_REG_BADPAT, /* octal value is greater than \377 (not in UTF-8 mode) */
PCRE_REG_BADPAT, /* internal error: overran compiling workspace */
PCRE_REG_BADPAT, /* internal error: previously-checked referenced subpattern not found */
PCRE_REG_BADPAT, /* DEFINE group contains more than one branch */
/* 55 */
PCRE_REG_BADPAT, /* repeating a DEFINE group is not allowed */
PCRE_REG_INVARG, /* inconsistent NEWLINE options */
PCRE_REG_BADPAT, /* \g is not followed followed by an (optionally braced) non-zero number */
PCRE_REG_BADPAT, /* a numbered reference must not be zero */
PCRE_REG_BADPAT, /* an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT) */
/* 60 */
PCRE_REG_BADPAT, /* (*VERB) not recognized */
PCRE_REG_BADPAT, /* number is too big */
PCRE_REG_BADPAT, /* subpattern name expected */
PCRE_REG_BADPAT, /* digit expected after (?+ */
PCRE_REG_BADPAT, /* ] is an invalid data character in JavaScript compatibility mode */
/* 65 */
PCRE_REG_BADPAT, /* different names for subpatterns of the same number are not allowed */
PCRE_REG_BADPAT, /* (*MARK) must have an argument */
PCRE_REG_INVARG, /* this version of PCRE is not compiled with PCRE_UCP support */
PCRE_REG_BADPAT, /* \c must be followed by an ASCII character */
PCRE_REG_BADPAT, /* \k is not followed by a braced, angle-bracketed, or quoted name */
/* 70 */
PCRE_REG_BADPAT, /* internal error: unknown opcode in find_fixedlength() */
PCRE_REG_BADPAT, /* \N is not supported in a class */
PCRE_REG_BADPAT, /* too many forward references */
PCRE_REG_BADPAT, /* disallowed UTF-8/16/32 code point (>= 0xd800 && <= 0xdfff) */
PCRE_REG_BADPAT, /* invalid UTF-16 string (should not occur) */
/* 75 */
PCRE_REG_BADPAT, /* overlong MARK name */
PCRE_REG_BADPAT, /* character value in \u.... sequence is too large */
PCRE_REG_BADPAT, /* invalid UTF-32 string (should not occur) */
PCRE_REG_BADPAT, /* setting UTF is disabled by the application */
PCRE_REG_BADPAT, /* non-hex character in \\x{} (closing brace missing?) */
/* 80 */
PCRE_REG_BADPAT, /* non-octal character in \o{} (closing brace missing?) */
PCRE_REG_BADPAT, /* missing opening brace after \o */
PCRE_REG_BADPAT, /* parentheses too deeply nested */
PCRE_REG_BADPAT, /* invalid range in character class */
PCRE_REG_BADPAT, /* group name must start with a non-digit */
/* 85 */
PCRE_REG_BADPAT, /* parentheses too deeply nested (stack check) */
PCRE_REG_BADPAT, /* missing digits in \x{} or \o{} */
PCRE_REG_BADPAT /* pattern too complicated */
};
/* Table of texts corresponding to POSIX error codes */
static const char *const pstring[] = {
"", /* Dummy for value 0 */
"internal error", /* REG_ASSERT */
"invalid repeat counts in {}", /* BADBR */
"pattern error", /* BADPAT */
"? * + invalid", /* BADRPT */
"unbalanced {}", /* EBRACE */
"unbalanced []", /* EBRACK */
"collation error - not relevant", /* ECOLLATE */
"bad class", /* ECTYPE */
"bad escape sequence", /* EESCAPE */
"empty expression", /* EMPTY */
"unbalanced ()", /* EPAREN */
"bad range inside []", /* ERANGE */
"expression too big", /* ESIZE */
"failed to get memory", /* ESPACE */
"bad back reference", /* ESUBREG */
"bad argument", /* INVARG */
"match failed" /* NOMATCH */
};
/*************************************************
* Translate error code to string *
*************************************************/
PCREPOSIX_EXP_DEFN size_t PCRE_CALL_CONVENTION
pcre_regerror(int errcode, const pcre_regex_t *preg, char *errbuf, size_t errbuf_size)
{
const char *message, *addmessage;
size_t length, addlength;
message = (errcode >= (int)(sizeof(pstring)/sizeof(char *)))?
"unknown error code" : pstring[errcode];
length = strlen(message) + 1;
addmessage = " at offset ";
addlength = (preg != NULL && (int)preg->re_erroffset != -1)?
strlen(addmessage) + 6 : 0;
if (errbuf_size > 0)
{
if (addlength > 0 && errbuf_size >= length + addlength)
sprintf(errbuf, "%s%s%-6d", message, addmessage, (int)preg->re_erroffset);
else
{
strncpy(errbuf, message, errbuf_size - 1);
errbuf[errbuf_size-1] = 0;
}
}
return length + addlength;
}
/*************************************************
* Free store held by a regex *
*************************************************/
PCREPOSIX_EXP_DEFN void PCRE_CALL_CONVENTION
pcre_regfree(pcre_regex_t *preg)
{
(PUBL(free))(preg->re_pcre);
}
/*************************************************
* Compile a regular expression *
*************************************************/
/*
Arguments:
preg points to a structure for recording the compiled expression
pattern the pattern to compile
cflags compilation flags
Returns: 0 on success
various non-zero codes on failure
*/
PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_regcomp(pcre_regex_t *preg, const char *pattern, int cflags)
{
const char *errorptr;
int erroffset;
int errorcode;
int options = 0;
int re_nsub = 0;
if ((cflags & PCRE_REG_ICASE) != 0) options |= PCRE_CASELESS;
if ((cflags & PCRE_REG_NEWLINE) != 0) options |= PCRE_MULTILINE;
if ((cflags & PCRE_REG_DOTALL) != 0) options |= PCRE_DOTALL;
if ((cflags & PCRE_REG_NOSUB) != 0) options |= PCRE_NO_AUTO_CAPTURE;
if ((cflags & PCRE_REG_UTF8) != 0) options |= PCRE_UTF8;
if ((cflags & PCRE_REG_UCP) != 0) options |= PCRE_UCP;
if ((cflags & PCRE_REG_UNGREEDY) != 0) options |= PCRE_UNGREEDY;
preg->re_pcre = pcre_compile2(pattern, options, &errorcode, &errorptr,
&erroffset, NULL);
preg->re_erroffset = erroffset;
/* Safety: if the error code is too big for the translation vector (which
should not happen, but we all make mistakes), return REG_BADPAT. */
if (preg->re_pcre == NULL)
{
return (errorcode < (int)(sizeof(eint)/sizeof(const int)))?
eint[errorcode] : PCRE_REG_BADPAT;
}
(void)pcre_fullinfo((const pcre *)preg->re_pcre, NULL, PCRE_INFO_CAPTURECOUNT,
&re_nsub);
preg->re_nsub = (size_t)re_nsub;
return 0;
}
/*************************************************
* Match a regular expression *
*************************************************/
/* Unfortunately, PCRE requires 3 ints of working space for each captured
substring, so we have to get and release working store instead of just using
the POSIX structures as was done in earlier releases when PCRE needed only 2
ints. However, if the number of possible capturing brackets is small, use a
block of store on the stack, to reduce the use of malloc/free. The threshold is
in a macro that can be changed at configure time.
If REG_NOSUB was specified at compile time, the PCRE_NO_AUTO_CAPTURE flag will
be set. When this is the case, the nmatch and pmatch arguments are ignored, and
the only result is yes/no/error. */
PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION
pcre_regexec(const pcre_regex_t *preg, const char *string, size_t nmatch,
pcre_regmatch_t pmatch[], int eflags)
{
int rc, so, eo;
int options = 0;
int *ovector = NULL;
int small_ovector[POSIX_MALLOC_THRESHOLD * 3];
BOOL allocated_ovector = FALSE;
BOOL nosub =
(REAL_PCRE_OPTIONS((const pcre *)preg->re_pcre) & PCRE_NO_AUTO_CAPTURE) != 0;
if ((eflags & PCRE_REG_NOTBOL) != 0) options |= PCRE_NOTBOL;
if ((eflags & PCRE_REG_NOTEOL) != 0) options |= PCRE_NOTEOL;
if ((eflags & PCRE_REG_NOTEMPTY) != 0) options |= PCRE_NOTEMPTY;
((pcre_regex_t *)preg)->re_erroffset = (size_t)(-1); /* Only has meaning after compile */
/* When no string data is being returned, or no vector has been passed in which
to put it, ensure that nmatch is zero. Otherwise, ensure the vector for holding
the return data is large enough. */
if (nosub || pmatch == NULL) nmatch = 0;
else if (nmatch > 0)
{
if (nmatch <= POSIX_MALLOC_THRESHOLD)
{
ovector = &(small_ovector[0]);
}
else
{
if (nmatch > INT_MAX/(sizeof(int) * 3)) return PCRE_REG_ESPACE;
ovector = (int *)malloc(sizeof(int) * nmatch * 3);
if (ovector == NULL) return PCRE_REG_ESPACE;
allocated_ovector = TRUE;
}
}
/* REG_STARTEND is a BSD extension, to allow for non-NUL-terminated strings.
The man page from OS X says "REG_STARTEND affects only the location of the
string, not how it is matched". That is why the "so" value is used to bump the
start location rather than being passed as a PCRE "starting offset". */
if ((eflags & PCRE_REG_STARTEND) != 0)
{
if (pmatch == NULL) return PCRE_REG_INVARG;
so = pmatch[0].rm_so;
eo = pmatch[0].rm_eo;
}
else
{
so = 0;
eo = (int)strlen(string);
}
rc = pcre_exec((const pcre *)preg->re_pcre, NULL, string + so, (eo - so),
0, options, ovector, (int)(nmatch * 3));
if (rc == 0) rc = (int)nmatch; /* All captured slots were filled in */
/* Successful match */
if (rc >= 0)
{
size_t i;
if (!nosub)
{
for (i = 0; i < (size_t)rc; i++)
{
pmatch[i].rm_so = (ovector[i*2] < 0)? -1 : ovector[i*2] + so;
pmatch[i].rm_eo = (ovector[i*2+1] < 0)? -1: ovector[i*2+1] + so;
}
if (allocated_ovector) free(ovector);
for (; i < nmatch; i++) pmatch[i].rm_so = pmatch[i].rm_eo = -1;
}
return 0;
}
/* Unsuccessful match */
if (allocated_ovector) free(ovector);
switch(rc)
{
/* ========================================================================== */
/* These cases are never obeyed. This is a fudge that causes a compile-time
error if the vector eint, which is indexed by compile-time error number, is
not the correct length. It seems to be the only way to do such a check at
compile time, as the sizeof() operator does not work in the C preprocessor.
As all the PCRE_ERROR_xxx values are negative, we can use 0 and 1. */
case 0:
case (sizeof(eint)/sizeof(int) == ERRCOUNT):
return PCRE_REG_ASSERT;
/* ========================================================================== */
case PCRE_ERROR_NOMATCH: return PCRE_REG_NOMATCH;
case PCRE_ERROR_NULL: return PCRE_REG_INVARG;
case PCRE_ERROR_BADOPTION: return PCRE_REG_INVARG;
case PCRE_ERROR_BADMAGIC: return PCRE_REG_INVARG;
case PCRE_ERROR_UNKNOWN_NODE: return PCRE_REG_ASSERT;
case PCRE_ERROR_NOMEMORY: return PCRE_REG_ESPACE;
case PCRE_ERROR_MATCHLIMIT: return PCRE_REG_ESPACE;
case PCRE_ERROR_BADUTF8: return PCRE_REG_INVARG;
case PCRE_ERROR_BADUTF8_OFFSET: return PCRE_REG_INVARG;
case PCRE_ERROR_BADMODE: return PCRE_REG_INVARG;
default: return PCRE_REG_ASSERT;
}
}
/* End of pcreposix.c */
deps/pcre/pcreposix.h 0 → 100644
/*************************************************
* Perl-Compatible Regular Expressions *
*************************************************/
#ifndef _PCREPOSIX_H
#define _PCREPOSIX_H
/* This is the header for the POSIX wrapper interface to the PCRE Perl-
Compatible Regular Expression library. It defines the things POSIX says should
be there. I hope.
Copyright (c) 1997-2012 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.
-----------------------------------------------------------------------------
*/
/* Have to include stdlib.h in order to ensure that size_t is defined. */
#include <stdlib.h>
#define PCREPOSIX_EXP_DEFN extern
#define PCREPOSIX_EXP_DECL extern
/* Options, mostly defined by POSIX, but with some extras. */
#define PCRE_REG_ICASE 0x0001 /* Maps to PCRE_CASELESS */
#define PCRE_REG_NEWLINE 0x0002 /* Maps to PCRE_MULTILINE */
#define PCRE_REG_NOTBOL 0x0004 /* Maps to PCRE_NOTBOL */
#define PCRE_REG_NOTEOL 0x0008 /* Maps to PCRE_NOTEOL */
#define PCRE_REG_DOTALL 0x0010 /* NOT defined by POSIX; maps to PCRE_DOTALL */
#define PCRE_REG_NOSUB 0x0020 /* Maps to PCRE_NO_AUTO_CAPTURE */
#define PCRE_REG_UTF8 0x0040 /* NOT defined by POSIX; maps to PCRE_UTF8 */
#define PCRE_REG_STARTEND 0x0080 /* BSD feature: pass subject string by so,eo */
#define PCRE_REG_NOTEMPTY 0x0100 /* NOT defined by POSIX; maps to PCRE_NOTEMPTY */
#define PCRE_REG_UNGREEDY 0x0200 /* NOT defined by POSIX; maps to PCRE_UNGREEDY */
#define PCRE_REG_UCP 0x0400 /* NOT defined by POSIX; maps to PCRE_UCP */
/* This is not used by PCRE, but by defining it we make it easier
to slot PCRE into existing programs that make POSIX calls. */
#define PCRE_REG_EXTENDED 0
/* Error values. Not all these are relevant or used by the wrapper. */
enum {
PCRE_REG_ASSERT = 1, /* internal error ? */
PCRE_REG_BADBR, /* invalid repeat counts in {} */
PCRE_REG_BADPAT, /* pattern error */
PCRE_REG_BADRPT, /* ? * + invalid */
PCRE_REG_EBRACE, /* unbalanced {} */
PCRE_REG_EBRACK, /* unbalanced [] */
PCRE_REG_ECOLLATE, /* collation error - not relevant */
PCRE_REG_ECTYPE, /* bad class */
PCRE_REG_EESCAPE, /* bad escape sequence */
PCRE_REG_EMPTY, /* empty expression */
PCRE_REG_EPAREN, /* unbalanced () */
PCRE_REG_ERANGE, /* bad range inside [] */
PCRE_REG_ESIZE, /* expression too big */
PCRE_REG_ESPACE, /* failed to get memory */
PCRE_REG_ESUBREG, /* bad back reference */
PCRE_REG_INVARG, /* bad argument */
PCRE_REG_NOMATCH /* match failed */
};
/* The structure representing a compiled regular expression. */
typedef struct {
void *re_pcre;
size_t re_nsub;
size_t re_erroffset;
} pcre_regex_t;
/* The structure in which a captured offset is returned. */
typedef int pcre_regoff_t;
typedef struct {
pcre_regoff_t rm_so;
pcre_regoff_t rm_eo;
} pcre_regmatch_t;
/* The functions */
PCREPOSIX_EXP_DECL int pcre_regcomp(pcre_regex_t *, const char *, int);
PCREPOSIX_EXP_DECL int pcre_regexec(const pcre_regex_t *, const char *, size_t,
pcre_regmatch_t *, int);
PCREPOSIX_EXP_DECL size_t pcre_regerror(int, const pcre_regex_t *, char *, size_t);
PCREPOSIX_EXP_DECL void pcre_regfree(pcre_regex_t *);
#endif /* End of pcreposix.h */
deps/pcre/ucp.h 0 → 100644
/*************************************************
* Unicode Property Table handler *
*************************************************/
#ifndef _UCP_H
#define _UCP_H
/* This file contains definitions of the property values that are returned by
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.
IMPORTANT: Note also that the specific numeric values of the enums have to be
the same as the values that are generated by the maint/MultiStage2.py script,
where the equivalent property descriptive names are listed in vectors.
ALSO: The specific values of the first two enums are assumed for the table
called catposstab in pcre_compile.c. */
/* These are the general character categories. */
enum {
ucp_C, /* Other */
ucp_L, /* Letter */
ucp_M, /* Mark */
ucp_N, /* Number */
ucp_P, /* Punctuation */
ucp_S, /* Symbol */
ucp_Z /* Separator */
};
/* These are the particular character categories. */
enum {
ucp_Cc, /* Control */
ucp_Cf, /* Format */
ucp_Cn, /* Unassigned */
ucp_Co, /* Private use */
ucp_Cs, /* Surrogate */
ucp_Ll, /* Lower case letter */
ucp_Lm, /* Modifier letter */
ucp_Lo, /* Other letter */
ucp_Lt, /* Title case letter */
ucp_Lu, /* Upper case letter */
ucp_Mc, /* Spacing mark */
ucp_Me, /* Enclosing mark */
ucp_Mn, /* Non-spacing mark */
ucp_Nd, /* Decimal number */
ucp_Nl, /* Letter number */
ucp_No, /* Other number */
ucp_Pc, /* Connector punctuation */
ucp_Pd, /* Dash punctuation */
ucp_Pe, /* Close punctuation */
ucp_Pf, /* Final punctuation */
ucp_Pi, /* Initial punctuation */
ucp_Po, /* Other punctuation */
ucp_Ps, /* Open punctuation */
ucp_Sc, /* Currency symbol */
ucp_Sk, /* Modifier symbol */
ucp_Sm, /* Mathematical symbol */
ucp_So, /* Other symbol */
ucp_Zl, /* Line separator */
ucp_Zp, /* Paragraph separator */
ucp_Zs /* Space separator */
};
/* These are grapheme break properties. Note that the code for processing them
assumes that the values are less than 16. If more values are added that take
the number to 16 or more, the code will have to be rewritten. */
enum {
ucp_gbCR, /* 0 */
ucp_gbLF, /* 1 */
ucp_gbControl, /* 2 */
ucp_gbExtend, /* 3 */
ucp_gbPrepend, /* 4 */
ucp_gbSpacingMark, /* 5 */
ucp_gbL, /* 6 Hangul syllable type L */
ucp_gbV, /* 7 Hangul syllable type V */
ucp_gbT, /* 8 Hangul syllable type T */
ucp_gbLV, /* 9 Hangul syllable type LV */
ucp_gbLVT, /* 10 Hangul syllable type LVT */
ucp_gbRegionalIndicator, /* 11 */
ucp_gbOther /* 12 */
};
/* These are the script identifications. */
enum {
ucp_Arabic,
ucp_Armenian,
ucp_Bengali,
ucp_Bopomofo,
ucp_Braille,
ucp_Buginese,
ucp_Buhid,
ucp_Canadian_Aboriginal,
ucp_Cherokee,
ucp_Common,
ucp_Coptic,
ucp_Cypriot,
ucp_Cyrillic,
ucp_Deseret,
ucp_Devanagari,
ucp_Ethiopic,
ucp_Georgian,
ucp_Glagolitic,
ucp_Gothic,
ucp_Greek,
ucp_Gujarati,
ucp_Gurmukhi,
ucp_Han,
ucp_Hangul,
ucp_Hanunoo,
ucp_Hebrew,
ucp_Hiragana,
ucp_Inherited,
ucp_Kannada,
ucp_Katakana,
ucp_Kharoshthi,
ucp_Khmer,
ucp_Lao,
ucp_Latin,
ucp_Limbu,
ucp_Linear_B,
ucp_Malayalam,
ucp_Mongolian,
ucp_Myanmar,
ucp_New_Tai_Lue,
ucp_Ogham,
ucp_Old_Italic,
ucp_Old_Persian,
ucp_Oriya,
ucp_Osmanya,
ucp_Runic,
ucp_Shavian,
ucp_Sinhala,
ucp_Syloti_Nagri,
ucp_Syriac,
ucp_Tagalog,
ucp_Tagbanwa,
ucp_Tai_Le,
ucp_Tamil,
ucp_Telugu,
ucp_Thaana,
ucp_Thai,
ucp_Tibetan,
ucp_Tifinagh,
ucp_Ugaritic,
ucp_Yi,
/* 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,
/* New for Unicode 6.0.0: */
ucp_Batak,
ucp_Brahmi,
ucp_Mandaic,
/* New for Unicode 6.1.0: */
ucp_Chakma,
ucp_Meroitic_Cursive,
ucp_Meroitic_Hieroglyphs,
ucp_Miao,
ucp_Sharada,
ucp_Sora_Sompeng,
ucp_Takri,
/* New for Unicode 7.0.0: */
ucp_Bassa_Vah,
ucp_Caucasian_Albanian,
ucp_Duployan,
ucp_Elbasan,
ucp_Grantha,
ucp_Khojki,
ucp_Khudawadi,
ucp_Linear_A,
ucp_Mahajani,
ucp_Manichaean,
ucp_Mende_Kikakui,
ucp_Modi,
ucp_Mro,
ucp_Nabataean,
ucp_Old_North_Arabian,
ucp_Old_Permic,
ucp_Pahawh_Hmong,
ucp_Palmyrene,
ucp_Psalter_Pahlavi,
ucp_Pau_Cin_Hau,
ucp_Siddham,
ucp_Tirhuta,
ucp_Warang_Citi
};
#endif
/* End of ucp.h */
deps/regex/CMakeLists.txt deleted 100644 → 0
INCLUDE_DIRECTORIES(".")
ADD_LIBRARY(regex OBJECT "regex.c" "regex.h")
deps/regex/COPYING deleted 100644 → 0
GNU LESSER GENERAL PUBLIC LICENSE
Version 2.1, February 1999
Copyright (C) 1991, 1999 Free Software Foundation, Inc.
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
[This is the first released version of the Lesser GPL. It also counts
as the successor of the GNU Library Public License, version 2, hence
the version number 2.1.]
Preamble
The licenses for most software are designed to take away your
freedom to share and change it. By contrast, the GNU General Public
Licenses are intended to guarantee your freedom to share and change
free software--to make sure the software is free for all its users.
This license, the Lesser General Public License, applies to some
specially designated software packages--typically libraries--of the
Free Software Foundation and other authors who decide to use it. You
can use it too, but we suggest you first think carefully about whether
this license or the ordinary General Public License is the better
strategy to use in any particular case, based on the explanations below.
When we speak of free software, we are referring to freedom of use,
not price. Our General Public Licenses are designed to make sure that
you have the freedom to distribute copies of free software (and charge
for this service if you wish); that you receive source code or can get
it if you want it; that you can change the software and use pieces of
it in new free programs; and that you are informed that you can do
these things.
To protect your rights, we need to make restrictions that forbid
distributors to deny you these rights or to ask you to surrender these
rights. These restrictions translate to certain responsibilities for
you if you distribute copies of the library or if you modify it.
For example, if you distribute copies of the library, whether gratis
or for a fee, you must give the recipients all the rights that we gave
you. You must make sure that they, too, receive or can get the source
code. If you link other code with the library, you must provide
complete object files to the recipients, so that they can relink them
with the library after making changes to the library and recompiling
it. And you must show them these terms so they know their rights.
We protect your rights with a two-step method: (1) we copyright the
library, and (2) we offer you this license, which gives you legal
permission to copy, distribute and/or modify the library.
To protect each distributor, we want to make it very clear that
there is no warranty for the free library. Also, if the library is
modified by someone else and passed on, the recipients should know
that what they have is not the original version, so that the original
author's reputation will not be affected by problems that might be
introduced by others.
Finally, software patents pose a constant threat to the existence of
any free program. We wish to make sure that a company cannot
effectively restrict the users of a free program by obtaining a
restrictive license from a patent holder. Therefore, we insist that
any patent license obtained for a version of the library must be
consistent with the full freedom of use specified in this license.
Most GNU software, including some libraries, is covered by the
ordinary GNU General Public License. This license, the GNU Lesser
General Public License, applies to certain designated libraries, and
is quite different from the ordinary General Public License. We use
this license for certain libraries in order to permit linking those
libraries into non-free programs.
When a program is linked with a library, whether statically or using
a shared library, the combination of the two is legally speaking a
combined work, a derivative of the original library. The ordinary
General Public License therefore permits such linking only if the
entire combination fits its criteria of freedom. The Lesser General
Public License permits more lax criteria for linking other code with
the library.
We call this license the "Lesser" General Public License because it
does Less to protect the user's freedom than the ordinary General
Public License. It also provides other free software developers Less
of an advantage over competing non-free programs. These disadvantages
are the reason we use the ordinary General Public License for many
libraries. However, the Lesser license provides advantages in certain
special circumstances.
For example, on rare occasions, there may be a special need to
encourage the widest possible use of a certain library, so that it becomes
a de-facto standard. To achieve this, non-free programs must be
allowed to use the library. A more frequent case is that a free
library does the same job as widely used non-free libraries. In this
case, there is little to gain by limiting the free library to free
software only, so we use the Lesser General Public License.
In other cases, permission to use a particular library in non-free
programs enables a greater number of people to use a large body of
free software. For example, permission to use the GNU C Library in
non-free programs enables many more people to use the whole GNU
operating system, as well as its variant, the GNU/Linux operating
system.
Although the Lesser General Public License is Less protective of the
users' freedom, it does ensure that the user of a program that is
linked with the Library has the freedom and the wherewithal to run
that program using a modified version of the Library.
The precise terms and conditions for copying, distribution and
modification follow. Pay close attention to the difference between a
"work based on the library" and a "work that uses the library". The
former contains code derived from the library, whereas the latter must
be combined with the library in order to run.
GNU LESSER GENERAL PUBLIC LICENSE
TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
0. This License Agreement applies to any software library or other
program which contains a notice placed by the copyright holder or
other authorized party saying it may be distributed under the terms of
this Lesser General Public License (also called "this License").
Each licensee is addressed as "you".
A "library" means a collection of software functions and/or data
prepared so as to be conveniently linked with application programs
(which use some of those functions and data) to form executables.
The "Library", below, refers to any such software library or work
which has been distributed under these terms. A "work based on the
Library" means either the Library or any derivative work under
copyright law: that is to say, a work containing the Library or a
portion of it, either verbatim or with modifications and/or translated
straightforwardly into another language. (Hereinafter, translation is
included without limitation in the term "modification".)
"Source code" for a work means the preferred form of the work for
making modifications to it. For a library, complete source code means
all the source code for all modules it contains, plus any associated
interface definition files, plus the scripts used to control compilation
and installation of the library.
Activities other than copying, distribution and modification are not
covered by this License; they are outside its scope. The act of
running a program using the Library is not restricted, and output from
such a program is covered only if its contents constitute a work based
on the Library (independent of the use of the Library in a tool for
writing it). Whether that is true depends on what the Library does
and what the program that uses the Library does.
1. You may copy and distribute verbatim copies of the Library's
complete source code as you receive it, in any medium, provided that
you conspicuously and appropriately publish on each copy an
appropriate copyright notice and disclaimer of warranty; keep intact
all the notices that refer to this License and to the absence of any
warranty; and distribute a copy of this License along with the
Library.
You may charge a fee for the physical act of transferring a copy,
and you may at your option offer warranty protection in exchange for a
fee.
2. You may modify your copy or copies of the Library or any portion
of it, thus forming a work based on the Library, and copy and
distribute such modifications or work under the terms of Section 1
above, provided that you also meet all of these conditions:
a) The modified work must itself be a software library.
b) You must cause the files modified to carry prominent notices
stating that you changed the files and the date of any change.
c) You must cause the whole of the work to be licensed at no
charge to all third parties under the terms of this License.
d) If a facility in the modified Library refers to a function or a
table of data to be supplied by an application program that uses
the facility, other than as an argument passed when the facility
is invoked, then you must make a good faith effort to ensure that,
in the event an application does not supply such function or
table, the facility still operates, and performs whatever part of
its purpose remains meaningful.
(For example, a function in a library to compute square roots has
a purpose that is entirely well-defined independent of the
application. Therefore, Subsection 2d requires that any
application-supplied function or table used by this function must
be optional: if the application does not supply it, the square
root function must still compute square roots.)
These requirements apply to the modified work as a whole. If
identifiable sections of that work are not derived from the Library,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works. But when you
distribute the same sections as part of a whole which is a work based
on the Library, the distribution of the whole must be on the terms of
this License, whose permissions for other licensees extend to the
entire whole, and thus to each and every part regardless of who wrote
it.
Thus, it is not the intent of this section to claim rights or contest
your rights to work written entirely by you; rather, the intent is to
exercise the right to control the distribution of derivative or
collective works based on the Library.
In addition, mere aggregation of another work not based on the Library
with the Library (or with a work based on the Library) on a volume of
a storage or distribution medium does not bring the other work under
the scope of this License.
3. You may opt to apply the terms of the ordinary GNU General Public
License instead of this License to a given copy of the Library. To do
this, you must alter all the notices that refer to this License, so
that they refer to the ordinary GNU General Public License, version 2,
instead of to this License. (If a newer version than version 2 of the
ordinary GNU General Public License has appeared, then you can specify
that version instead if you wish.) Do not make any other change in
these notices.
Once this change is made in a given copy, it is irreversible for
that copy, so the ordinary GNU General Public License applies to all
subsequent copies and derivative works made from that copy.
This option is useful when you wish to copy part of the code of
the Library into a program that is not a library.
4. You may copy and distribute the Library (or a portion or
derivative of it, under Section 2) in object code or executable form
under the terms of Sections 1 and 2 above provided that you accompany
it with the complete corresponding machine-readable source code, which
must be distributed under the terms of Sections 1 and 2 above on a
medium customarily used for software interchange.
If distribution of object code is made by offering access to copy
from a designated place, then offering equivalent access to copy the
source code from the same place satisfies the requirement to
distribute the source code, even though third parties are not
compelled to copy the source along with the object code.
5. A program that contains no derivative of any portion of the
Library, but is designed to work with the Library by being compiled or
linked with it, is called a "work that uses the Library". Such a
work, in isolation, is not a derivative work of the Library, and
therefore falls outside the scope of this License.
However, linking a "work that uses the Library" with the Library
creates an executable that is a derivative of the Library (because it
contains portions of the Library), rather than a "work that uses the
library". The executable is therefore covered by this License.
Section 6 states terms for distribution of such executables.
When a "work that uses the Library" uses material from a header file
that is part of the Library, the object code for the work may be a
derivative work of the Library even though the source code is not.
Whether this is true is especially significant if the work can be
linked without the Library, or if the work is itself a library. The
threshold for this to be true is not precisely defined by law.
If such an object file uses only numerical parameters, data
structure layouts and accessors, and small macros and small inline
functions (ten lines or less in length), then the use of the object
file is unrestricted, regardless of whether it is legally a derivative
work. (Executables containing this object code plus portions of the
Library will still fall under Section 6.)
Otherwise, if the work is a derivative of the Library, you may
distribute the object code for the work under the terms of Section 6.
Any executables containing that work also fall under Section 6,
whether or not they are linked directly with the Library itself.
6. As an exception to the Sections above, you may also combine or
link a "work that uses the Library" with the Library to produce a
work containing portions of the Library, and distribute that work
under terms of your choice, provided that the terms permit
modification of the work for the customer's own use and reverse
engineering for debugging such modifications.
You must give prominent notice with each copy of the work that the
Library is used in it and that the Library and its use are covered by
this License. You must supply a copy of this License. If the work
during execution displays copyright notices, you must include the
copyright notice for the Library among them, as well as a reference
directing the user to the copy of this License. Also, you must do one
of these things:
a) Accompany the work with the complete corresponding
machine-readable source code for the Library including whatever
changes were used in the work (which must be distributed under
Sections 1 and 2 above); and, if the work is an executable linked
with the Library, with the complete machine-readable "work that
uses the Library", as object code and/or source code, so that the
user can modify the Library and then relink to produce a modified
executable containing the modified Library. (It is understood
that the user who changes the contents of definitions files in the
Library will not necessarily be able to recompile the application
to use the modified definitions.)
b) Use a suitable shared library mechanism for linking with the
Library. A suitable mechanism is one that (1) uses at run time a
copy of the library already present on the user's computer system,
rather than copying library functions into the executable, and (2)
will operate properly with a modified version of the library, if
the user installs one, as long as the modified version is
interface-compatible with the version that the work was made with.
c) Accompany the work with a written offer, valid for at
least three years, to give the same user the materials
specified in Subsection 6a, above, for a charge no more
than the cost of performing this distribution.
d) If distribution of the work is made by offering access to copy
from a designated place, offer equivalent access to copy the above
specified materials from the same place.
e) Verify that the user has already received a copy of these
materials or that you have already sent this user a copy.
For an executable, the required form of the "work that uses the
Library" must include any data and utility programs needed for
reproducing the executable from it. However, as a special exception,
the materials to be distributed need not include anything that is
normally distributed (in either source or binary form) with the major
components (compiler, kernel, and so on) of the operating system on
which the executable runs, unless that component itself accompanies
the executable.
It may happen that this requirement contradicts the license
restrictions of other proprietary libraries that do not normally
accompany the operating system. Such a contradiction means you cannot
use both them and the Library together in an executable that you
distribute.
7. You may place library facilities that are a work based on the
Library side-by-side in a single library together with other library
facilities not covered by this License, and distribute such a combined
library, provided that the separate distribution of the work based on
the Library and of the other library facilities is otherwise
permitted, and provided that you do these two things:
a) Accompany the combined library with a copy of the same work
based on the Library, uncombined with any other library
facilities. This must be distributed under the terms of the
Sections above.
b) Give prominent notice with the combined library of the fact
that part of it is a work based on the Library, and explaining
where to find the accompanying uncombined form of the same work.
8. You may not copy, modify, sublicense, link with, or distribute
the Library except as expressly provided under this License. Any
attempt otherwise to copy, modify, sublicense, link with, or
distribute the Library is void, and will automatically terminate your
rights under this License. However, parties who have received copies,
or rights, from you under this License will not have their licenses
terminated so long as such parties remain in full compliance.
9. You are not required to accept this License, since you have not
signed it. However, nothing else grants you permission to modify or
distribute the Library or its derivative works. These actions are
prohibited by law if you do not accept this License. Therefore, by
modifying or distributing the Library (or any work based on the
Library), you indicate your acceptance of this License to do so, and
all its terms and conditions for copying, distributing or modifying
the Library or works based on it.
10. Each time you redistribute the Library (or any work based on the
Library), the recipient automatically receives a license from the
original licensor to copy, distribute, link with or modify the Library
subject to these terms and conditions. You may not impose any further
restrictions on the recipients' exercise of the rights granted herein.
You are not responsible for enforcing compliance by third parties with
this License.
11. If, as a consequence of a court judgment or allegation of patent
infringement or for any other reason (not limited to patent issues),
conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License. If you cannot
distribute so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you
may not distribute the Library at all. For example, if a patent
license would not permit royalty-free redistribution of the Library by
all those who receive copies directly or indirectly through you, then
the only way you could satisfy both it and this License would be to
refrain entirely from distribution of the Library.
If any portion of this section is held invalid or unenforceable under any
particular circumstance, the balance of the section is intended to apply,
and the section as a whole is intended to apply in other circumstances.
It is not the purpose of this section to induce you to infringe any
patents or other property right claims or to contest validity of any
such claims; this section has the sole purpose of protecting the
integrity of the free software distribution system which is
implemented by public license practices. Many people have made
generous contributions to the wide range of software distributed
through that system in reliance on consistent application of that
system; it is up to the author/donor to decide if he or she is willing
to distribute software through any other system and a licensee cannot
impose that choice.
This section is intended to make thoroughly clear what is believed to
be a consequence of the rest of this License.
12. If the distribution and/or use of the Library is restricted in
certain countries either by patents or by copyrighted interfaces, the
original copyright holder who places the Library under this License may add
an explicit geographical distribution limitation excluding those countries,
so that distribution is permitted only in or among countries not thus
excluded. In such case, this License incorporates the limitation as if
written in the body of this License.
13. The Free Software Foundation may publish revised and/or new
versions of the Lesser General Public License from time to time.
Such new versions will be similar in spirit to the present version,
but may differ in detail to address new problems or concerns.
Each version is given a distinguishing version number. If the Library
specifies a version number of this License which applies to it and
"any later version", you have the option of following the terms and
conditions either of that version or of any later version published by
the Free Software Foundation. If the Library does not specify a
license version number, you may choose any version ever published by
the Free Software Foundation.
14. If you wish to incorporate parts of the Library into other free
programs whose distribution conditions are incompatible with these,
write to the author to ask for permission. For software which is
copyrighted by the Free Software Foundation, write to the Free
Software Foundation; we sometimes make exceptions for this. Our
decision will be guided by the two goals of preserving the free status
of all derivatives of our free software and of promoting the sharing
and reuse of software generally.
NO WARRANTY
15. BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO
WARRANTY FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW.
EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR
OTHER PARTIES PROVIDE THE LIBRARY "AS IS" WITHOUT WARRANTY OF ANY
KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE
LIBRARY IS WITH YOU. SHOULD THE LIBRARY PROVE DEFECTIVE, YOU ASSUME
THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
16. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN
WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY
AND/OR REDISTRIBUTE THE LIBRARY AS PERMITTED ABOVE, BE LIABLE TO YOU
FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR
CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE
LIBRARY (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING
RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A
FAILURE OF THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF
SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
DAMAGES.
END OF TERMS AND CONDITIONS
How to Apply These Terms to Your New Libraries
If you develop a new library, and you want it to be of the greatest
possible use to the public, we recommend making it free software that
everyone can redistribute and change. You can do so by permitting
redistribution under these terms (or, alternatively, under the terms of the
ordinary General Public License).
To apply these terms, attach the following notices to the library. It is
safest to attach them to the start of each source file to most effectively
convey the exclusion of warranty; and each file should have at least the
"copyright" line and a pointer to where the full notice is found.
<one line to give the library's name and a brief idea of what it does.>
Copyright (C) <year> <name of author>
This library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with this library; if not, write to the Free Software
Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Also add information on how to contact you by electronic and paper mail.
You should also get your employer (if you work as a programmer) or your
school, if any, to sign a "copyright disclaimer" for the library, if
necessary. Here is a sample; alter the names:
Yoyodyne, Inc., hereby disclaims all copyright interest in the
library `Frob' (a library for tweaking knobs) written by James Random Hacker.
<signature of Ty Coon>, 1 April 1990
Ty Coon, President of Vice
That's all there is to it!
deps/regex/config.h deleted 100644 → 0
#ifndef _REGEX_CONFIG_H_
#define _REGEX_CONFIG_H_
# define GAWK
# define NO_MBSUPPORT
#endif
deps/regex/regcomp.c deleted 100644 → 0
This source diff could not be displayed because it is too large. You can view the blob instead.
deps/regex/regex.c deleted 100644 → 0
/* Extended regular expression matching and search library.
Copyright (C) 2002, 2003, 2005 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, write to the Free
Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
02110-1301 USA. */
#include "config.h"
/* Make sure noone compiles this code with a C++ compiler. */
#ifdef __cplusplus
# error "This is C code, use a C compiler"
#endif
#ifdef _LIBC
/* We have to keep the namespace clean. */
# define regfree(preg) __regfree (preg)
# define regexec(pr, st, nm, pm, ef) __regexec (pr, st, nm, pm, ef)
# define regcomp(preg, pattern, cflags) __regcomp (preg, pattern, cflags)
# define regerror(errcode, preg, errbuf, errbuf_size) \
__regerror(errcode, preg, errbuf, errbuf_size)
# define re_set_registers(bu, re, nu, st, en) \
__re_set_registers (bu, re, nu, st, en)
# define re_match_2(bufp, string1, size1, string2, size2, pos, regs, stop) \
__re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop)
# define re_match(bufp, string, size, pos, regs) \
__re_match (bufp, string, size, pos, regs)
# define re_search(bufp, string, size, startpos, range, regs) \
__re_search (bufp, string, size, startpos, range, regs)
# define re_compile_pattern(pattern, length, bufp) \
__re_compile_pattern (pattern, length, bufp)
# define re_set_syntax(syntax) __re_set_syntax (syntax)
# define re_search_2(bufp, st1, s1, st2, s2, startpos, range, regs, stop) \
__re_search_2 (bufp, st1, s1, st2, s2, startpos, range, regs, stop)
# define re_compile_fastmap(bufp) __re_compile_fastmap (bufp)
# include "../locale/localeinfo.h"
#endif
#if defined (_MSC_VER)
#include <stdio.h> /* for size_t */
#endif
/* On some systems, limits.h sets RE_DUP_MAX to a lower value than
GNU regex allows. Include it before <regex.h>, which correctly
#undefs RE_DUP_MAX and sets it to the right value. */
#include <limits.h>
#ifdef GAWK
#undef alloca
#define alloca alloca_is_bad_you_should_never_use_it
#endif
#include <regex.h>
#include "regex_internal.h"
#include "regex_internal.c"
#ifdef GAWK
# define bool int
# ifndef true
# define true (1)
# endif
# ifndef false
# define false (0)
# endif
#endif
#include "regcomp.c"
#include "regexec.c"
/* Binary backward compatibility. */
#if _LIBC
# include <shlib-compat.h>
# if SHLIB_COMPAT (libc, GLIBC_2_0, GLIBC_2_3)
link_warning (re_max_failures, "the 're_max_failures' variable is obsolete and will go away.")
int re_max_failures = 2000;
# endif
#endif
deps/regex/regex.h deleted 100644 → 0
#include <stdio.h>
#include <stddef.h>
/* Definitions for data structures and routines for the regular
expression library.
Copyright (C) 1985,1989-93,1995-98,2000,2001,2002,2003,2005,2006,2008
Free Software Foundation, Inc.
This file is part of the GNU C Library.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, write to the Free
Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
02110-1301 USA. */
#ifndef _REGEX_H
#define _REGEX_H 1
#ifdef HAVE_STDDEF_H
#include <stddef.h>
#endif
#ifdef HAVE_SYS_TYPES_H
#include <sys/types.h>
#endif
#ifndef _LIBC
#define __USE_GNU 1
#endif
/* Allow the use in C++ code. */
#ifdef __cplusplus
extern "C" {
#endif
/* The following two types have to be signed and unsigned integer type
wide enough to hold a value of a pointer. For most ANSI compilers
ptrdiff_t and size_t should be likely OK. Still size of these two
types is 2 for Microsoft C. Ugh... */
typedef long int s_reg_t;
typedef unsigned long int active_reg_t;
/* The following bits are used to determine the regexp syntax we
recognize. The set/not-set meanings are chosen so that Emacs syntax
remains the value 0. The bits are given in alphabetical order, and
the definitions shifted by one from the previous bit; thus, when we
add or remove a bit, only one other definition need change. */
typedef unsigned long int reg_syntax_t;
#ifdef __USE_GNU
/* If this bit is not set, then \ inside a bracket expression is literal.
If set, then such a \ quotes the following character. */
# define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
/* If this bit is not set, then + and ? are operators, and \+ and \? are
literals.
If set, then \+ and \? are operators and + and ? are literals. */
# define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
/* If this bit is set, then character classes are supported. They are:
[:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
[:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
If not set, then character classes are not supported. */
# define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
/* If this bit is set, then ^ and $ are always anchors (outside bracket
expressions, of course).
If this bit is not set, then it depends:
^ is an anchor if it is at the beginning of a regular
expression or after an open-group or an alternation operator;
$ is an anchor if it is at the end of a regular expression, or
before a close-group or an alternation operator.
This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
POSIX draft 11.2 says that * etc. in leading positions is undefined.
We already implemented a previous draft which made those constructs
invalid, though, so we haven't changed the code back. */
# define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
/* If this bit is set, then special characters are always special
regardless of where they are in the pattern.
If this bit is not set, then special characters are special only in
some contexts; otherwise they are ordinary. Specifically,
* + ? and intervals are only special when not after the beginning,
open-group, or alternation operator. */
# define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
/* If this bit is set, then *, +, ?, and { cannot be first in an re or
immediately after an alternation or begin-group operator. */
# define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
/* If this bit is set, then . matches newline.
If not set, then it doesn't. */
# define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
/* If this bit is set, then . doesn't match NUL.
If not set, then it does. */
# define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
/* If this bit is set, nonmatching lists [^...] do not match newline.
If not set, they do. */
# define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
/* If this bit is set, either \{...\} or {...} defines an
interval, depending on RE_NO_BK_BRACES.
If not set, \{, \}, {, and } are literals. */
# define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
/* If this bit is set, +, ? and | aren't recognized as operators.
If not set, they are. */
# define RE_LIMITED_OPS (RE_INTERVALS << 1)
/* If this bit is set, newline is an alternation operator.
If not set, newline is literal. */
# define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
/* If this bit is set, then `{...}' defines an interval, and \{ and \}
are literals.
If not set, then `\{...\}' defines an interval. */
# define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
/* If this bit is set, (...) defines a group, and \( and \) are literals.
If not set, \(...\) defines a group, and ( and ) are literals. */
# define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
/* If this bit is set, then \<digit> matches <digit>.
If not set, then \<digit> is a back-reference. */
# define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
/* If this bit is set, then | is an alternation operator, and \| is literal.
If not set, then \| is an alternation operator, and | is literal. */
# define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
/* If this bit is set, then an ending range point collating higher
than the starting range point, as in [z-a], is invalid.
If not set, then when ending range point collates higher than the
starting range point, the range is ignored. */
# define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
/* If this bit is set, then an unmatched ) is ordinary.
If not set, then an unmatched ) is invalid. */
# define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
/* If this bit is set, succeed as soon as we match the whole pattern,
without further backtracking. */
# define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
/* If this bit is set, do not process the GNU regex operators.
If not set, then the GNU regex operators are recognized. */
# define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
/* If this bit is set, a syntactically invalid interval is treated as
a string of ordinary characters. For example, the ERE 'a{1' is
treated as 'a\{1'. */
# define RE_INVALID_INTERVAL_ORD (RE_NO_GNU_OPS << 1)
/* If this bit is set, then ignore case when matching.
If not set, then case is significant. */
# define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1)
/* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only
for ^, because it is difficult to scan the regex backwards to find
whether ^ should be special. */
# define RE_CARET_ANCHORS_HERE (RE_ICASE << 1)
/* If this bit is set, then \{ cannot be first in an bre or
immediately after an alternation or begin-group operator. */
# define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1)
/* If this bit is set, then no_sub will be set to 1 during
re_compile_pattern. */
#define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1)
#endif
/* This global variable defines the particular regexp syntax to use (for
some interfaces). When a regexp is compiled, the syntax used is
stored in the pattern buffer, so changing this does not affect
already-compiled regexps. */
extern reg_syntax_t re_syntax_options;
#ifdef __USE_GNU
/* Define combinations of the above bits for the standard possibilities.
(The [[[ comments delimit what gets put into the Texinfo file, so
don't delete them!) */
/* [[[begin syntaxes]]] */
#define RE_SYNTAX_EMACS 0
#define RE_SYNTAX_AWK \
(RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
| RE_NO_BK_PARENS | RE_NO_BK_REFS \
| RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
| RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
| RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
#define RE_SYNTAX_GNU_AWK \
((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
| RE_INVALID_INTERVAL_ORD) \
& ~(RE_DOT_NOT_NULL | RE_CONTEXT_INDEP_OPS \
| RE_CONTEXT_INVALID_OPS ))
#define RE_SYNTAX_POSIX_AWK \
(RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
| RE_INTERVALS | RE_NO_GNU_OPS \
| RE_INVALID_INTERVAL_ORD)
#define RE_SYNTAX_GREP \
(RE_BK_PLUS_QM | RE_CHAR_CLASSES \
| RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
| RE_NEWLINE_ALT)
#define RE_SYNTAX_EGREP \
(RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
| RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
| RE_NEWLINE_ALT | RE_NO_BK_PARENS \
| RE_NO_BK_VBAR)
#define RE_SYNTAX_POSIX_EGREP \
(RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES \
| RE_INVALID_INTERVAL_ORD)
/* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
#define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
#define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
/* Syntax bits common to both basic and extended POSIX regex syntax. */
#define _RE_SYNTAX_POSIX_COMMON \
(RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
| RE_INTERVALS | RE_NO_EMPTY_RANGES)
#define RE_SYNTAX_POSIX_BASIC \
(_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP)
/* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
isn't minimal, since other operators, such as \`, aren't disabled. */
#define RE_SYNTAX_POSIX_MINIMAL_BASIC \
(_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
#define RE_SYNTAX_POSIX_EXTENDED \
(_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
| RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
| RE_NO_BK_PARENS | RE_NO_BK_VBAR \
| RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
removed and RE_NO_BK_REFS is added. */
#define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
(_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
| RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
| RE_NO_BK_PARENS | RE_NO_BK_REFS \
| RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
/* [[[end syntaxes]]] */
/* Maximum number of duplicates an interval can allow. Some systems
(erroneously) define this in other header files, but we want our
value, so remove any previous define. */
# ifdef RE_DUP_MAX
# undef RE_DUP_MAX
# endif
/* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */
# define RE_DUP_MAX (0x7fff)
#endif
/* POSIX `cflags' bits (i.e., information for `regcomp'). */
/* If this bit is set, then use extended regular expression syntax.
If not set, then use basic regular expression syntax. */
#define REG_EXTENDED 1
/* If this bit is set, then ignore case when matching.
If not set, then case is significant. */
#define REG_ICASE (REG_EXTENDED << 1)
/* If this bit is set, then anchors do not match at newline
characters in the string.
If not set, then anchors do match at newlines. */
#define REG_NEWLINE (REG_ICASE << 1)
/* If this bit is set, then report only success or fail in regexec.
If not set, then returns differ between not matching and errors. */
#define REG_NOSUB (REG_NEWLINE << 1)
/* POSIX `eflags' bits (i.e., information for regexec). */
/* If this bit is set, then the beginning-of-line operator doesn't match
the beginning of the string (presumably because it's not the
beginning of a line).
If not set, then the beginning-of-line operator does match the
beginning of the string. */
#define REG_NOTBOL 1
/* Like REG_NOTBOL, except for the end-of-line. */
#define REG_NOTEOL (1 << 1)
/* Use PMATCH[0] to delimit the start and end of the search in the
buffer. */
#define REG_STARTEND (1 << 2)
/* If any error codes are removed, changed, or added, update the
`re_error_msg' table in regex.c. */
typedef enum
{
#if defined _XOPEN_SOURCE || defined __USE_XOPEN2K
REG_ENOSYS = -1, /* This will never happen for this implementation. */
#endif
REG_NOERROR = 0, /* Success. */
REG_NOMATCH, /* Didn't find a match (for regexec). */
/* POSIX regcomp return error codes. (In the order listed in the
standard.) */
REG_BADPAT, /* Invalid pattern. */
REG_ECOLLATE, /* Inalid collating element. */
REG_ECTYPE, /* Invalid character class name. */
REG_EESCAPE, /* Trailing backslash. */
REG_ESUBREG, /* Invalid back reference. */
REG_EBRACK, /* Unmatched left bracket. */
REG_EPAREN, /* Parenthesis imbalance. */
REG_EBRACE, /* Unmatched \{. */
REG_BADBR, /* Invalid contents of \{\}. */
REG_ERANGE, /* Invalid range end. */
REG_ESPACE, /* Ran out of memory. */
REG_BADRPT, /* No preceding re for repetition op. */
/* Error codes we've added. */
REG_EEND, /* Premature end. */
REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
} reg_errcode_t;
/* This data structure represents a compiled pattern. Before calling
the pattern compiler, the fields `buffer', `allocated', `fastmap',
`translate', and `no_sub' can be set. After the pattern has been
compiled, the `re_nsub' field is available. All other fields are
private to the regex routines. */
#ifndef RE_TRANSLATE_TYPE
# define __RE_TRANSLATE_TYPE unsigned char *
# ifdef __USE_GNU
# define RE_TRANSLATE_TYPE __RE_TRANSLATE_TYPE
# endif
#endif
#ifdef __USE_GNU
# define __REPB_PREFIX(name) name
#else
# define __REPB_PREFIX(name) __##name
#endif
struct re_pattern_buffer
{
/* Space that holds the compiled pattern. It is declared as
`unsigned char *' because its elements are sometimes used as
array indexes. */
unsigned char *__REPB_PREFIX(buffer);
/* Number of bytes to which `buffer' points. */
unsigned long int __REPB_PREFIX(allocated);
/* Number of bytes actually used in `buffer'. */
unsigned long int __REPB_PREFIX(used);
/* Syntax setting with which the pattern was compiled. */
reg_syntax_t __REPB_PREFIX(syntax);
/* Pointer to a fastmap, if any, otherwise zero. re_search uses the
fastmap, if there is one, to skip over impossible starting points
for matches. */
char *__REPB_PREFIX(fastmap);
/* Either a translate table to apply to all characters before
comparing them, or zero for no translation. The translation is
applied to a pattern when it is compiled and to a string when it
is matched. */
__RE_TRANSLATE_TYPE __REPB_PREFIX(translate);
/* Number of subexpressions found by the compiler. */
size_t re_nsub;
/* Zero if this pattern cannot match the empty string, one else.
Well, in truth it's used only in `re_search_2', to see whether or
not we should use the fastmap, so we don't set this absolutely
perfectly; see `re_compile_fastmap' (the `duplicate' case). */
unsigned __REPB_PREFIX(can_be_null) : 1;
/* If REGS_UNALLOCATED, allocate space in the `regs' structure
for `max (RE_NREGS, re_nsub + 1)' groups.
If REGS_REALLOCATE, reallocate space if necessary.
If REGS_FIXED, use what's there. */
#ifdef __USE_GNU
# define REGS_UNALLOCATED 0
# define REGS_REALLOCATE 1
# define REGS_FIXED 2
#endif
unsigned __REPB_PREFIX(regs_allocated) : 2;
/* Set to zero when `regex_compile' compiles a pattern; set to one
by `re_compile_fastmap' if it updates the fastmap. */
unsigned __REPB_PREFIX(fastmap_accurate) : 1;
/* If set, `re_match_2' does not return information about
subexpressions. */
unsigned __REPB_PREFIX(no_sub) : 1;
/* If set, a beginning-of-line anchor doesn't match at the beginning
of the string. */
unsigned __REPB_PREFIX(not_bol) : 1;
/* Similarly for an end-of-line anchor. */
unsigned __REPB_PREFIX(not_eol) : 1;
/* If true, an anchor at a newline matches. */
unsigned __REPB_PREFIX(newline_anchor) : 1;
};
typedef struct re_pattern_buffer regex_t;
/* Type for byte offsets within the string. POSIX mandates this. */
typedef int regoff_t;
#ifdef __USE_GNU
/* This is the structure we store register match data in. See
regex.texinfo for a full description of what registers match. */
struct re_registers
{
unsigned num_regs;
regoff_t *start;
regoff_t *end;
};
/* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
`re_match_2' returns information about at least this many registers
the first time a `regs' structure is passed. */
# ifndef RE_NREGS
# define RE_NREGS 30
# endif
#endif
/* POSIX specification for registers. Aside from the different names than
`re_registers', POSIX uses an array of structures, instead of a
structure of arrays. */
typedef struct
{
regoff_t rm_so; /* Byte offset from string's start to substring's start. */
regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
} regmatch_t;
/* Declarations for routines. */
#ifdef __USE_GNU
/* Sets the current default syntax to SYNTAX, and return the old syntax.
You can also simply assign to the `re_syntax_options' variable. */
extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
/* Compile the regular expression PATTERN, with length LENGTH
and syntax given by the global `re_syntax_options', into the buffer
BUFFER. Return NULL if successful, and an error string if not. */
extern const char *re_compile_pattern (const char *__pattern, size_t __length,
struct re_pattern_buffer *__buffer);
/* Compile a fastmap for the compiled pattern in BUFFER; used to
accelerate searches. Return 0 if successful and -2 if was an
internal error. */
extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
/* Search in the string STRING (with length LENGTH) for the pattern
compiled into BUFFER. Start searching at position START, for RANGE
characters. Return the starting position of the match, -1 for no
match, or -2 for an internal error. Also return register
information in REGS (if REGS and BUFFER->no_sub are nonzero). */
extern int re_search (struct re_pattern_buffer *__buffer, const char *__cstring,
int __length, int __start, int __range,
struct re_registers *__regs);
/* Like `re_search', but search in the concatenation of STRING1 and
STRING2. Also, stop searching at index START + STOP. */
extern int re_search_2 (struct re_pattern_buffer *__buffer,
const char *__string1, int __length1,
const char *__string2, int __length2, int __start,
int __range, struct re_registers *__regs, int __stop);
/* Like `re_search', but return how many characters in STRING the regexp
in BUFFER matched, starting at position START. */
extern int re_match (struct re_pattern_buffer *__buffer, const char *__cstring,
int __length, int __start, struct re_registers *__regs);
/* Relates to `re_match' as `re_search_2' relates to `re_search'. */
extern int re_match_2 (struct re_pattern_buffer *__buffer,
const char *__string1, int __length1,
const char *__string2, int __length2, int __start,
struct re_registers *__regs, int __stop);
/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
ENDS. Subsequent matches using BUFFER and REGS will use this memory
for recording register information. STARTS and ENDS must be
allocated with malloc, and must each be at least `NUM_REGS * sizeof
(regoff_t)' bytes long.
If NUM_REGS == 0, then subsequent matches should allocate their own
register data.
Unless this function is called, the first search or match using
PATTERN_BUFFER will allocate its own register data, without
freeing the old data. */
extern void re_set_registers (struct re_pattern_buffer *__buffer,
struct re_registers *__regs,
unsigned int __num_regs,
regoff_t *__starts, regoff_t *__ends);
#endif /* Use GNU */
#if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_BSD)
# ifndef _CRAY
/* 4.2 bsd compatibility. */
extern char *re_comp (const char *);
extern int re_exec (const char *);
# endif
#endif
/* GCC 2.95 and later have "__restrict"; C99 compilers have
"restrict", and "configure" may have defined "restrict". */
#ifndef __restrict
# if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__))
# if defined restrict || 199901L <= __STDC_VERSION__
# define __restrict restrict
# else
# define __restrict
# endif
# endif
#endif
/* gcc 3.1 and up support the [restrict] syntax. */
#ifndef __restrict_arr
# if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) \
&& !defined __GNUG__
# define __restrict_arr __restrict
# else
# define __restrict_arr
# endif
#endif
/* POSIX compatibility. */
extern int regcomp (regex_t *__restrict __preg,
const char *__restrict __pattern,
int __cflags);
extern int regexec (const regex_t *__restrict __preg,
const char *__restrict __cstring, size_t __nmatch,
regmatch_t __pmatch[__restrict_arr],
int __eflags);
extern size_t regerror (int __errcode, const regex_t *__restrict __preg,
char *__restrict __errbuf, size_t __errbuf_size);
extern void regfree (regex_t *__preg);
#ifdef __cplusplus
}
#endif /* C++ */
#endif /* regex.h */
deps/regex/regex_internal.c deleted 100644 → 0
/* Extended regular expression matching and search library.
Copyright (C) 2002-2006, 2010 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, write to the Free
Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
02110-1301 USA. */
static void re_string_construct_common (const char *str, int len,
re_string_t *pstr,
RE_TRANSLATE_TYPE trans, int icase,
const re_dfa_t *dfa) internal_function;
static re_dfastate_t *create_ci_newstate (const re_dfa_t *dfa,
const re_node_set *nodes,
unsigned int hash) internal_function;
static re_dfastate_t *create_cd_newstate (const re_dfa_t *dfa,
const re_node_set *nodes,
unsigned int context,
unsigned int hash) internal_function;
#ifdef GAWK
#undef MAX /* safety */
static size_t
MAX(size_t a, size_t b)
{
return (a > b ? a : b);
}
#endif
/* Functions for string operation. */
/* This function allocate the buffers. It is necessary to call
re_string_reconstruct before using the object. */
static reg_errcode_t
internal_function
re_string_allocate (re_string_t *pstr, const char *str, int len, int init_len,
RE_TRANSLATE_TYPE trans, int icase, const re_dfa_t *dfa)
{
reg_errcode_t ret;
int init_buf_len;
/* Ensure at least one character fits into the buffers. */
if (init_len < dfa->mb_cur_max)
init_len = dfa->mb_cur_max;
init_buf_len = (len + 1 < init_len) ? len + 1: init_len;
re_string_construct_common (str, len, pstr, trans, icase, dfa);
ret = re_string_realloc_buffers (pstr, init_buf_len);
if (BE (ret != REG_NOERROR, 0))
return ret;
pstr->word_char = dfa->word_char;
pstr->word_ops_used = dfa->word_ops_used;
pstr->mbs = pstr->mbs_allocated ? pstr->mbs : (unsigned char *) str;
pstr->valid_len = (pstr->mbs_allocated || dfa->mb_cur_max > 1) ? 0 : len;
pstr->valid_raw_len = pstr->valid_len;
return REG_NOERROR;
}
/* This function allocate the buffers, and initialize them. */
static reg_errcode_t
internal_function
re_string_construct (re_string_t *pstr, const char *str, int len,
RE_TRANSLATE_TYPE trans, int icase, const re_dfa_t *dfa)
{
reg_errcode_t ret;
memset (pstr, '\0', sizeof (re_string_t));
re_string_construct_common (str, len, pstr, trans, icase, dfa);
if (len > 0)
{
ret = re_string_realloc_buffers (pstr, len + 1);
if (BE (ret != REG_NOERROR, 0))
return ret;
}
pstr->mbs = pstr->mbs_allocated ? pstr->mbs : (unsigned char *) str;
if (icase)
{
#ifdef RE_ENABLE_I18N
if (dfa->mb_cur_max > 1)
{
while (1)
{
ret = build_wcs_upper_buffer (pstr);
if (BE (ret != REG_NOERROR, 0))
return ret;
if (pstr->valid_raw_len >= len)
break;
if (pstr->bufs_len > pstr->valid_len + dfa->mb_cur_max)
break;
ret = re_string_realloc_buffers (pstr, pstr->bufs_len * 2);
if (BE (ret != REG_NOERROR, 0))
return ret;
}
}
else
#endif /* RE_ENABLE_I18N */
build_upper_buffer (pstr);
}
else
{
#ifdef RE_ENABLE_I18N
if (dfa->mb_cur_max > 1)
build_wcs_buffer (pstr);
else
#endif /* RE_ENABLE_I18N */
{
if (trans != NULL)
re_string_translate_buffer (pstr);
else
{
pstr->valid_len = pstr->bufs_len;
pstr->valid_raw_len = pstr->bufs_len;
}
}
}
return REG_NOERROR;
}
/* Helper functions for re_string_allocate, and re_string_construct. */
static reg_errcode_t
internal_function
re_string_realloc_buffers (re_string_t *pstr, int new_buf_len)
{
#ifdef RE_ENABLE_I18N
if (pstr->mb_cur_max > 1)
{
wint_t *new_wcs;
/* Avoid overflow in realloc. */
const size_t max_object_size = MAX (sizeof (wint_t), sizeof (int));
if (BE (SIZE_MAX / max_object_size < new_buf_len, 0))
return REG_ESPACE;
new_wcs = re_realloc (pstr->wcs, wint_t, new_buf_len);
if (BE (new_wcs == NULL, 0))
return REG_ESPACE;
pstr->wcs = new_wcs;
if (pstr->offsets != NULL)
{
int *new_offsets = re_realloc (pstr->offsets, int, new_buf_len);
if (BE (new_offsets == NULL, 0))
return REG_ESPACE;
pstr->offsets = new_offsets;
}
}
#endif /* RE_ENABLE_I18N */
if (pstr->mbs_allocated)
{
unsigned char *new_mbs = re_realloc (pstr->mbs, unsigned char,
new_buf_len);
if (BE (new_mbs == NULL, 0))
return REG_ESPACE;
pstr->mbs = new_mbs;
}
pstr->bufs_len = new_buf_len;
return REG_NOERROR;
}
static void
internal_function
re_string_construct_common (const char *str, int len, re_string_t *pstr,
RE_TRANSLATE_TYPE trans, int icase,
const re_dfa_t *dfa)
{
pstr->raw_mbs = (const unsigned char *) str;
pstr->len = len;
pstr->raw_len = len;
pstr->trans = trans;
pstr->icase = icase ? 1 : 0;
pstr->mbs_allocated = (trans != NULL || icase);
pstr->mb_cur_max = dfa->mb_cur_max;
pstr->is_utf8 = dfa->is_utf8;
pstr->map_notascii = dfa->map_notascii;
pstr->stop = pstr->len;
pstr->raw_stop = pstr->stop;
}
#ifdef RE_ENABLE_I18N
/* Build wide character buffer PSTR->WCS.
If the byte sequence of the string are:
<mb1>(0), <mb1>(1), <mb2>(0), <mb2>(1), <sb3>
Then wide character buffer will be:
<wc1> , WEOF , <wc2> , WEOF , <wc3>
We use WEOF for padding, they indicate that the position isn't
a first byte of a multibyte character.
Note that this function assumes PSTR->VALID_LEN elements are already
built and starts from PSTR->VALID_LEN. */
static void
internal_function
build_wcs_buffer (re_string_t *pstr)
{
#ifdef _LIBC
unsigned char buf[MB_LEN_MAX];
assert (MB_LEN_MAX >= pstr->mb_cur_max);
#else
unsigned char buf[64];
#endif
mbstate_t prev_st;
int byte_idx, end_idx, remain_len;
size_t mbclen;
/* Build the buffers from pstr->valid_len to either pstr->len or
pstr->bufs_len. */
end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len;
for (byte_idx = pstr->valid_len; byte_idx < end_idx;)
{
wchar_t wc;
const char *p;
remain_len = end_idx - byte_idx;
prev_st = pstr->cur_state;
/* Apply the translation if we need. */
if (BE (pstr->trans != NULL, 0))
{
int i, ch;
for (i = 0; i < pstr->mb_cur_max && i < remain_len; ++i)
{
ch = pstr->raw_mbs [pstr->raw_mbs_idx + byte_idx + i];
buf[i] = pstr->mbs[byte_idx + i] = pstr->trans[ch];
}
p = (const char *) buf;
}
else
p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + byte_idx;
mbclen = __mbrtowc (&wc, p, remain_len, &pstr->cur_state);
if (BE (mbclen == (size_t) -2, 0))
{
/* The buffer doesn't have enough space, finish to build. */
pstr->cur_state = prev_st;
break;
}
else if (BE (mbclen == (size_t) -1 || mbclen == 0, 0))
{
/* We treat these cases as a singlebyte character. */
mbclen = 1;
wc = (wchar_t) pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx];
if (BE (pstr->trans != NULL, 0))
wc = pstr->trans[wc];
pstr->cur_state = prev_st;
}
/* Write wide character and padding. */
pstr->wcs[byte_idx++] = wc;
/* Write paddings. */
for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;)
pstr->wcs[byte_idx++] = WEOF;
}
pstr->valid_len = byte_idx;
pstr->valid_raw_len = byte_idx;
}
/* Build wide character buffer PSTR->WCS like build_wcs_buffer,
but for REG_ICASE. */
static reg_errcode_t
internal_function
build_wcs_upper_buffer (re_string_t *pstr)
{
mbstate_t prev_st;
int src_idx, byte_idx, end_idx, remain_len;
size_t mbclen;
#ifdef _LIBC
char buf[MB_LEN_MAX];
assert (MB_LEN_MAX >= pstr->mb_cur_max);
#else
char buf[64];
#endif
byte_idx = pstr->valid_len;
end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len;
/* The following optimization assumes that ASCII characters can be
mapped to wide characters with a simple cast. */
if (! pstr->map_notascii && pstr->trans == NULL && !pstr->offsets_needed)
{
while (byte_idx < end_idx)
{
wchar_t wc;
if (isascii (pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx])
&& mbsinit (&pstr->cur_state))
{
/* In case of a singlebyte character. */
pstr->mbs[byte_idx]
= toupper (pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]);
/* The next step uses the assumption that wchar_t is encoded
ASCII-safe: all ASCII values can be converted like this. */
pstr->wcs[byte_idx] = (wchar_t) pstr->mbs[byte_idx];
++byte_idx;
continue;
}
remain_len = end_idx - byte_idx;
prev_st = pstr->cur_state;
mbclen = __mbrtowc (&wc,
((const char *) pstr->raw_mbs + pstr->raw_mbs_idx
+ byte_idx), remain_len, &pstr->cur_state);
if (BE (mbclen + 2 > 2, 1))
{
wchar_t wcu = wc;
if (iswlower (wc))
{
size_t mbcdlen;
wcu = towupper (wc);
mbcdlen = wcrtomb (buf, wcu, &prev_st);
if (BE (mbclen == mbcdlen, 1))
memcpy (pstr->mbs + byte_idx, buf, mbclen);
else
{
src_idx = byte_idx;
goto offsets_needed;
}
}
else
memcpy (pstr->mbs + byte_idx,
pstr->raw_mbs + pstr->raw_mbs_idx + byte_idx, mbclen);
pstr->wcs[byte_idx++] = wcu;
/* Write paddings. */
for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;)
pstr->wcs[byte_idx++] = WEOF;
}
else if (mbclen == (size_t) -1 || mbclen == 0)
{
/* It is an invalid character or '\0'. Just use the byte. */
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx];
pstr->mbs[byte_idx] = ch;
/* And also cast it to wide char. */
pstr->wcs[byte_idx++] = (wchar_t) ch;
if (BE (mbclen == (size_t) -1, 0))
pstr->cur_state = prev_st;
}
else
{
/* The buffer doesn't have enough space, finish to build. */
pstr->cur_state = prev_st;
break;
}
}
pstr->valid_len = byte_idx;
pstr->valid_raw_len = byte_idx;
return REG_NOERROR;
}
else
for (src_idx = pstr->valid_raw_len; byte_idx < end_idx;)
{
wchar_t wc;
const char *p;
offsets_needed:
remain_len = end_idx - byte_idx;
prev_st = pstr->cur_state;
if (BE (pstr->trans != NULL, 0))
{
int i, ch;
for (i = 0; i < pstr->mb_cur_max && i < remain_len; ++i)
{
ch = pstr->raw_mbs [pstr->raw_mbs_idx + src_idx + i];
buf[i] = pstr->trans[ch];
}
p = (const char *) buf;
}
else
p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + src_idx;
mbclen = __mbrtowc (&wc, p, remain_len, &pstr->cur_state);
if (BE (mbclen + 2 > 2, 1))
{
wchar_t wcu = wc;
if (iswlower (wc))
{
size_t mbcdlen;
wcu = towupper (wc);
mbcdlen = wcrtomb ((char *) buf, wcu, &prev_st);
if (BE (mbclen == mbcdlen, 1))
memcpy (pstr->mbs + byte_idx, buf, mbclen);
else if (mbcdlen != (size_t) -1)
{
size_t i;
if (byte_idx + mbcdlen > pstr->bufs_len)
{
pstr->cur_state = prev_st;
break;
}
if (pstr->offsets == NULL)
{
pstr->offsets = re_malloc (int, pstr->bufs_len);
if (pstr->offsets == NULL)
return REG_ESPACE;
}
if (!pstr->offsets_needed)
{
for (i = 0; i < (size_t) byte_idx; ++i)
pstr->offsets[i] = i;
pstr->offsets_needed = 1;
}
memcpy (pstr->mbs + byte_idx, buf, mbcdlen);
pstr->wcs[byte_idx] = wcu;
pstr->offsets[byte_idx] = src_idx;
for (i = 1; i < mbcdlen; ++i)
{
pstr->offsets[byte_idx + i]
= src_idx + (i < mbclen ? i : mbclen - 1);
pstr->wcs[byte_idx + i] = WEOF;
}
pstr->len += mbcdlen - mbclen;
if (pstr->raw_stop > src_idx)
pstr->stop += mbcdlen - mbclen;
end_idx = (pstr->bufs_len > pstr->len)
? pstr->len : pstr->bufs_len;
byte_idx += mbcdlen;
src_idx += mbclen;
continue;
}
else
memcpy (pstr->mbs + byte_idx, p, mbclen);
}
else
memcpy (pstr->mbs + byte_idx, p, mbclen);
if (BE (pstr->offsets_needed != 0, 0))
{
size_t i;
for (i = 0; i < mbclen; ++i)
pstr->offsets[byte_idx + i] = src_idx + i;
}
src_idx += mbclen;
pstr->wcs[byte_idx++] = wcu;
/* Write paddings. */
for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;)
pstr->wcs[byte_idx++] = WEOF;
}
else if (mbclen == (size_t) -1 || mbclen == 0)
{
/* It is an invalid character or '\0'. Just use the byte. */
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + src_idx];
if (BE (pstr->trans != NULL, 0))
ch = pstr->trans [ch];
pstr->mbs[byte_idx] = ch;
if (BE (pstr->offsets_needed != 0, 0))
pstr->offsets[byte_idx] = src_idx;
++src_idx;
/* And also cast it to wide char. */
pstr->wcs[byte_idx++] = (wchar_t) ch;
if (BE (mbclen == (size_t) -1, 0))
pstr->cur_state = prev_st;
}
else
{
/* The buffer doesn't have enough space, finish to build. */
pstr->cur_state = prev_st;
break;
}
}
pstr->valid_len = byte_idx;
pstr->valid_raw_len = src_idx;
return REG_NOERROR;
}
/* Skip characters until the index becomes greater than NEW_RAW_IDX.
Return the index. */
static int
internal_function
re_string_skip_chars (re_string_t *pstr, int new_raw_idx, wint_t *last_wc)
{
mbstate_t prev_st;
int rawbuf_idx;
size_t mbclen;
wint_t wc = WEOF;
/* Skip the characters which are not necessary to check. */
for (rawbuf_idx = pstr->raw_mbs_idx + pstr->valid_raw_len;
rawbuf_idx < new_raw_idx;)
{
wchar_t wc2;
int remain_len = pstr->len - rawbuf_idx;
prev_st = pstr->cur_state;
mbclen = __mbrtowc (&wc2, (const char *) pstr->raw_mbs + rawbuf_idx,
remain_len, &pstr->cur_state);
if (BE (mbclen == (size_t) -2 || mbclen == (size_t) -1 || mbclen == 0, 0))
{
/* We treat these cases as a single byte character. */
if (mbclen == 0 || remain_len == 0)
wc = L'\0';
else
wc = *(unsigned char *) (pstr->raw_mbs + rawbuf_idx);
mbclen = 1;
pstr->cur_state = prev_st;
}
else
wc = (wint_t) wc2;
/* Then proceed the next character. */
rawbuf_idx += mbclen;
}
*last_wc = (wint_t) wc;
return rawbuf_idx;
}
#endif /* RE_ENABLE_I18N */
/* Build the buffer PSTR->MBS, and apply the translation if we need.
This function is used in case of REG_ICASE. */
static void
internal_function
build_upper_buffer (re_string_t *pstr)
{
int char_idx, end_idx;
end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len;
for (char_idx = pstr->valid_len; char_idx < end_idx; ++char_idx)
{
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + char_idx];
if (BE (pstr->trans != NULL, 0))
ch = pstr->trans[ch];
if (islower (ch))
pstr->mbs[char_idx] = toupper (ch);
else
pstr->mbs[char_idx] = ch;
}
pstr->valid_len = char_idx;
pstr->valid_raw_len = char_idx;
}
/* Apply TRANS to the buffer in PSTR. */
static void
internal_function
re_string_translate_buffer (re_string_t *pstr)
{
int buf_idx, end_idx;
end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len;
for (buf_idx = pstr->valid_len; buf_idx < end_idx; ++buf_idx)
{
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + buf_idx];
pstr->mbs[buf_idx] = pstr->trans[ch];
}
pstr->valid_len = buf_idx;
pstr->valid_raw_len = buf_idx;
}
/* This function re-construct the buffers.
Concretely, convert to wide character in case of pstr->mb_cur_max > 1,
convert to upper case in case of REG_ICASE, apply translation. */
static reg_errcode_t
internal_function
re_string_reconstruct (re_string_t *pstr, int idx, int eflags)
{
int offset = idx - pstr->raw_mbs_idx;
if (BE (offset < 0, 0))
{
/* Reset buffer. */
#ifdef RE_ENABLE_I18N
if (pstr->mb_cur_max > 1)
memset (&pstr->cur_state, '\0', sizeof (mbstate_t));
#endif /* RE_ENABLE_I18N */
pstr->len = pstr->raw_len;
pstr->stop = pstr->raw_stop;
pstr->valid_len = 0;
pstr->raw_mbs_idx = 0;
pstr->valid_raw_len = 0;
pstr->offsets_needed = 0;
pstr->tip_context = ((eflags & REG_NOTBOL) ? CONTEXT_BEGBUF
: CONTEXT_NEWLINE | CONTEXT_BEGBUF);
if (!pstr->mbs_allocated)
pstr->mbs = (unsigned char *) pstr->raw_mbs;
offset = idx;
}
if (BE (offset != 0, 1))
{
/* Should the already checked characters be kept? */
if (BE (offset < pstr->valid_raw_len, 1))
{
/* Yes, move them to the front of the buffer. */
#ifdef RE_ENABLE_I18N
if (BE (pstr->offsets_needed, 0))
{
int low = 0, high = pstr->valid_len, mid;
do
{
mid = (high + low) / 2;
if (pstr->offsets[mid] > offset)
high = mid;
else if (pstr->offsets[mid] < offset)
low = mid + 1;
else
break;
}
while (low < high);
if (pstr->offsets[mid] < offset)
++mid;
pstr->tip_context = re_string_context_at (pstr, mid - 1,
eflags);
/* This can be quite complicated, so handle specially
only the common and easy case where the character with
different length representation of lower and upper
case is present at or after offset. */
if (pstr->valid_len > offset
&& mid == offset && pstr->offsets[mid] == offset)
{
memmove (pstr->wcs, pstr->wcs + offset,
(pstr->valid_len - offset) * sizeof (wint_t));
memmove (pstr->mbs, pstr->mbs + offset, pstr->valid_len - offset);
pstr->valid_len -= offset;
pstr->valid_raw_len -= offset;
for (low = 0; low < pstr->valid_len; low++)
pstr->offsets[low] = pstr->offsets[low + offset] - offset;
}
else
{
/* Otherwise, just find out how long the partial multibyte
character at offset is and fill it with WEOF/255. */
pstr->len = pstr->raw_len - idx + offset;
pstr->stop = pstr->raw_stop - idx + offset;
pstr->offsets_needed = 0;
while (mid > 0 && pstr->offsets[mid - 1] == offset)
--mid;
while (mid < pstr->valid_len)
if (pstr->wcs[mid] != WEOF)
break;
else
++mid;
if (mid == pstr->valid_len)
pstr->valid_len = 0;
else
{
pstr->valid_len = pstr->offsets[mid] - offset;
if (pstr->valid_len)
{
for (low = 0; low < pstr->valid_len; ++low)
pstr->wcs[low] = WEOF;
memset (pstr->mbs, 255, pstr->valid_len);
}
}
pstr->valid_raw_len = pstr->valid_len;
}
}
else
#endif
{
pstr->tip_context = re_string_context_at (pstr, offset - 1,
eflags);
#ifdef RE_ENABLE_I18N
if (pstr->mb_cur_max > 1)
memmove (pstr->wcs, pstr->wcs + offset,
(pstr->valid_len - offset) * sizeof (wint_t));
#endif /* RE_ENABLE_I18N */
if (BE (pstr->mbs_allocated, 0))
memmove (pstr->mbs, pstr->mbs + offset,
pstr->valid_len - offset);
pstr->valid_len -= offset;
pstr->valid_raw_len -= offset;
#if DEBUG
assert (pstr->valid_len > 0);
#endif
}
}
else
{
#ifdef RE_ENABLE_I18N
/* No, skip all characters until IDX. */
int prev_valid_len = pstr->valid_len;
if (BE (pstr->offsets_needed, 0))
{
pstr->len = pstr->raw_len - idx + offset;
pstr->stop = pstr->raw_stop - idx + offset;
pstr->offsets_needed = 0;
}
#endif
pstr->valid_len = 0;
#ifdef RE_ENABLE_I18N
if (pstr->mb_cur_max > 1)
{
int wcs_idx;
wint_t wc = WEOF;
if (pstr->is_utf8)
{
const unsigned char *raw, *p, *end;
/* Special case UTF-8. Multi-byte chars start with any
byte other than 0x80 - 0xbf. */
raw = pstr->raw_mbs + pstr->raw_mbs_idx;
end = raw + (offset - pstr->mb_cur_max);
if (end < pstr->raw_mbs)
end = pstr->raw_mbs;
p = raw + offset - 1;
#ifdef _LIBC
/* We know the wchar_t encoding is UCS4, so for the simple
case, ASCII characters, skip the conversion step. */
if (isascii (*p) && BE (pstr->trans == NULL, 1))
{
memset (&pstr->cur_state, '\0', sizeof (mbstate_t));
/* pstr->valid_len = 0; */
wc = (wchar_t) *p;
}
else
#endif
for (; p >= end; --p)
if ((*p & 0xc0) != 0x80)
{
mbstate_t cur_state;
wchar_t wc2;
int mlen = raw + pstr->len - p;
unsigned char buf[6];
size_t mbclen;
if (BE (pstr->trans != NULL, 0))
{
int i = mlen < 6 ? mlen : 6;
while (--i >= 0)
buf[i] = pstr->trans[p[i]];
}
/* XXX Don't use mbrtowc, we know which conversion
to use (UTF-8 -> UCS4). */
memset (&cur_state, 0, sizeof (cur_state));
mbclen = __mbrtowc (&wc2, (const char *) p, mlen,
&cur_state);
if (raw + offset - p <= mbclen
&& mbclen < (size_t) -2)
{
memset (&pstr->cur_state, '\0',
sizeof (mbstate_t));
pstr->valid_len = mbclen - (raw + offset - p);
wc = wc2;
}
break;
}
}
if (wc == WEOF)
pstr->valid_len = re_string_skip_chars (pstr, idx, &wc) - idx;
if (wc == WEOF)
pstr->tip_context
= re_string_context_at (pstr, prev_valid_len - 1, eflags);
else
pstr->tip_context = ((BE (pstr->word_ops_used != 0, 0)
&& IS_WIDE_WORD_CHAR (wc))
? CONTEXT_WORD
: ((IS_WIDE_NEWLINE (wc)
&& pstr->newline_anchor)
? CONTEXT_NEWLINE : 0));
if (BE (pstr->valid_len, 0))
{
for (wcs_idx = 0; wcs_idx < pstr->valid_len; ++wcs_idx)
pstr->wcs[wcs_idx] = WEOF;
if (pstr->mbs_allocated)
memset (pstr->mbs, 255, pstr->valid_len);
}
pstr->valid_raw_len = pstr->valid_len;
}
else
#endif /* RE_ENABLE_I18N */
{
int c = pstr->raw_mbs[pstr->raw_mbs_idx + offset - 1];
pstr->valid_raw_len = 0;
if (pstr->trans)
c = pstr->trans[c];
pstr->tip_context = (bitset_contain (pstr->word_char, c)
? CONTEXT_WORD
: ((IS_NEWLINE (c) && pstr->newline_anchor)
? CONTEXT_NEWLINE : 0));
}
}
if (!BE (pstr->mbs_allocated, 0))
pstr->mbs += offset;
}
pstr->raw_mbs_idx = idx;
pstr->len -= offset;
pstr->stop -= offset;
/* Then build the buffers. */
#ifdef RE_ENABLE_I18N
if (pstr->mb_cur_max > 1)
{
if (pstr->icase)
{
reg_errcode_t ret = build_wcs_upper_buffer (pstr);
if (BE (ret != REG_NOERROR, 0))
return ret;
}
else
build_wcs_buffer (pstr);
}
else
#endif /* RE_ENABLE_I18N */
if (BE (pstr->mbs_allocated, 0))
{
if (pstr->icase)
build_upper_buffer (pstr);
else if (pstr->trans != NULL)
re_string_translate_buffer (pstr);
}
else
pstr->valid_len = pstr->len;
pstr->cur_idx = 0;
return REG_NOERROR;
}
static unsigned char
internal_function __attribute ((pure))
re_string_peek_byte_case (const re_string_t *pstr, int idx)
{
int ch, off;
/* Handle the common (easiest) cases first. */
if (BE (!pstr->mbs_allocated, 1))
return re_string_peek_byte (pstr, idx);
#ifdef RE_ENABLE_I18N
if (pstr->mb_cur_max > 1
&& ! re_string_is_single_byte_char (pstr, pstr->cur_idx + idx))
return re_string_peek_byte (pstr, idx);
#endif
off = pstr->cur_idx + idx;
#ifdef RE_ENABLE_I18N
if (pstr->offsets_needed)
off = pstr->offsets[off];
#endif
ch = pstr->raw_mbs[pstr->raw_mbs_idx + off];
#ifdef RE_ENABLE_I18N
/* Ensure that e.g. for tr_TR.UTF-8 BACKSLASH DOTLESS SMALL LETTER I
this function returns CAPITAL LETTER I instead of first byte of
DOTLESS SMALL LETTER I. The latter would confuse the parser,
since peek_byte_case doesn't advance cur_idx in any way. */
if (pstr->offsets_needed && !isascii (ch))
return re_string_peek_byte (pstr, idx);
#endif
return ch;
}
static unsigned char
internal_function __attribute ((pure))
re_string_fetch_byte_case (re_string_t *pstr)
{
if (BE (!pstr->mbs_allocated, 1))
return re_string_fetch_byte (pstr);
#ifdef RE_ENABLE_I18N
if (pstr->offsets_needed)
{
int off, ch;
/* For tr_TR.UTF-8 [[:islower:]] there is
[[: CAPITAL LETTER I WITH DOT lower:]] in mbs. Skip
in that case the whole multi-byte character and return
the original letter. On the other side, with
[[: DOTLESS SMALL LETTER I return [[:I, as doing
anything else would complicate things too much. */
if (!re_string_first_byte (pstr, pstr->cur_idx))
return re_string_fetch_byte (pstr);
off = pstr->offsets[pstr->cur_idx];
ch = pstr->raw_mbs[pstr->raw_mbs_idx + off];
if (! isascii (ch))
return re_string_fetch_byte (pstr);
re_string_skip_bytes (pstr,
re_string_char_size_at (pstr, pstr->cur_idx));
return ch;
}
#endif
return pstr->raw_mbs[pstr->raw_mbs_idx + pstr->cur_idx++];
}
static void
internal_function
re_string_destruct (re_string_t *pstr)
{
#ifdef RE_ENABLE_I18N
re_free (pstr->wcs);
re_free (pstr->offsets);
#endif /* RE_ENABLE_I18N */
if (pstr->mbs_allocated)
re_free (pstr->mbs);
}
/* Return the context at IDX in INPUT. */
static unsigned int
internal_function
re_string_context_at (const re_string_t *input, int idx, int eflags)
{
int c;
if (BE (idx < 0, 0))
/* In this case, we use the value stored in input->tip_context,
since we can't know the character in input->mbs[-1] here. */
return input->tip_context;
if (BE (idx == input->len, 0))
return ((eflags & REG_NOTEOL) ? CONTEXT_ENDBUF
: CONTEXT_NEWLINE | CONTEXT_ENDBUF);
#ifdef RE_ENABLE_I18N
if (input->mb_cur_max > 1)
{
wint_t wc;
int wc_idx = idx;
while(input->wcs[wc_idx] == WEOF)
{
#ifdef DEBUG
/* It must not happen. */
assert (wc_idx >= 0);
#endif
--wc_idx;
if (wc_idx < 0)
return input->tip_context;
}
wc = input->wcs[wc_idx];
if (BE (input->word_ops_used != 0, 0) && IS_WIDE_WORD_CHAR (wc))
return CONTEXT_WORD;
return (IS_WIDE_NEWLINE (wc) && input->newline_anchor
? CONTEXT_NEWLINE : 0);
}
else
#endif
{
c = re_string_byte_at (input, idx);
if (bitset_contain (input->word_char, c))
return CONTEXT_WORD;
return IS_NEWLINE (c) && input->newline_anchor ? CONTEXT_NEWLINE : 0;
}
}
/* Functions for set operation. */
static reg_errcode_t
internal_function
re_node_set_alloc (re_node_set *set, int size)
{
/*
* ADR: valgrind says size can be 0, which then doesn't
* free the block of size 0. Harumph. This seems
* to work ok, though.
*/
if (size == 0)
{
memset(set, 0, sizeof(*set));
return REG_NOERROR;
}
set->alloc = size;
set->nelem = 0;
set->elems = re_malloc (int, size);
if (BE (set->elems == NULL, 0))
return REG_ESPACE;
return REG_NOERROR;
}
static reg_errcode_t
internal_function
re_node_set_init_1 (re_node_set *set, int elem)
{
set->alloc = 1;
set->nelem = 1;
set->elems = re_malloc (int, 1);
if (BE (set->elems == NULL, 0))
{
set->alloc = set->nelem = 0;
return REG_ESPACE;
}
set->elems[0] = elem;
return REG_NOERROR;
}
static reg_errcode_t
internal_function
re_node_set_init_2 (re_node_set *set, int elem1, int elem2)
{
set->alloc = 2;
set->elems = re_malloc (int, 2);
if (BE (set->elems == NULL, 0))
return REG_ESPACE;
if (elem1 == elem2)
{
set->nelem = 1;
set->elems[0] = elem1;
}
else
{
set->nelem = 2;
if (elem1 < elem2)
{
set->elems[0] = elem1;
set->elems[1] = elem2;
}
else
{
set->elems[0] = elem2;
set->elems[1] = elem1;
}
}
return REG_NOERROR;
}
static reg_errcode_t
internal_function
re_node_set_init_copy (re_node_set *dest, const re_node_set *src)
{
dest->nelem = src->nelem;
if (src->nelem > 0)
{
dest->alloc = dest->nelem;
dest->elems = re_malloc (int, dest->alloc);
if (BE (dest->elems == NULL, 0))
{
dest->alloc = dest->nelem = 0;
return REG_ESPACE;
}
memcpy (dest->elems, src->elems, src->nelem * sizeof (int));
}
else
re_node_set_init_empty (dest);
return REG_NOERROR;
}
/* Calculate the intersection of the sets SRC1 and SRC2. And merge it to
DEST. Return value indicate the error code or REG_NOERROR if succeeded.
Note: We assume dest->elems is NULL, when dest->alloc is 0. */
static reg_errcode_t
internal_function
re_node_set_add_intersect (re_node_set *dest, const re_node_set *src1,
const re_node_set *src2)
{
int i1, i2, is, id, delta, sbase;
if (src1->nelem == 0 || src2->nelem == 0)
return REG_NOERROR;
/* We need dest->nelem + 2 * elems_in_intersection; this is a
conservative estimate. */
if (src1->nelem + src2->nelem + dest->nelem > dest->alloc)
{
int new_alloc = src1->nelem + src2->nelem + dest->alloc;
int *new_elems = re_realloc (dest->elems, int, new_alloc);
if (BE (new_elems == NULL, 0))
return REG_ESPACE;
dest->elems = new_elems;
dest->alloc = new_alloc;
}
/* Find the items in the intersection of SRC1 and SRC2, and copy
into the top of DEST those that are not already in DEST itself. */
sbase = dest->nelem + src1->nelem + src2->nelem;
i1 = src1->nelem - 1;
i2 = src2->nelem - 1;
id = dest->nelem - 1;
for (;;)
{
if (src1->elems[i1] == src2->elems[i2])
{
/* Try to find the item in DEST. Maybe we could binary search? */
while (id >= 0 && dest->elems[id] > src1->elems[i1])
--id;
if (id < 0 || dest->elems[id] != src1->elems[i1])
dest->elems[--sbase] = src1->elems[i1];
if (--i1 < 0 || --i2 < 0)
break;
}
/* Lower the highest of the two items. */
else if (src1->elems[i1] < src2->elems[i2])
{
if (--i2 < 0)
break;
}
else
{
if (--i1 < 0)
break;
}
}
id = dest->nelem - 1;
is = dest->nelem + src1->nelem + src2->nelem - 1;
delta = is - sbase + 1;
/* Now copy. When DELTA becomes zero, the remaining
DEST elements are already in place; this is more or
less the same loop that is in re_node_set_merge. */
dest->nelem += delta;
if (delta > 0 && id >= 0)
for (;;)
{
if (dest->elems[is] > dest->elems[id])
{
/* Copy from the top. */
dest->elems[id + delta--] = dest->elems[is--];
if (delta == 0)
break;
}
else
{
/* Slide from the bottom. */
dest->elems[id + delta] = dest->elems[id];
if (--id < 0)
break;
}
}
/* Copy remaining SRC elements. */
memcpy (dest->elems, dest->elems + sbase, delta * sizeof (int));
return REG_NOERROR;
}
/* Calculate the union set of the sets SRC1 and SRC2. And store it to
DEST. Return value indicate the error code or REG_NOERROR if succeeded. */
static reg_errcode_t
internal_function
re_node_set_init_union (re_node_set *dest, const re_node_set *src1,
const re_node_set *src2)
{
int i1, i2, id;
if (src1 != NULL && src1->nelem > 0 && src2 != NULL && src2->nelem > 0)
{
dest->alloc = src1->nelem + src2->nelem;
dest->elems = re_malloc (int, dest->alloc);
if (BE (dest->elems == NULL, 0))
return REG_ESPACE;
}
else
{
if (src1 != NULL && src1->nelem > 0)
return re_node_set_init_copy (dest, src1);
else if (src2 != NULL && src2->nelem > 0)
return re_node_set_init_copy (dest, src2);
else
re_node_set_init_empty (dest);
return REG_NOERROR;
}
for (i1 = i2 = id = 0 ; i1 < src1->nelem && i2 < src2->nelem ;)
{
if (src1->elems[i1] > src2->elems[i2])
{
dest->elems[id++] = src2->elems[i2++];
continue;
}
if (src1->elems[i1] == src2->elems[i2])
++i2;
dest->elems[id++] = src1->elems[i1++];
}
if (i1 < src1->nelem)
{
memcpy (dest->elems + id, src1->elems + i1,
(src1->nelem - i1) * sizeof (int));
id += src1->nelem - i1;
}
else if (i2 < src2->nelem)
{
memcpy (dest->elems + id, src2->elems + i2,
(src2->nelem - i2) * sizeof (int));
id += src2->nelem - i2;
}
dest->nelem = id;
return REG_NOERROR;
}
/* Calculate the union set of the sets DEST and SRC. And store it to
DEST. Return value indicate the error code or REG_NOERROR if succeeded. */
static reg_errcode_t
internal_function
re_node_set_merge (re_node_set *dest, const re_node_set *src)
{
int is, id, sbase, delta;
if (src == NULL || src->nelem == 0)
return REG_NOERROR;
if (dest->alloc < 2 * src->nelem + dest->nelem)
{
int new_alloc = 2 * (src->nelem + dest->alloc);
int *new_buffer = re_realloc (dest->elems, int, new_alloc);
if (BE (new_buffer == NULL, 0))
return REG_ESPACE;
dest->elems = new_buffer;
dest->alloc = new_alloc;
}
if (BE (dest->nelem == 0, 0))
{
dest->nelem = src->nelem;
memcpy (dest->elems, src->elems, src->nelem * sizeof (int));
return REG_NOERROR;
}
/* Copy into the top of DEST the items of SRC that are not
found in DEST. Maybe we could binary search in DEST? */
for (sbase = dest->nelem + 2 * src->nelem,
is = src->nelem - 1, id = dest->nelem - 1; is >= 0 && id >= 0; )
{
if (dest->elems[id] == src->elems[is])
is--, id--;
else if (dest->elems[id] < src->elems[is])
dest->elems[--sbase] = src->elems[is--];
else /* if (dest->elems[id] > src->elems[is]) */
--id;
}
if (is >= 0)
{
/* If DEST is exhausted, the remaining items of SRC must be unique. */
sbase -= is + 1;
memcpy (dest->elems + sbase, src->elems, (is + 1) * sizeof (int));
}
id = dest->nelem - 1;
is = dest->nelem + 2 * src->nelem - 1;
delta = is - sbase + 1;
if (delta == 0)
return REG_NOERROR;
/* Now copy. When DELTA becomes zero, the remaining
DEST elements are already in place. */
dest->nelem += delta;
for (;;)
{
if (dest->elems[is] > dest->elems[id])
{
/* Copy from the top. */
dest->elems[id + delta--] = dest->elems[is--];
if (delta == 0)
break;
}
else
{
/* Slide from the bottom. */
dest->elems[id + delta] = dest->elems[id];
if (--id < 0)
{
/* Copy remaining SRC elements. */
memcpy (dest->elems, dest->elems + sbase,
delta * sizeof (int));
break;
}
}
}
return REG_NOERROR;
}
/* Insert the new element ELEM to the re_node_set* SET.
SET should not already have ELEM.
return -1 if an error is occured, return 1 otherwise. */
static int
internal_function
re_node_set_insert (re_node_set *set, int elem)
{
int idx;
/* In case the set is empty. */
if (set->alloc == 0)
{
if (BE (re_node_set_init_1 (set, elem) == REG_NOERROR, 1))
return 1;
else
return -1;
}
if (BE (set->nelem, 0) == 0)
{
/* We already guaranteed above that set->alloc != 0. */
set->elems[0] = elem;
++set->nelem;
return 1;
}
/* Realloc if we need. */
if (set->alloc == set->nelem)
{
int *new_elems;
set->alloc = set->alloc * 2;
new_elems = re_realloc (set->elems, int, set->alloc);
if (BE (new_elems == NULL, 0))
return -1;
set->elems = new_elems;
}
/* Move the elements which follows the new element. Test the
first element separately to skip a check in the inner loop. */
if (elem < set->elems[0])
{
idx = 0;
for (idx = set->nelem; idx > 0; idx--)
set->elems[idx] = set->elems[idx - 1];
}
else
{
for (idx = set->nelem; set->elems[idx - 1] > elem; idx--)
set->elems[idx] = set->elems[idx - 1];
}
/* Insert the new element. */
set->elems[idx] = elem;
++set->nelem;
return 1;
}
/* Insert the new element ELEM to the re_node_set* SET.
SET should not already have any element greater than or equal to ELEM.
Return -1 if an error is occured, return 1 otherwise. */
static int
internal_function
re_node_set_insert_last (re_node_set *set, int elem)
{
/* Realloc if we need. */
if (set->alloc == set->nelem)
{
int *new_elems;
set->alloc = (set->alloc + 1) * 2;
new_elems = re_realloc (set->elems, int, set->alloc);
if (BE (new_elems == NULL, 0))
return -1;
set->elems = new_elems;
}
/* Insert the new element. */
set->elems[set->nelem++] = elem;
return 1;
}
/* Compare two node sets SET1 and SET2.
return 1 if SET1 and SET2 are equivalent, return 0 otherwise. */
static int
internal_function __attribute ((pure))
re_node_set_compare (const re_node_set *set1, const re_node_set *set2)
{
int i;
if (set1 == NULL || set2 == NULL || set1->nelem != set2->nelem)
return 0;
for (i = set1->nelem ; --i >= 0 ; )
if (set1->elems[i] != set2->elems[i])
return 0;
return 1;
}
/* Return (idx + 1) if SET contains the element ELEM, return 0 otherwise. */
static int
internal_function __attribute ((pure))
re_node_set_contains (const re_node_set *set, int elem)
{
unsigned int idx, right, mid;
if (set->nelem <= 0)
return 0;
/* Binary search the element. */
idx = 0;
right = set->nelem - 1;
while (idx < right)
{
mid = (idx + right) / 2;
if (set->elems[mid] < elem)
idx = mid + 1;
else
right = mid;
}
return set->elems[idx] == elem ? idx + 1 : 0;
}
static void
internal_function
re_node_set_remove_at (re_node_set *set, int idx)
{
if (idx < 0 || idx >= set->nelem)
return;
--set->nelem;
for (; idx < set->nelem; idx++)
set->elems[idx] = set->elems[idx + 1];
}
/* Add the token TOKEN to dfa->nodes, and return the index of the token.
Or return -1, if an error will be occured. */
static int
internal_function
re_dfa_add_node (re_dfa_t *dfa, re_token_t token)
{
if (BE (dfa->nodes_len >= dfa->nodes_alloc, 0))
{
size_t new_nodes_alloc = dfa->nodes_alloc * 2;
int *new_nexts, *new_indices;
re_node_set *new_edests, *new_eclosures;
re_token_t *new_nodes;
/* Avoid overflows in realloc. */
const size_t max_object_size = MAX (sizeof (re_token_t),
MAX (sizeof (re_node_set),
sizeof (int)));
if (BE (SIZE_MAX / max_object_size < new_nodes_alloc, 0))
return -1;
new_nodes = re_realloc (dfa->nodes, re_token_t, new_nodes_alloc);
if (BE (new_nodes == NULL, 0))
return -1;
dfa->nodes = new_nodes;
new_nexts = re_realloc (dfa->nexts, int, new_nodes_alloc);
new_indices = re_realloc (dfa->org_indices, int, new_nodes_alloc);
new_edests = re_realloc (dfa->edests, re_node_set, new_nodes_alloc);
new_eclosures = re_realloc (dfa->eclosures, re_node_set, new_nodes_alloc);
if (BE (new_nexts == NULL || new_indices == NULL
|| new_edests == NULL || new_eclosures == NULL, 0))
return -1;
dfa->nexts = new_nexts;
dfa->org_indices = new_indices;
dfa->edests = new_edests;
dfa->eclosures = new_eclosures;
dfa->nodes_alloc = new_nodes_alloc;
}
dfa->nodes[dfa->nodes_len] = token;
dfa->nodes[dfa->nodes_len].constraint = 0;
#ifdef RE_ENABLE_I18N
dfa->nodes[dfa->nodes_len].accept_mb =
(token.type == OP_PERIOD && dfa->mb_cur_max > 1) || token.type == COMPLEX_BRACKET;
#endif
dfa->nexts[dfa->nodes_len] = -1;
re_node_set_init_empty (dfa->edests + dfa->nodes_len);
re_node_set_init_empty (dfa->eclosures + dfa->nodes_len);
return dfa->nodes_len++;
}
static inline unsigned int
internal_function
calc_state_hash (const re_node_set *nodes, unsigned int context)
{
unsigned int hash = nodes->nelem + context;
int i;
for (i = 0 ; i < nodes->nelem ; i++)
hash += nodes->elems[i];
return hash;
}
/* Search for the state whose node_set is equivalent to NODES.
Return the pointer to the state, if we found it in the DFA.
Otherwise create the new one and return it. In case of an error
return NULL and set the error code in ERR.
Note: - We assume NULL as the invalid state, then it is possible that
return value is NULL and ERR is REG_NOERROR.
- We never return non-NULL value in case of any errors, it is for
optimization. */
static re_dfastate_t *
internal_function
re_acquire_state (reg_errcode_t *err, const re_dfa_t *dfa,
const re_node_set *nodes)
{
unsigned int hash;
re_dfastate_t *new_state;
struct re_state_table_entry *spot;
int i;
if (BE (nodes->nelem == 0, 0))
{
*err = REG_NOERROR;
return NULL;
}
hash = calc_state_hash (nodes, 0);
spot = dfa->state_table + (hash & dfa->state_hash_mask);
for (i = 0 ; i < spot->num ; i++)
{
re_dfastate_t *state = spot->array[i];
if (hash != state->hash)
continue;
if (re_node_set_compare (&state->nodes, nodes))
return state;
}
/* There are no appropriate state in the dfa, create the new one. */
new_state = create_ci_newstate (dfa, nodes, hash);
if (BE (new_state == NULL, 0))
*err = REG_ESPACE;
return new_state;
}
/* Search for the state whose node_set is equivalent to NODES and
whose context is equivalent to CONTEXT.
Return the pointer to the state, if we found it in the DFA.
Otherwise create the new one and return it. In case of an error
return NULL and set the error code in ERR.
Note: - We assume NULL as the invalid state, then it is possible that
return value is NULL and ERR is REG_NOERROR.
- We never return non-NULL value in case of any errors, it is for
optimization. */
static re_dfastate_t *
internal_function
re_acquire_state_context (reg_errcode_t *err, const re_dfa_t *dfa,
const re_node_set *nodes, unsigned int context)
{
unsigned int hash;
re_dfastate_t *new_state;
struct re_state_table_entry *spot;
int i;
if (nodes->nelem == 0)
{
*err = REG_NOERROR;
return NULL;
}
hash = calc_state_hash (nodes, context);
spot = dfa->state_table + (hash & dfa->state_hash_mask);
for (i = 0 ; i < spot->num ; i++)
{
re_dfastate_t *state = spot->array[i];
if (state->hash == hash
&& state->context == context
&& re_node_set_compare (state->entrance_nodes, nodes))
return state;
}
/* There are no appropriate state in `dfa', create the new one. */
new_state = create_cd_newstate (dfa, nodes, context, hash);
if (BE (new_state == NULL, 0))
*err = REG_ESPACE;
return new_state;
}
/* Finish initialization of the new state NEWSTATE, and using its hash value
HASH put in the appropriate bucket of DFA's state table. Return value
indicates the error code if failed. */
static reg_errcode_t
register_state (const re_dfa_t *dfa, re_dfastate_t *newstate,
unsigned int hash)
{
struct re_state_table_entry *spot;
reg_errcode_t err;
int i;
newstate->hash = hash;
err = re_node_set_alloc (&newstate->non_eps_nodes, newstate->nodes.nelem);
if (BE (err != REG_NOERROR, 0))
return REG_ESPACE;
for (i = 0; i < newstate->nodes.nelem; i++)
{
int elem = newstate->nodes.elems[i];
if (!IS_EPSILON_NODE (dfa->nodes[elem].type))
if (re_node_set_insert_last (&newstate->non_eps_nodes, elem) < 0)
return REG_ESPACE;
}
spot = dfa->state_table + (hash & dfa->state_hash_mask);
if (BE (spot->alloc <= spot->num, 0))
{
int new_alloc = 2 * spot->num + 2;
re_dfastate_t **new_array = re_realloc (spot->array, re_dfastate_t *,
new_alloc);
if (BE (new_array == NULL, 0))
return REG_ESPACE;
spot->array = new_array;
spot->alloc = new_alloc;
}
spot->array[spot->num++] = newstate;
return REG_NOERROR;
}
static void
free_state (re_dfastate_t *state)
{
re_node_set_free (&state->non_eps_nodes);
re_node_set_free (&state->inveclosure);
if (state->entrance_nodes != &state->nodes)
{
re_node_set_free (state->entrance_nodes);
re_free (state->entrance_nodes);
}
re_node_set_free (&state->nodes);
re_free (state->word_trtable);
re_free (state->trtable);
re_free (state);
}
/* Create the new state which is independ of contexts.
Return the new state if succeeded, otherwise return NULL. */
static re_dfastate_t *
internal_function
create_ci_newstate (const re_dfa_t *dfa, const re_node_set *nodes,
unsigned int hash)
{
int i;
reg_errcode_t err;
re_dfastate_t *newstate;
newstate = (re_dfastate_t *) calloc (sizeof (re_dfastate_t), 1);
if (BE (newstate == NULL, 0))
return NULL;
err = re_node_set_init_copy (&newstate->nodes, nodes);
if (BE (err != REG_NOERROR, 0))
{
re_free (newstate);
return NULL;
}
newstate->entrance_nodes = &newstate->nodes;
for (i = 0 ; i < nodes->nelem ; i++)
{
re_token_t *node = dfa->nodes + nodes->elems[i];
re_token_type_t type = node->type;
if (type == CHARACTER && !node->constraint)
continue;
#ifdef RE_ENABLE_I18N
newstate->accept_mb |= node->accept_mb;
#endif /* RE_ENABLE_I18N */
/* If the state has the halt node, the state is a halt state. */
if (type == END_OF_RE)
newstate->halt = 1;
else if (type == OP_BACK_REF)
newstate->has_backref = 1;
else if (type == ANCHOR || node->constraint)
newstate->has_constraint = 1;
}
err = register_state (dfa, newstate, hash);
if (BE (err != REG_NOERROR, 0))
{
free_state (newstate);
newstate = NULL;
}
return newstate;
}
/* Create the new state which is depend on the context CONTEXT.
Return the new state if succeeded, otherwise return NULL. */
static re_dfastate_t *
internal_function
create_cd_newstate (const re_dfa_t *dfa, const re_node_set *nodes,
unsigned int context, unsigned int hash)
{
int i, nctx_nodes = 0;
reg_errcode_t err;
re_dfastate_t *newstate;
newstate = (re_dfastate_t *) calloc (sizeof (re_dfastate_t), 1);
if (BE (newstate == NULL, 0))
return NULL;
err = re_node_set_init_copy (&newstate->nodes, nodes);
if (BE (err != REG_NOERROR, 0))
{
re_free (newstate);
return NULL;
}
newstate->context = context;
newstate->entrance_nodes = &newstate->nodes;
for (i = 0 ; i < nodes->nelem ; i++)
{
re_token_t *node = dfa->nodes + nodes->elems[i];
re_token_type_t type = node->type;
unsigned int constraint = node->constraint;
if (type == CHARACTER && !constraint)
continue;
#ifdef RE_ENABLE_I18N
newstate->accept_mb |= node->accept_mb;
#endif /* RE_ENABLE_I18N */
/* If the state has the halt node, the state is a halt state. */
if (type == END_OF_RE)
newstate->halt = 1;
else if (type == OP_BACK_REF)
newstate->has_backref = 1;
if (constraint)
{
if (newstate->entrance_nodes == &newstate->nodes)
{
newstate->entrance_nodes = re_malloc (re_node_set, 1);
if (BE (newstate->entrance_nodes == NULL, 0))
{
free_state (newstate);
return NULL;
}
if (re_node_set_init_copy (newstate->entrance_nodes, nodes)
!= REG_NOERROR)
return NULL;
nctx_nodes = 0;
newstate->has_constraint = 1;
}
if (NOT_SATISFY_PREV_CONSTRAINT (constraint,context))
{
re_node_set_remove_at (&newstate->nodes, i - nctx_nodes);
++nctx_nodes;
}
}
}
err = register_state (dfa, newstate, hash);
if (BE (err != REG_NOERROR, 0))
{
free_state (newstate);
newstate = NULL;
}
return newstate;
}
deps/regex/regex_internal.h deleted 100644 → 0
/* Extended regular expression matching and search library.
Copyright (C) 2002-2005, 2007, 2008, 2010 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, write to the Free
Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
02111-1307 USA. */
#ifndef _REGEX_INTERNAL_H
#define _REGEX_INTERNAL_H 1
#include <assert.h>
#include <ctype.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#ifndef UNUSED
# ifdef __GNUC__
# define UNUSED __attribute__((unused))
# else
# define UNUSED
# endif
#endif
#if defined HAVE_LANGINFO_H || defined HAVE_LANGINFO_CODESET || defined _LIBC
# include <langinfo.h>
#endif
#if defined HAVE_LOCALE_H || defined _LIBC
# include <locale.h>
#endif
#if defined HAVE_WCHAR_H || defined _LIBC
# include <wchar.h>
#endif /* HAVE_WCHAR_H || _LIBC */
#if defined HAVE_WCTYPE_H || defined _LIBC
# include <wctype.h>
#endif /* HAVE_WCTYPE_H || _LIBC */
#if defined HAVE_STDBOOL_H || defined _LIBC
# include <stdbool.h>
#endif /* HAVE_STDBOOL_H || _LIBC */
#if !defined(ZOS_USS)
#if defined HAVE_STDINT_H || defined _LIBC
# include <stdint.h>
#endif /* HAVE_STDINT_H || _LIBC */
#endif /* !ZOS_USS */
#if defined _LIBC
# include <bits/libc-lock.h>
#else
# define __libc_lock_define(CLASS,NAME)
# define __libc_lock_init(NAME) do { } while (0)
# define __libc_lock_lock(NAME) do { } while (0)
# define __libc_lock_unlock(NAME) do { } while (0)
#endif
#ifndef GAWK
/* In case that the system doesn't have isblank(). */
#if !defined _LIBC && !defined HAVE_ISBLANK && !defined isblank
# define isblank(ch) ((ch) == ' ' || (ch) == '\t')
#endif
#else /* GAWK */
/*
* This is a mess. On glibc systems you have to define
* a magic constant to get isblank() out of <ctype.h>, since it's
* a C99 function. To heck with all that and borrow a page from
* dfa.c's book.
*/
static int
is_blank (int c)
{
return (c == ' ' || c == '\t');
}
#endif /* GAWK */
#ifdef _LIBC
# ifndef _RE_DEFINE_LOCALE_FUNCTIONS
# define _RE_DEFINE_LOCALE_FUNCTIONS 1
# include <locale/localeinfo.h>
# include <locale/elem-hash.h>
# include <locale/coll-lookup.h>
# endif
#endif
/* This is for other GNU distributions with internationalized messages. */
#if (HAVE_LIBINTL_H && ENABLE_NLS) || defined _LIBC
# include <libintl.h>
# ifdef _LIBC
# undef gettext
# define gettext(msgid) \
INTUSE(__dcgettext) (_libc_intl_domainname, msgid, LC_MESSAGES)
# endif
#else
# define gettext(msgid) (msgid)
#endif
#ifndef gettext_noop
/* This define is so xgettext can find the internationalizable
strings. */
# define gettext_noop(String) String
#endif
/* For loser systems without the definition. */
#ifndef SIZE_MAX
# define SIZE_MAX ((size_t) -1)
#endif
#ifndef NO_MBSUPPORT
#include "mbsupport.h" /* gawk */
#endif
#ifndef MB_CUR_MAX
#define MB_CUR_MAX 1
#endif
#if (defined MBS_SUPPORT) || _LIBC
# define RE_ENABLE_I18N
#endif
#if __GNUC__ >= 3
# define BE(expr, val) __builtin_expect (expr, val)
#else
# define BE(expr, val) (expr)
# ifdef inline
# undef inline
# endif
# define inline
#endif
/* Number of single byte character. */
#define SBC_MAX 256
#define COLL_ELEM_LEN_MAX 8
/* The character which represents newline. */
#define NEWLINE_CHAR '\n'
#define WIDE_NEWLINE_CHAR L'\n'
/* Rename to standard API for using out of glibc. */
#ifndef _LIBC
# ifdef __wctype
# undef __wctype
# endif
# define __wctype wctype
# ifdef __iswctype
# undef __iswctype
# endif
# define __iswctype iswctype
# define __btowc btowc
# define __mbrtowc mbrtowc
#undef __mempcpy /* GAWK */
# define __mempcpy mempcpy
# define __wcrtomb wcrtomb
# define __regfree regfree
# define attribute_hidden
#endif /* not _LIBC */
#ifdef __GNUC__
# define __attribute(arg) __attribute__ (arg)
#else
# define __attribute(arg)
#endif
extern const char __re_error_msgid[] attribute_hidden;
extern const size_t __re_error_msgid_idx[] attribute_hidden;
/* An integer used to represent a set of bits. It must be unsigned,
and must be at least as wide as unsigned int. */
typedef unsigned long int bitset_word_t;
/* All bits set in a bitset_word_t. */
#define BITSET_WORD_MAX ULONG_MAX
/* Number of bits in a bitset_word_t. Cast to int as most code use it
* like that for counting */
#define BITSET_WORD_BITS ((int)(sizeof (bitset_word_t) * CHAR_BIT))
/* Number of bitset_word_t in a bit_set. */
#define BITSET_WORDS (SBC_MAX / BITSET_WORD_BITS)
typedef bitset_word_t bitset_t[BITSET_WORDS];
typedef bitset_word_t *re_bitset_ptr_t;
typedef const bitset_word_t *re_const_bitset_ptr_t;
#define bitset_set(set,i) \
(set[i / BITSET_WORD_BITS] |= (bitset_word_t) 1 << i % BITSET_WORD_BITS)
#define bitset_clear(set,i) \
(set[i / BITSET_WORD_BITS] &= ~((bitset_word_t) 1 << i % BITSET_WORD_BITS))
#define bitset_contain(set,i) \
(set[i / BITSET_WORD_BITS] & ((bitset_word_t) 1 << i % BITSET_WORD_BITS))
#define bitset_empty(set) memset (set, '\0', sizeof (bitset_t))
#define bitset_set_all(set) memset (set, '\xff', sizeof (bitset_t))
#define bitset_copy(dest,src) memcpy (dest, src, sizeof (bitset_t))
#define PREV_WORD_CONSTRAINT 0x0001
#define PREV_NOTWORD_CONSTRAINT 0x0002
#define NEXT_WORD_CONSTRAINT 0x0004
#define NEXT_NOTWORD_CONSTRAINT 0x0008
#define PREV_NEWLINE_CONSTRAINT 0x0010
#define NEXT_NEWLINE_CONSTRAINT 0x0020
#define PREV_BEGBUF_CONSTRAINT 0x0040
#define NEXT_ENDBUF_CONSTRAINT 0x0080
#define WORD_DELIM_CONSTRAINT 0x0100
#define NOT_WORD_DELIM_CONSTRAINT 0x0200
typedef enum
{
INSIDE_WORD = PREV_WORD_CONSTRAINT | NEXT_WORD_CONSTRAINT,
WORD_FIRST = PREV_NOTWORD_CONSTRAINT | NEXT_WORD_CONSTRAINT,
WORD_LAST = PREV_WORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT,
INSIDE_NOTWORD = PREV_NOTWORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT,
LINE_FIRST = PREV_NEWLINE_CONSTRAINT,
LINE_LAST = NEXT_NEWLINE_CONSTRAINT,
BUF_FIRST = PREV_BEGBUF_CONSTRAINT,
BUF_LAST = NEXT_ENDBUF_CONSTRAINT,
WORD_DELIM = WORD_DELIM_CONSTRAINT,
NOT_WORD_DELIM = NOT_WORD_DELIM_CONSTRAINT
} re_context_type;
typedef struct
{
int alloc;
int nelem;
int *elems;
} re_node_set;
typedef enum
{
NON_TYPE = 0,
/* Node type, These are used by token, node, tree. */
CHARACTER = 1,
END_OF_RE = 2,
SIMPLE_BRACKET = 3,
OP_BACK_REF = 4,
OP_PERIOD = 5,
#ifdef RE_ENABLE_I18N
COMPLEX_BRACKET = 6,
OP_UTF8_PERIOD = 7,
#endif /* RE_ENABLE_I18N */
/* We define EPSILON_BIT as a macro so that OP_OPEN_SUBEXP is used
when the debugger shows values of this enum type. */
#define EPSILON_BIT 8
OP_OPEN_SUBEXP = EPSILON_BIT | 0,
OP_CLOSE_SUBEXP = EPSILON_BIT | 1,
OP_ALT = EPSILON_BIT | 2,
OP_DUP_ASTERISK = EPSILON_BIT | 3,
ANCHOR = EPSILON_BIT | 4,
/* Tree type, these are used only by tree. */
CONCAT = 16,
SUBEXP = 17,
/* Token type, these are used only by token. */
OP_DUP_PLUS = 18,
OP_DUP_QUESTION,
OP_OPEN_BRACKET,
OP_CLOSE_BRACKET,
OP_CHARSET_RANGE,
OP_OPEN_DUP_NUM,
OP_CLOSE_DUP_NUM,
OP_NON_MATCH_LIST,
OP_OPEN_COLL_ELEM,
OP_CLOSE_COLL_ELEM,
OP_OPEN_EQUIV_CLASS,
OP_CLOSE_EQUIV_CLASS,
OP_OPEN_CHAR_CLASS,
OP_CLOSE_CHAR_CLASS,
OP_WORD,
OP_NOTWORD,
OP_SPACE,
OP_NOTSPACE,
BACK_SLASH
} re_token_type_t;
#ifdef RE_ENABLE_I18N
typedef struct
{
/* Multibyte characters. */
wchar_t *mbchars;
/* Collating symbols. */
# ifdef _LIBC
int32_t *coll_syms;
# endif
/* Equivalence classes. */
# ifdef _LIBC
int32_t *equiv_classes;
# endif
/* Range expressions. */
# ifdef _LIBC
uint32_t *range_starts;
uint32_t *range_ends;
# else /* not _LIBC */
wchar_t *range_starts;
wchar_t *range_ends;
# endif /* not _LIBC */
/* Character classes. */
wctype_t *char_classes;
/* If this character set is the non-matching list. */
unsigned int non_match : 1;
/* # of multibyte characters. */
int nmbchars;
/* # of collating symbols. */
int ncoll_syms;
/* # of equivalence classes. */
int nequiv_classes;
/* # of range expressions. */
int nranges;
/* # of character classes. */
int nchar_classes;
} re_charset_t;
#endif /* RE_ENABLE_I18N */
typedef struct
{
union
{
unsigned char c; /* for CHARACTER */
re_bitset_ptr_t sbcset; /* for SIMPLE_BRACKET */
#ifdef RE_ENABLE_I18N
re_charset_t *mbcset; /* for COMPLEX_BRACKET */
#endif /* RE_ENABLE_I18N */
int idx; /* for BACK_REF */
re_context_type ctx_type; /* for ANCHOR */
} opr;
#if __GNUC__ >= 2
re_token_type_t type : 8;
#else
re_token_type_t type;
#endif
unsigned int constraint : 10; /* context constraint */
unsigned int duplicated : 1;
unsigned int opt_subexp : 1;
#ifdef RE_ENABLE_I18N
unsigned int accept_mb : 1;
/* These 2 bits can be moved into the union if needed (e.g. if running out
of bits; move opr.c to opr.c.c and move the flags to opr.c.flags). */
unsigned int mb_partial : 1;
#endif
unsigned int word_char : 1;
} re_token_t;
#define IS_EPSILON_NODE(type) ((type) & EPSILON_BIT)
struct re_string_t
{
/* Indicate the raw buffer which is the original string passed as an
argument of regexec(), re_search(), etc.. */
const unsigned char *raw_mbs;
/* Store the multibyte string. In case of "case insensitive mode" like
REG_ICASE, upper cases of the string are stored, otherwise MBS points
the same address that RAW_MBS points. */
unsigned char *mbs;
#ifdef RE_ENABLE_I18N
/* Store the wide character string which is corresponding to MBS. */
wint_t *wcs;
int *offsets;
mbstate_t cur_state;
#endif
/* Index in RAW_MBS. Each character mbs[i] corresponds to
raw_mbs[raw_mbs_idx + i]. */
int raw_mbs_idx;
/* The length of the valid characters in the buffers. */
int valid_len;
/* The corresponding number of bytes in raw_mbs array. */
int valid_raw_len;
/* The length of the buffers MBS and WCS. */
int bufs_len;
/* The index in MBS, which is updated by re_string_fetch_byte. */
int cur_idx;
/* length of RAW_MBS array. */
int raw_len;
/* This is RAW_LEN - RAW_MBS_IDX + VALID_LEN - VALID_RAW_LEN. */
int len;
/* End of the buffer may be shorter than its length in the cases such
as re_match_2, re_search_2. Then, we use STOP for end of the buffer
instead of LEN. */
int raw_stop;
/* This is RAW_STOP - RAW_MBS_IDX adjusted through OFFSETS. */
int stop;
/* The context of mbs[0]. We store the context independently, since
the context of mbs[0] may be different from raw_mbs[0], which is
the beginning of the input string. */
unsigned int tip_context;
/* The translation passed as a part of an argument of re_compile_pattern. */
RE_TRANSLATE_TYPE trans;
/* Copy of re_dfa_t's word_char. */
re_const_bitset_ptr_t word_char;
/* 1 if REG_ICASE. */
unsigned char icase;
unsigned char is_utf8;
unsigned char map_notascii;
unsigned char mbs_allocated;
unsigned char offsets_needed;
unsigned char newline_anchor;
unsigned char word_ops_used;
int mb_cur_max;
};
typedef struct re_string_t re_string_t;
struct re_dfa_t;
typedef struct re_dfa_t re_dfa_t;
#ifndef _LIBC
# ifdef __i386__
# define internal_function __attribute ((regparm (3), stdcall))
# else
# define internal_function
# endif
#endif
#ifndef NOT_IN_libc
static reg_errcode_t re_string_realloc_buffers (re_string_t *pstr,
int new_buf_len)
internal_function;
# ifdef RE_ENABLE_I18N
static void build_wcs_buffer (re_string_t *pstr) internal_function;
static reg_errcode_t build_wcs_upper_buffer (re_string_t *pstr)
internal_function;
# endif /* RE_ENABLE_I18N */
static void build_upper_buffer (re_string_t *pstr) internal_function;
static void re_string_translate_buffer (re_string_t *pstr) internal_function;
static unsigned int re_string_context_at (const re_string_t *input, int idx,
int eflags)
internal_function __attribute ((pure));
#endif
#define re_string_peek_byte(pstr, offset) \
((pstr)->mbs[(pstr)->cur_idx + offset])
#define re_string_fetch_byte(pstr) \
((pstr)->mbs[(pstr)->cur_idx++])
#define re_string_first_byte(pstr, idx) \
((idx) == (pstr)->valid_len || (pstr)->wcs[idx] != WEOF)
#define re_string_is_single_byte_char(pstr, idx) \
((pstr)->wcs[idx] != WEOF && ((pstr)->valid_len == (idx) + 1 \
|| (pstr)->wcs[(idx) + 1] != WEOF))
#define re_string_eoi(pstr) ((pstr)->stop <= (pstr)->cur_idx)
#define re_string_cur_idx(pstr) ((pstr)->cur_idx)
#define re_string_get_buffer(pstr) ((pstr)->mbs)
#define re_string_length(pstr) ((pstr)->len)
#define re_string_byte_at(pstr,idx) ((pstr)->mbs[idx])
#define re_string_skip_bytes(pstr,idx) ((pstr)->cur_idx += (idx))
#define re_string_set_index(pstr,idx) ((pstr)->cur_idx = (idx))
#ifndef _LIBC
# if HAVE_ALLOCA
# if (_MSC_VER)
# include <malloc.h>
# define __libc_use_alloca(n) 0
# else
# include <alloca.h>
/* The OS usually guarantees only one guard page at the bottom of the stack,
and a page size can be as small as 4096 bytes. So we cannot safely
allocate anything larger than 4096 bytes. Also care for the possibility
of a few compiler-allocated temporary stack slots. */
# define __libc_use_alloca(n) ((n) < 4032)
# endif
# else
/* alloca is implemented with malloc, so just use malloc. */
# define __libc_use_alloca(n) 0
# endif
#endif
#define re_malloc(t,n) ((t *) malloc ((n) * sizeof (t)))
/* SunOS 4.1.x realloc doesn't accept null pointers: pre-Standard C. Sigh. */
#define re_realloc(p,t,n) ((p != NULL) ? (t *) realloc (p,(n)*sizeof(t)) : (t *) calloc(n,sizeof(t)))
#define re_free(p) free (p)
struct bin_tree_t
{
struct bin_tree_t *parent;
struct bin_tree_t *left;
struct bin_tree_t *right;
struct bin_tree_t *first;
struct bin_tree_t *next;
re_token_t token;
/* `node_idx' is the index in dfa->nodes, if `type' == 0.
Otherwise `type' indicate the type of this node. */
int node_idx;
};
typedef struct bin_tree_t bin_tree_t;
#define BIN_TREE_STORAGE_SIZE \
((1024 - sizeof (void *)) / sizeof (bin_tree_t))
struct bin_tree_storage_t
{
struct bin_tree_storage_t *next;
bin_tree_t data[BIN_TREE_STORAGE_SIZE];
};
typedef struct bin_tree_storage_t bin_tree_storage_t;
#define CONTEXT_WORD 1
#define CONTEXT_NEWLINE (CONTEXT_WORD << 1)
#define CONTEXT_BEGBUF (CONTEXT_NEWLINE << 1)
#define CONTEXT_ENDBUF (CONTEXT_BEGBUF << 1)
#define IS_WORD_CONTEXT(c) ((c) & CONTEXT_WORD)
#define IS_NEWLINE_CONTEXT(c) ((c) & CONTEXT_NEWLINE)
#define IS_BEGBUF_CONTEXT(c) ((c) & CONTEXT_BEGBUF)
#define IS_ENDBUF_CONTEXT(c) ((c) & CONTEXT_ENDBUF)
#define IS_ORDINARY_CONTEXT(c) ((c) == 0)
#define IS_WORD_CHAR(ch) (isalnum (ch) || (ch) == '_')
#define IS_NEWLINE(ch) ((ch) == NEWLINE_CHAR)
#define IS_WIDE_WORD_CHAR(ch) (iswalnum (ch) || (ch) == L'_')
#define IS_WIDE_NEWLINE(ch) ((ch) == WIDE_NEWLINE_CHAR)
#define NOT_SATISFY_PREV_CONSTRAINT(constraint,context) \
((((constraint) & PREV_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \
|| ((constraint & PREV_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \
|| ((constraint & PREV_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT (context))\
|| ((constraint & PREV_BEGBUF_CONSTRAINT) && !IS_BEGBUF_CONTEXT (context)))
#define NOT_SATISFY_NEXT_CONSTRAINT(constraint,context) \
((((constraint) & NEXT_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \
|| (((constraint) & NEXT_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \
|| (((constraint) & NEXT_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT (context)) \
|| (((constraint) & NEXT_ENDBUF_CONSTRAINT) && !IS_ENDBUF_CONTEXT (context)))
struct re_dfastate_t
{
unsigned int hash;
re_node_set nodes;
re_node_set non_eps_nodes;
re_node_set inveclosure;
re_node_set *entrance_nodes;
struct re_dfastate_t **trtable, **word_trtable;
unsigned int context : 4;
unsigned int halt : 1;
/* If this state can accept `multi byte'.
Note that we refer to multibyte characters, and multi character
collating elements as `multi byte'. */
unsigned int accept_mb : 1;
/* If this state has backreference node(s). */
unsigned int has_backref : 1;
unsigned int has_constraint : 1;
};
typedef struct re_dfastate_t re_dfastate_t;
struct re_state_table_entry
{
int num;
int alloc;
re_dfastate_t **array;
};
/* Array type used in re_sub_match_last_t and re_sub_match_top_t. */
typedef struct
{
int next_idx;
int alloc;
re_dfastate_t **array;
} state_array_t;
/* Store information about the node NODE whose type is OP_CLOSE_SUBEXP. */
typedef struct
{
int node;
int str_idx; /* The position NODE match at. */
state_array_t path;
} re_sub_match_last_t;
/* Store information about the node NODE whose type is OP_OPEN_SUBEXP.
And information about the node, whose type is OP_CLOSE_SUBEXP,
corresponding to NODE is stored in LASTS. */
typedef struct
{
int str_idx;
int node;
state_array_t *path;
int alasts; /* Allocation size of LASTS. */
int nlasts; /* The number of LASTS. */
re_sub_match_last_t **lasts;
} re_sub_match_top_t;
struct re_backref_cache_entry
{
int node;
int str_idx;
int subexp_from;
int subexp_to;
char more;
char unused;
unsigned short int eps_reachable_subexps_map;
};
typedef struct
{
/* The string object corresponding to the input string. */
re_string_t input;
#if defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L)
const re_dfa_t *const dfa;
#else
const re_dfa_t *dfa;
#endif
/* EFLAGS of the argument of regexec. */
int eflags;
/* Where the matching ends. */
int match_last;
int last_node;
/* The state log used by the matcher. */
re_dfastate_t **state_log;
int state_log_top;
/* Back reference cache. */
int nbkref_ents;
int abkref_ents;
struct re_backref_cache_entry *bkref_ents;
int max_mb_elem_len;
int nsub_tops;
int asub_tops;
re_sub_match_top_t **sub_tops;
} re_match_context_t;
typedef struct
{
re_dfastate_t **sifted_states;
re_dfastate_t **limited_states;
int last_node;
int last_str_idx;
re_node_set limits;
} re_sift_context_t;
struct re_fail_stack_ent_t
{
int idx;
int node;
regmatch_t *regs;
re_node_set eps_via_nodes;
};
struct re_fail_stack_t
{
int num;
int alloc;
struct re_fail_stack_ent_t *stack;
};
struct re_dfa_t
{
re_token_t *nodes;
size_t nodes_alloc;
size_t nodes_len;
int *nexts;
int *org_indices;
re_node_set *edests;
re_node_set *eclosures;
re_node_set *inveclosures;
struct re_state_table_entry *state_table;
re_dfastate_t *init_state;
re_dfastate_t *init_state_word;
re_dfastate_t *init_state_nl;
re_dfastate_t *init_state_begbuf;
bin_tree_t *str_tree;
bin_tree_storage_t *str_tree_storage;
re_bitset_ptr_t sb_char;
int str_tree_storage_idx;
/* number of subexpressions `re_nsub' is in regex_t. */
unsigned int state_hash_mask;
int init_node;
int nbackref; /* The number of backreference in this dfa. */
/* Bitmap expressing which backreference is used. */
bitset_word_t used_bkref_map;
bitset_word_t completed_bkref_map;
unsigned int has_plural_match : 1;
/* If this dfa has "multibyte node", which is a backreference or
a node which can accept multibyte character or multi character
collating element. */
unsigned int has_mb_node : 1;
unsigned int is_utf8 : 1;
unsigned int map_notascii : 1;
unsigned int word_ops_used : 1;
int mb_cur_max;
bitset_t word_char;
reg_syntax_t syntax;
int *subexp_map;
#ifdef DEBUG
char* re_str;
#endif
#if defined _LIBC
__libc_lock_define (, lock)
#endif
};
#define re_node_set_init_empty(set) memset (set, '\0', sizeof (re_node_set))
#define re_node_set_remove(set,id) \
(re_node_set_remove_at (set, re_node_set_contains (set, id) - 1))
#define re_node_set_empty(p) ((p)->nelem = 0)
#define re_node_set_free(set) re_free ((set)->elems)
typedef enum
{
SB_CHAR,
MB_CHAR,
EQUIV_CLASS,
COLL_SYM,
CHAR_CLASS
} bracket_elem_type;
typedef struct
{
bracket_elem_type type;
union
{
unsigned char ch;
unsigned char *name;
wchar_t wch;
} opr;
} bracket_elem_t;
/* Inline functions for bitset operation. */
static inline void
bitset_not (bitset_t set)
{
int bitset_i;
for (bitset_i = 0; bitset_i < BITSET_WORDS; ++bitset_i)
set[bitset_i] = ~set[bitset_i];
}
static inline void
bitset_merge (bitset_t dest, const bitset_t src)
{
int bitset_i;
for (bitset_i = 0; bitset_i < BITSET_WORDS; ++bitset_i)
dest[bitset_i] |= src[bitset_i];
}
static inline void
bitset_mask (bitset_t dest, const bitset_t src)
{
int bitset_i;
for (bitset_i = 0; bitset_i < BITSET_WORDS; ++bitset_i)
dest[bitset_i] &= src[bitset_i];
}
#ifdef RE_ENABLE_I18N
/* Inline functions for re_string. */
static inline int
internal_function __attribute ((pure))
re_string_char_size_at (const re_string_t *pstr, int idx)
{
int byte_idx;
if (pstr->mb_cur_max == 1)
return 1;
for (byte_idx = 1; idx + byte_idx < pstr->valid_len; ++byte_idx)
if (pstr->wcs[idx + byte_idx] != WEOF)
break;
return byte_idx;
}
static inline wint_t
internal_function __attribute ((pure))
re_string_wchar_at (const re_string_t *pstr, int idx)
{
if (pstr->mb_cur_max == 1)
return (wint_t) pstr->mbs[idx];
return (wint_t) pstr->wcs[idx];
}
# ifndef NOT_IN_libc
static int
internal_function __attribute ((pure))
re_string_elem_size_at (const re_string_t *pstr, int idx)
{
# ifdef _LIBC
const unsigned char *p, *extra;
const int32_t *table, *indirect;
int32_t tmp;
# include <locale/weight.h>
uint_fast32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
if (nrules != 0)
{
table = (const int32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB);
extra = (const unsigned char *)
_NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB);
indirect = (const int32_t *) _NL_CURRENT (LC_COLLATE,
_NL_COLLATE_INDIRECTMB);
p = pstr->mbs + idx;
tmp = findidx (&p);
return p - pstr->mbs - idx;
}
else
# endif /* _LIBC */
return 1;
}
# endif
#endif /* RE_ENABLE_I18N */
#endif /* _REGEX_INTERNAL_H */
deps/regex/regexec.c deleted 100644 → 0
This source diff could not be displayed because it is too large. You can view the blob instead.
src/CMakeLists.txt
......@@ -306,6 +306,16 @@ ELSE()
MESSAGE(FATAL_ERROR "Asked for unknown SHA1 backend ${SHA1_BACKEND}")
ENDIF()
# Include PCRE and its POSIX regex compatibility layer when it is required
IF (HAVE_REGCOMP_L)
ADD_FEATURE_INFO(regex ON "using system regcomp_l")
ELSE()
ADD_SUBDIRECTORY("${libgit2_SOURCE_DIR}/deps/pcre" "${libgit2_BINARY_DIR}/deps/pcre")
LIST(APPEND LIBGIT2_INCLUDES "${libgit2_SOURCE_DIR}/deps/pcre")
LIST(APPEND LIBGIT2_OBJECTS $<TARGET_OBJECTS:pcre>)
ADD_FEATURE_INFO(regex ON "using bundled PCRE")
ENDIF()
# Optional external dependency: http-parser
FIND_PACKAGE(HTTP_Parser)
IF (USE_EXT_HTTP_PARSER AND HTTP_PARSER_FOUND AND HTTP_PARSER_VERSION_MAJOR EQUAL 2)
......
src/posix_regex.h
......@@ -8,14 +8,17 @@
#define INCLUDE_posix_regex_h__
#include "common.h"
#include <regex.h>
/*
* Regular expressions: if the operating system has p_regcomp_l,
* use that as our p_regcomp implementation, otherwise fall back
* to standard regcomp.
* use it so that we can override the locale environment variable.
* Otherwise, use our bundled PCRE implementation.
*/
#ifdef GIT_USE_REGCOMP_L
# include <regex.h>
# include <xlocale.h>
#define P_REG_EXTENDED REG_EXTENDED
#define P_REG_ICASE REG_ICASE
#define P_REG_NOMATCH REG_NOMATCH
......@@ -23,19 +26,28 @@
#define p_regex_t regex_t
#define p_regmatch_t regmatch_t
#define p_regerror regerror
#define p_regexec regexec
#define p_regfree regfree
#ifdef GIT_USE_REGCOMP_L
#include <xlocale.h>
GIT_INLINE(int) p_regcomp(p_regex_t *preg, const char *pattern, int cflags)
{
return regcomp_l(preg, pattern, cflags, (locale_t) 0);
}
#define p_regerror regerror
#define p_regexec regexec
#define p_regfree regfree
#else
# define p_regcomp regcomp
# include "pcreposix.h"
# define P_REG_EXTENDED PCRE_REG_EXTENDED
# define P_REG_ICASE PCRE_REG_ICASE
# define P_REG_NOMATCH PCRE_REG_NOMATCH
# define p_regex_t pcre_regex_t
# define p_regmatch_t pcre_regmatch_t
# define p_regcomp pcre_regcomp
# define p_regerror pcre_regerror
# define p_regexec pcre_regexec
# define p_regfree pcre_regfree
#endif
#endif
src/win32/precompiled.h
......@@ -13,8 +13,6 @@
#include <sys/types.h>
#include <sys/stat.h>
#include <regex.h>
#include <io.h>
#include <direct.h>
#ifdef GIT_THREADS
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment