procyberian/emacs
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Use Gnulib regex for lib-src

Browse source 
Emacs regular expressions forked from everyone else long ago.
This makes it official and should allow simplification later.
etags.c now uses the glibc regex API, falling back on a
Gnulib-supplied substitute lib/regex.c if necessary.
Emacs proper now uses its own regular expression module.
Although this patch may look dauntingly large, most of it
was generated automatically by admin/merge-gnulib
and contains an exact copy of the glibc regex source,
and the by-hand changes do not grow the Emacs source code.
* admin/merge-gnulib (GNULIB_MODULES): Add regex.
(AVOIDED_MODULES): Add btowc, langinfo, lock, mbrtowc, mbsinit,
nl_langinfo, wchar, wcrtomb, wctype-h.
* lib-src/Makefile.in (regex-emacs.o): Remove; Gnulib does it now.
(etags_deps, etags_libs): Remove regex-emacs.o.
* lib-src/etags.c: Go back to including regex.h.
(add_regex): Use unsigned char translation array,
since glibc regex requires that.
* lib/Makefile.in (not_emacs_OBJECTS, for_emacs_OBJECTS): New macros.
(libegnu_a_OBJECTS): Use them, to avoid building e-regex.o.
* lib/gnulib.mk.in, m4/gnulib-comp.m4: Regenerate.
* lib/regcomp.c, lib/regex.c, lib/regex.h, lib/regex_internal.c:
* lib/regex_internal.h, lib/regexec.c, m4/builtin-expect.m4:
* m4/eealloc.m4, m4/glibc21.m4, m4/mbstate_t.m4, m4/regex.m4:
New files, copied from Gnulib.
* src/regex-emacs.h, src/conf_post.h:
(RE_TRANSLATE_TYPE, RE_TRANSLATE, RE_TRANSLATE_P):
Move from src/conf_post.h to src/regex-emacs.h,
so that they don’t interfere with compiling lib/regex.c.
This commit is contained in:
Paul Eggert 2018-08-05 18:41:20 -07:00
parent e5652268a9
commit d904cc83f3
20 changed files with 12194 additions and 20 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

7
admin/merge-gnulib
View file

@ -37,7 +37,7 @@ GNULIB_MODULES='
getloadavg getopt-gnu gettime gettimeofday gitlog-to-changelog
ieee754-h ignore-value intprops largefile lstat
manywarnings memrchr minmax mkostemp mktime nstrftime
pipe2 pselect pthread_sigmask putenv qcopy-acl readlink readlinkat
pipe2 pselect pthread_sigmask putenv qcopy-acl readlink readlinkat regex
sig2str socklen stat-time std-gnu11 stdalign stddef stdio
stpcpy strtoimax symlink sys_stat sys_time
tempname time time_r time_rz timegm timer-time timespec-add timespec-sub
@ -46,11 +46,12 @@ GNULIB_MODULES='
'
AVOIDED_MODULES='
close dup fchdir fstat
malloc-posix msvc-inval msvc-nothrow
btowc close dup fchdir fstat langinfo lock
malloc-posix mbrtowc mbsinit msvc-inval msvc-nothrow nl_langinfo
openat-die opendir raise
save-cwd select setenv sigprocmask stat stdarg stdbool
threadlib tzset unsetenv utime utime-h
wchar wcrtomb wctype-h
'
GNULIB_TOOL_FLAGS='

7
etc/NEWS
View file

@ -31,6 +31,13 @@ functions 'json-serialize', 'json-insert', 'json-parse-string', and
'json-parse-buffer' are typically much faster than their Lisp
counterparts from json.el.
** The etags program now uses the C library's regular expression matcher
when possible, and a compatible regex substitute otherwise. This will
let developers maintain Emacs's own regex code without having to also
support other programs. The new configure option '--without-included-regex'
forces etags to use the C library's regex matcher even if the regex
substitute ordinarily would be used to work around compatibility problems.
** Emacs has been ported to the -fcheck-pointer-bounds option of GCC.
This causes Emacs to check bounds of some arrays addressed by its
internal pointers, which can be helpful when debugging the Emacs

8
lib-src/Makefile.in
View file

@ -361,13 +361,9 @@ TAGS: etags${EXEEXT} ${tagsfiles}
../lib/libgnu.a: $(config_h)
$(MAKE) -C ../lib all
regex-emacs.o: $(srcdir)/../src/regex-emacs.c $(srcdir)/../src/regex-emacs.h $(config_h)
$(AM_V_CC)$(CC) -c $(CPP_CFLAGS) $<
etags_deps = ${srcdir}/etags.c regex-emacs.o $(NTLIB) $(config_h)
etags_deps = ${srcdir}/etags.c $(NTLIB) $(config_h)
etags_cflags = -DEMACS_NAME="\"GNU Emacs\"" -DVERSION="\"${version}\"" -o $@
etags_libs = regex-emacs.o $(NTLIB) $(LOADLIBES)
etags_libs = $(NTLIB) $(LOADLIBES)
etags${EXEEXT}: ${etags_deps}
$(AM_V_CCLD)$(CC) ${ALL_CFLAGS} $(etags_cflags) $< $(etags_libs)

4
lib-src/etags.c
View file

@ -135,7 +135,7 @@ char pot_etags_version[] = "@(#) pot revision number is 17.38.1.4";
#endif
#include <getopt.h>
#include <regex-emacs.h>
#include <regex.h>
/* Define CTAGS to make the program "ctags" compatible with the usual one.
Leave it undefined to make the program "etags", which makes emacs-style
@ -6401,7 +6401,7 @@ add_regex (char *regexp_pattern, language *lang)
*patbuf = zeropattern;
if (ignore_case)
{
static char lc_trans[UCHAR_MAX + 1];
static unsigned char lc_trans[UCHAR_MAX + 1];
int i;
for (i = 0; i < UCHAR_MAX + 1; i++)
lc_trans[i] = c_tolower (i);

8
lib/Makefile.in
View file

@ -79,9 +79,15 @@ endif
Makefile: ../config.status $(srcdir)/Makefile.in
$(MAKE) -C .. src/$@
# Object modules that need not be built for Emacs.
# Emacs does not need e-regex.o (it has its own regex-emacs.c),
# and building it would just waste time.
not_emacs_OBJECTS = regex.o
libgnu_a_OBJECTS = $(gl_LIBOBJS) \
$(patsubst %.c,%.o,$(filter %.c,$(libgnu_a_SOURCES)))
libegnu_a_OBJECTS = $(patsubst %.o,e-%.o,$(libgnu_a_OBJECTS))
for_emacs_OBJECTS = $(filter-out $(not_emacs_OBJECTS),$(libgnu_a_OBJECTS))
libegnu_a_OBJECTS = $(patsubst %.o,e-%.o,$(for_emacs_OBJECTS))
$(libegnu_a_OBJECTS) $(libgnu_a_OBJECTS): $(BUILT_SOURCES)

23
lib/gnulib.mk.in
View file

@ -34,13 +34,19 @@
# --no-libtool \
# --macro-prefix=gl \
# --no-vc-files \
# --avoid=btowc \
# --avoid=close \
# --avoid=dup \
# --avoid=fchdir \
# --avoid=fstat \
# --avoid=langinfo \
# --avoid=lock \
# --avoid=malloc-posix \
# --avoid=mbrtowc \
# --avoid=mbsinit \
# --avoid=msvc-inval \
# --avoid=msvc-nothrow \
# --avoid=nl_langinfo \
# --avoid=openat-die \
# --avoid=opendir \
# --avoid=raise \
@ -56,6 +62,9 @@
# --avoid=unsetenv \
# --avoid=utime \
# --avoid=utime-h \
# --avoid=wchar \
# --avoid=wcrtomb \
# --avoid=wctype-h \
# alloca-opt \
# binary-io \
# byteswap \
@ -113,6 +122,7 @@
# qcopy-acl \
# readlink \
# readlinkat \
# regex \
# sig2str \
# socklen \
# stat-time \
@ -216,6 +226,7 @@ GETOPT_CDEFS_H = @GETOPT_CDEFS_H@
GETOPT_H = @GETOPT_H@
GFILENOTIFY_CFLAGS = @GFILENOTIFY_CFLAGS@
GFILENOTIFY_LIBS = @GFILENOTIFY_LIBS@
GLIBC21 = @GLIBC21@
GL_COND_LIBTOOL = @GL_COND_LIBTOOL@
GL_GENERATE_ALLOCA_H = @GL_GENERATE_ALLOCA_H@
GL_GENERATE_BYTESWAP_H = @GL_GENERATE_BYTESWAP_H@
@ -1024,6 +1035,7 @@ gameuser = @gameuser@
gl_GNULIB_ENABLED_03e0aaad4cb89ca757653bd367a6ccb7 = @gl_GNULIB_ENABLED_03e0aaad4cb89ca757653bd367a6ccb7@
gl_GNULIB_ENABLED_2049e887c7e5308faad27b3f894bb8c9 = @gl_GNULIB_ENABLED_2049e887c7e5308faad27b3f894bb8c9@
gl_GNULIB_ENABLED_260941c0e5dc67ec9e87d1fb321c300b = @gl_GNULIB_ENABLED_260941c0e5dc67ec9e87d1fb321c300b@
gl_GNULIB_ENABLED_37f71b604aa9c54446783d80f42fe547 = @gl_GNULIB_ENABLED_37f71b604aa9c54446783d80f42fe547@
gl_GNULIB_ENABLED_5264294aa0a5557541b53c8c741f7f31 = @gl_GNULIB_ENABLED_5264294aa0a5557541b53c8c741f7f31@
gl_GNULIB_ENABLED_6099e9737f757db36c47fa9d9f02e88c = @gl_GNULIB_ENABLED_6099e9737f757db36c47fa9d9f02e88c@
gl_GNULIB_ENABLED_682e609604ccaac6be382e4ee3a4eaec = @gl_GNULIB_ENABLED_682e609604ccaac6be382e4ee3a4eaec@
@ -2095,6 +2107,17 @@ EXTRA_libgnu_a_SOURCES += at-func.c readlinkat.c
endif
## end gnulib module readlinkat
## begin gnulib module regex
ifeq (,$(OMIT_GNULIB_MODULE_regex))
EXTRA_DIST += regcomp.c regex.c regex.h regex_internal.c regex_internal.h regexec.c
EXTRA_libgnu_a_SOURCES += regcomp.c regex.c regex_internal.c regexec.c
endif
## end gnulib module regex
## begin gnulib module root-uid
ifeq (,$(OMIT_GNULIB_MODULE_root-uid))

3944
lib/regcomp.c Normal file
View file

File diff suppressed because it is too large Load diff

81
lib/regex.c Normal file
View file

@ -0,0 +1,81 @@
/* Extended regular expression matching and search library.
Copyright (C) 2002-2018 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 General Public
License as published by the Free Software Foundation; either
version 3 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
General Public License for more details.
You should have received a copy of the GNU General Public
License along with the GNU C Library; if not, see
<https://www.gnu.org/licenses/>. */
#ifndef _LIBC
# include <config.h>
# if (__GNUC__ == 4 && 6 <= __GNUC_MINOR__) || 4 < __GNUC__
# pragma GCC diagnostic ignored "-Wsuggest-attribute=pure"
# endif
# if (__GNUC__ == 4 && 3 <= __GNUC_MINOR__) || 4 < __GNUC__
# pragma GCC diagnostic ignored "-Wold-style-definition"
# pragma GCC diagnostic ignored "-Wtype-limits"
# endif
#endif
/* Make sure no one compiles this code with a C++ compiler. */
#if defined __cplusplus && defined _LIBC
# 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
/* 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>
#include <regex.h>
#include "regex_internal.h"
#include "regex_internal.c"
#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

658
lib/regex.h Normal file
View file

@ -0,0 +1,658 @@
/* Definitions for data structures and routines for the regular
expression library.
Copyright (C) 1985, 1989-2018 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 General Public
License as published by the Free Software Foundation; either
version 3 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
General Public License for more details.
You should have received a copy of the GNU General Public
License along with the GNU C Library; if not, see
<https://www.gnu.org/licenses/>. */
#ifndef _REGEX_H
#define _REGEX_H 1
#include <sys/types.h>
/* Allow the use in C++ code. */
#ifdef __cplusplus
extern "C" {
#endif
/* Define __USE_GNU to declare GNU extensions that violate the
POSIX name space rules. */
#ifdef _GNU_SOURCE
# define __USE_GNU 1
#endif
#ifdef _REGEX_LARGE_OFFSETS
/* Use types and values that are wide enough to represent signed and
unsigned byte offsets in memory. This currently works only when
the regex code is used outside of the GNU C library; it is not yet
supported within glibc itself, and glibc users should not define
_REGEX_LARGE_OFFSETS. */
/* The type of object sizes. */
typedef size_t __re_size_t;
/* The type of object sizes, in places where the traditional code
uses unsigned long int. */
typedef size_t __re_long_size_t;
#else
/* The traditional GNU regex implementation mishandles strings longer
than INT_MAX. */
typedef unsigned int __re_size_t;
typedef unsigned long int __re_long_size_t;
#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, turn on internal regex debugging.
If not set, and debugging was on, turn it off.
This only works if regex.c is compiled -DDEBUG.
We define this bit always, so that all that's needed to turn on
debugging is to recompile regex.c; the calling code can always have
this bit set, and it won't affect anything in the normal case. */
# define RE_DEBUG (RE_NO_GNU_OPS << 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_DEBUG << 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 a regex or
immediately after an alternation, open-group or \} 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_CHAR_CLASSES \
| 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_SYNTAX_POSIX_BASIC | RE_NEWLINE_ALT) \
& ~(RE_CONTEXT_INVALID_DUP | RE_DOT_NOT_NULL))
# define RE_SYNTAX_EGREP \
((RE_SYNTAX_POSIX_EXTENDED | RE_INVALID_INTERVAL_ORD | RE_NEWLINE_ALT) \
& ~(RE_CONTEXT_INVALID_OPS | RE_DOT_NOT_NULL))
/* POSIX grep -E behavior is no longer incompatible with GNU. */
# define RE_SYNTAX_POSIX_EGREP \
RE_SYNTAX_EGREP
/* 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. POSIX-conforming
systems might define this in <limits.h>, but we want our
value, so remove any previous define. */
# ifdef _REGEX_INCLUDE_LIMITS_H
# include <limits.h>
# endif
# ifdef RE_DUP_MAX
# undef RE_DUP_MAX
# endif
/* RE_DUP_MAX is 2**15 - 1 because an earlier implementation stored
the counter as a 2-byte signed integer. This is no longer true, so
RE_DUP_MAX could be increased to (INT_MAX / 10 - 1), or to
((SIZE_MAX - 9) / 10) if _REGEX_LARGE_OFFSETS is defined.
However, there would be a huge performance problem if someone
actually used a pattern like a\{214748363\}, so RE_DUP_MAX retains
its historical value. */
# 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 (1 << 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 (1 << 2)
/* 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 (1 << 3)
/* 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_msgid' table in regcomp.c. */
typedef enum
{
_REG_ENOSYS = -1, /* This will never happen for this implementation. */
_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, /* Invalid 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, /* Too large (e.g., repeat count too large). */
_REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
} reg_errcode_t;
#if defined _XOPEN_SOURCE || defined __USE_XOPEN2K
# define REG_ENOSYS _REG_ENOSYS
#endif
#define REG_NOERROR _REG_NOERROR
#define REG_NOMATCH _REG_NOMATCH
#define REG_BADPAT _REG_BADPAT
#define REG_ECOLLATE _REG_ECOLLATE
#define REG_ECTYPE _REG_ECTYPE
#define REG_EESCAPE _REG_EESCAPE
#define REG_ESUBREG _REG_ESUBREG
#define REG_EBRACK _REG_EBRACK
#define REG_EPAREN _REG_EPAREN
#define REG_EBRACE _REG_EBRACE
#define REG_BADBR _REG_BADBR
#define REG_ERANGE _REG_ERANGE
#define REG_ESPACE _REG_ESPACE
#define REG_BADRPT _REG_BADRPT
#define REG_EEND _REG_EEND
#define REG_ESIZE _REG_ESIZE
#define REG_ERPAREN _REG_ERPAREN
/* This data structure represents a compiled pattern. Before calling
the pattern compiler, the fields 'buffer', 'allocated', 'fastmap',
and 'translate' can be set. After the pattern has been compiled,
the fields 're_nsub', 'not_bol' and 'not_eol' are 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. The type
'struct re_dfa_t' is private and is not declared here. */
struct re_dfa_t *__REPB_PREFIX(buffer);
/* Number of bytes to which 'buffer' points. */
__re_long_size_t __REPB_PREFIX(allocated);
/* Number of bytes actually used in 'buffer'. */
__re_long_size_t __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 're_compile_pattern' 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. */
#ifdef _REGEX_LARGE_OFFSETS
/* POSIX 1003.1-2008 requires that regoff_t be at least as wide as
ptrdiff_t and ssize_t. We don't know of any hosts where ptrdiff_t
is wider than ssize_t, so ssize_t is safe. ptrdiff_t is not
visible here, so use ssize_t. */
typedef ssize_t regoff_t;
#else
/* The traditional GNU regex implementation mishandles strings longer
than INT_MAX. */
typedef int regoff_t;
#endif
#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
{
__re_size_t 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.
To free the allocated storage, you must call 'regfree' on BUFFER.
Note that the translate table must either have been initialized by
'regcomp', with a malloc'ed value, or set to NULL before calling
'regfree'. */
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 regoff_t re_search (struct re_pattern_buffer *__buffer,
const char *__String, regoff_t __length,
regoff_t __start, regoff_t __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 regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
const char *__string1, regoff_t __length1,
const char *__string2, regoff_t __length2,
regoff_t __start, regoff_t __range,
struct re_registers *__regs,
regoff_t __stop);
/* Like 're_search', but return how many characters in STRING the regexp
in BUFFER matched, starting at position START. */
extern regoff_t re_match (struct re_pattern_buffer *__buffer,
const char *__String, regoff_t __length,
regoff_t __start, struct re_registers *__regs);
/* Relates to 're_match' as 're_search_2' relates to 're_search'. */
extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
const char *__string1, regoff_t __length1,
const char *__string2, regoff_t __length2,
regoff_t __start, struct re_registers *__regs,
regoff_t __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
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,
__re_size_t __num_regs,
regoff_t *__starts, regoff_t *__ends);
#endif /* Use GNU */
#if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_MISC)
# ifndef _CRAY
/* 4.2 bsd compatibility. */
extern char *re_comp (const char *);
extern int re_exec (const char *);
# endif
#endif
/* For plain 'restrict', use glibc's __restrict if defined.
Otherwise, GCC 2.95 and later have "__restrict"; C99 compilers have
"restrict", and "configure" may have defined "restrict".
Other compilers use __restrict, __restrict__, and _Restrict, and
'configure' might #define 'restrict' to those words, so pick a
different name. */
#ifndef _Restrict_
# if defined __restrict || 2 < __GNUC__ + (95 <= __GNUC_MINOR__)
# define _Restrict_ __restrict
# elif 199901L <= __STDC_VERSION__ || defined restrict
# define _Restrict_ restrict
# else
# define _Restrict_
# endif
#endif
/* For [restrict], use glibc's __restrict_arr if available.
Otherwise, GCC 3.1 (not in C++ mode) and C99 support [restrict]. */
#ifndef _Restrict_arr_
# ifdef __restrict_arr
# define _Restrict_arr_ __restrict_arr
# elif ((199901L <= __STDC_VERSION__ || 3 < __GNUC__ + (1 <= __GNUC_MINOR__)) \
&& !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_ __String, 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 */

1740
lib/regex_internal.c Normal file
View file

File diff suppressed because it is too large Load diff

911
lib/regex_internal.h Normal file
View file

@ -0,0 +1,911 @@
/* Extended regular expression matching and search library.
Copyright (C) 2002-2018 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 General Public
License as published by the Free Software Foundation; either
version 3 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
General Public License for more details.
You should have received a copy of the GNU General Public
License along with the GNU C Library; if not, see
<https://www.gnu.org/licenses/>. */
#ifndef _REGEX_INTERNAL_H
#define _REGEX_INTERNAL_H 1
#include <assert.h>
#include <ctype.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <langinfo.h>
#include <locale.h>
#include <wchar.h>
#include <wctype.h>
#include <stdbool.h>
#include <stdint.h>
/* Properties of integers. Although Gnulib has intprops.h, glibc does
without for now. */
#ifndef _LIBC
# include "intprops.h"
#else
/* True if the real type T is signed. */
# define TYPE_SIGNED(t) (! ((t) 0 < (t) -1))
/* True if adding the nonnegative Idx values A and B would overflow.
If false, set *R to A + B. A, B, and R may be evaluated more than
once, or zero times. Although this is not a full implementation of
Gnulib INT_ADD_WRAPV, it is good enough for glibc regex code.
FIXME: This implementation is a fragile stopgap, and this file would
be simpler and more robust if intprops.h were migrated into glibc. */
# define INT_ADD_WRAPV(a, b, r) \
(IDX_MAX - (a) < (b) ? true : (*(r) = (a) + (b), false))
#endif
#ifdef _LIBC
# include <libc-lock.h>
# define lock_define(name) __libc_lock_define (, name)
# define lock_init(lock) (__libc_lock_init (lock), 0)
# define lock_fini(lock) ((void) 0)
# define lock_lock(lock) __libc_lock_lock (lock)
# define lock_unlock(lock) __libc_lock_unlock (lock)
#elif defined GNULIB_LOCK && !defined USE_UNLOCKED_IO
# include "glthread/lock.h"
/* Use gl_lock_define if empty macro arguments are known to work.
Otherwise, fall back on less-portable substitutes. */
# if ((defined __GNUC__ && !defined __STRICT_ANSI__) \
|| (defined __STDC_VERSION__ && 199901L <= __STDC_VERSION__))
# define lock_define(name) gl_lock_define (, name)
# elif USE_POSIX_THREADS
# define lock_define(name) pthread_mutex_t name;
# elif USE_PTH_THREADS
# define lock_define(name) pth_mutex_t name;
# elif USE_SOLARIS_THREADS
# define lock_define(name) mutex_t name;
# elif USE_WINDOWS_THREADS
# define lock_define(name) gl_lock_t name;
# else
# define lock_define(name)
# endif
# define lock_init(lock) glthread_lock_init (&(lock))
# define lock_fini(lock) glthread_lock_destroy (&(lock))
# define lock_lock(lock) glthread_lock_lock (&(lock))
# define lock_unlock(lock) glthread_lock_unlock (&(lock))
#elif defined GNULIB_PTHREAD && !defined USE_UNLOCKED_IO
# include <pthread.h>
# define lock_define(name) pthread_mutex_t name;
# define lock_init(lock) pthread_mutex_init (&(lock), 0)
# define lock_fini(lock) pthread_mutex_destroy (&(lock))
# define lock_lock(lock) pthread_mutex_lock (&(lock))
# define lock_unlock(lock) pthread_mutex_unlock (&(lock))
#else
# define lock_define(name)
# define lock_init(lock) 0
# define lock_fini(lock) ((void) 0)
/* The 'dfa' avoids an "unused variable 'dfa'" warning from GCC. */
# define lock_lock(lock) ((void) dfa)
# define lock_unlock(lock) ((void) 0)
#endif
/* In case that the system doesn't have isblank(). */
#if !defined _LIBC && ! (defined isblank || (HAVE_ISBLANK && HAVE_DECL_ISBLANK))
# define isblank(ch) ((ch) == ' ' || (ch) == '\t')
#endif
#ifdef _LIBC
# ifndef _RE_DEFINE_LOCALE_FUNCTIONS
# define _RE_DEFINE_LOCALE_FUNCTIONS 1
# include <locale/localeinfo.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) \
__dcgettext (_libc_intl_domainname, msgid, LC_MESSAGES)
# endif
#else
# undef gettext
# define gettext(msgid) (msgid)
#endif
#ifndef gettext_noop
/* This define is so xgettext can find the internationalizable
strings. */
# define gettext_noop(String) String
#endif
#if (defined MB_CUR_MAX && HAVE_WCTYPE_H && HAVE_ISWCTYPE) || _LIBC
# define RE_ENABLE_I18N
#endif
#define BE(expr, val) __builtin_expect (expr, val)
/* Number of ASCII characters. */
#define ASCII_CHARS 0x80
/* Number of single byte characters. */
#define SBC_MAX (UCHAR_MAX + 1)
#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
# undef __wctype
# undef __iswctype
# define __wctype wctype
# define __iswalnum iswalnum
# define __iswctype iswctype
# define __towlower towlower
# define __towupper towupper
# define __btowc btowc
# define __mbrtowc mbrtowc
# define __wcrtomb wcrtomb
# define __regfree regfree
# define attribute_hidden
#endif /* not _LIBC */
#if __GNUC__ < 3 + (__GNUC_MINOR__ < 1)
# define __attribute__(arg)
#endif
#ifndef SSIZE_MAX
# define SSIZE_MAX ((ssize_t) (SIZE_MAX / 2))
#endif
/* The type of indexes into strings. This is signed, not size_t,
since the API requires indexes to fit in regoff_t anyway, and using
signed integers makes the code a bit smaller and presumably faster.
The traditional GNU regex implementation uses int for indexes.
The POSIX-compatible implementation uses a possibly-wider type.
The name 'Idx' is three letters to minimize the hassle of
reindenting a lot of regex code that formerly used 'int'. */
typedef regoff_t Idx;
#ifdef _REGEX_LARGE_OFFSETS
# define IDX_MAX SSIZE_MAX
#else
# define IDX_MAX INT_MAX
#endif
/* A hash value, suitable for computing hash tables. */
typedef __re_size_t re_hashval_t;
/* 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. For portability to hosts with
padding bits, do not use '(sizeof (bitset_word_t) * CHAR_BIT)';
instead, deduce it directly from BITSET_WORD_MAX. Avoid
greater-than-32-bit integers and unconditional shifts by more than
31 bits, as they're not portable. */
#if BITSET_WORD_MAX == 0xffffffffUL
# define BITSET_WORD_BITS 32
#elif BITSET_WORD_MAX >> 31 >> 4 == 1
# define BITSET_WORD_BITS 36
#elif BITSET_WORD_MAX >> 31 >> 16 == 1
# define BITSET_WORD_BITS 48
#elif BITSET_WORD_MAX >> 31 >> 28 == 1
# define BITSET_WORD_BITS 60
#elif BITSET_WORD_MAX >> 31 >> 31 >> 1 == 1
# define BITSET_WORD_BITS 64
#elif BITSET_WORD_MAX >> 31 >> 31 >> 9 == 1
# define BITSET_WORD_BITS 72
#elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 3 == 1
# define BITSET_WORD_BITS 128
#elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 7 == 1
# define BITSET_WORD_BITS 256
#elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 7 > 1
# define BITSET_WORD_BITS 257 /* any value > SBC_MAX will do here */
# if BITSET_WORD_BITS <= SBC_MAX
# error "Invalid SBC_MAX"
# endif
#else
# error "Add case for new bitset_word_t size"
#endif
/* Number of bitset_word_t values in a bitset_t. */
#define BITSET_WORDS ((SBC_MAX + BITSET_WORD_BITS - 1) / 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 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
{
Idx alloc;
Idx nelem;
Idx *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. */
Idx nmbchars;
/* # of collating symbols. */
Idx ncoll_syms;
/* # of equivalence classes. */
Idx nequiv_classes;
/* # of range expressions. */
Idx nranges;
/* # of character classes. */
Idx 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 */
Idx idx; /* for BACK_REF */
re_context_type ctx_type; /* for ANCHOR */
} opr;
#if __GNUC__ >= 2 && !defined __STRICT_ANSI__
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;
Idx *offsets;
mbstate_t cur_state;
#endif
/* Index in RAW_MBS. Each character mbs[i] corresponds to
raw_mbs[raw_mbs_idx + i]. */
Idx raw_mbs_idx;
/* The length of the valid characters in the buffers. */
Idx valid_len;
/* The corresponding number of bytes in raw_mbs array. */
Idx valid_raw_len;
/* The length of the buffers MBS and WCS. */
Idx bufs_len;
/* The index in MBS, which is updated by re_string_fetch_byte. */
Idx cur_idx;
/* length of RAW_MBS array. */
Idx raw_len;
/* This is RAW_LEN - RAW_MBS_IDX + VALID_LEN - VALID_RAW_LEN. */
Idx 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. */
Idx raw_stop;
/* This is RAW_STOP - RAW_MBS_IDX adjusted through OFFSETS. */
Idx 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;
/* true 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
# define IS_IN(libc) false
#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))
#if defined _LIBC || HAVE_ALLOCA
# include <alloca.h>
#endif
#ifndef _LIBC
# if HAVE_ALLOCA
/* 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)
# else
/* alloca is implemented with malloc, so just use malloc. */
# define __libc_use_alloca(n) 0
# undef alloca
# define alloca(n) malloc (n)
# endif
#endif
#ifdef _LIBC
# define MALLOC_0_IS_NONNULL 1
#elif !defined MALLOC_0_IS_NONNULL
# define MALLOC_0_IS_NONNULL 0
#endif
#ifndef MAX
# define MAX(a,b) ((a) < (b) ? (b) : (a))
#endif
#ifndef MIN
# define MIN(a,b) ((a) < (b) ? (a) : (b))
#endif
#define re_malloc(t,n) ((t *) malloc ((n) * sizeof (t)))
#define re_realloc(p,t,n) ((t *) realloc (p, (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. */
Idx 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
{
re_hashval_t 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
{
Idx num;
Idx alloc;
re_dfastate_t **array;
};
/* Array type used in re_sub_match_last_t and re_sub_match_top_t. */
typedef struct
{
Idx next_idx;
Idx alloc;
re_dfastate_t **array;
} state_array_t;
/* Store information about the node NODE whose type is OP_CLOSE_SUBEXP. */
typedef struct
{
Idx node;
Idx 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
{
Idx str_idx;
Idx node;
state_array_t *path;
Idx alasts; /* Allocation size of LASTS. */
Idx nlasts; /* The number of LASTS. */
re_sub_match_last_t **lasts;
} re_sub_match_top_t;
struct re_backref_cache_entry
{
Idx node;
Idx str_idx;
Idx subexp_from;
Idx 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. */
Idx match_last;
Idx last_node;
/* The state log used by the matcher. */
re_dfastate_t **state_log;
Idx state_log_top;
/* Back reference cache. */
Idx nbkref_ents;
Idx abkref_ents;
struct re_backref_cache_entry *bkref_ents;
int max_mb_elem_len;
Idx nsub_tops;
Idx 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;
Idx last_node;
Idx last_str_idx;
re_node_set limits;
} re_sift_context_t;
struct re_fail_stack_ent_t
{
Idx idx;
Idx node;
regmatch_t *regs;
re_node_set eps_via_nodes;
};
struct re_fail_stack_t
{
Idx num;
Idx alloc;
struct re_fail_stack_ent_t *stack;
};
struct re_dfa_t
{
re_token_t *nodes;
size_t nodes_alloc;
size_t nodes_len;
Idx *nexts;
Idx *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. */
re_hashval_t state_hash_mask;
Idx init_node;
Idx 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;
Idx *subexp_map;
#ifdef DEBUG
char* re_str;
#endif
lock_define (lock)
};
#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;
/* Functions for bitset_t operation. */
static inline void
bitset_set (bitset_t set, Idx i)
{
set[i / BITSET_WORD_BITS] |= (bitset_word_t) 1 << i % BITSET_WORD_BITS;
}
static inline void
bitset_clear (bitset_t set, Idx i)
{
set[i / BITSET_WORD_BITS] &= ~ ((bitset_word_t) 1 << i % BITSET_WORD_BITS);
}
static inline bool
bitset_contain (const bitset_t set, Idx i)
{
return (set[i / BITSET_WORD_BITS] >> i % BITSET_WORD_BITS) & 1;
}
static inline void
bitset_empty (bitset_t set)
{
memset (set, '\0', sizeof (bitset_t));
}
static inline void
bitset_set_all (bitset_t set)
{
memset (set, -1, sizeof (bitset_word_t) * (SBC_MAX / BITSET_WORD_BITS));
if (SBC_MAX % BITSET_WORD_BITS != 0)
set[BITSET_WORDS - 1] =
((bitset_word_t) 1 << SBC_MAX % BITSET_WORD_BITS) - 1;
}
static inline void
bitset_copy (bitset_t dest, const bitset_t src)
{
memcpy (dest, src, sizeof (bitset_t));
}
static inline void
bitset_not (bitset_t set)
{
int bitset_i;
for (bitset_i = 0; bitset_i < SBC_MAX / BITSET_WORD_BITS; ++bitset_i)
set[bitset_i] = ~set[bitset_i];
if (SBC_MAX % BITSET_WORD_BITS != 0)
set[BITSET_WORDS - 1] =
((((bitset_word_t) 1 << SBC_MAX % BITSET_WORD_BITS) - 1)
& ~set[BITSET_WORDS - 1]);
}
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
/* Functions for re_string. */
static int
__attribute__ ((pure, unused))
re_string_char_size_at (const re_string_t *pstr, Idx 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 wint_t
__attribute__ ((pure, unused))
re_string_wchar_at (const re_string_t *pstr, Idx idx)
{
if (pstr->mb_cur_max == 1)
return (wint_t) pstr->mbs[idx];
return (wint_t) pstr->wcs[idx];
}
# ifdef _LIBC
# include <locale/weight.h>
# endif
static int
__attribute__ ((pure, unused))
re_string_elem_size_at (const re_string_t *pstr, Idx idx)
{
# ifdef _LIBC
const unsigned char *p, *extra;
const int32_t *table, *indirect;
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;
findidx (table, indirect, extra, &p, pstr->len - idx);
return p - pstr->mbs - idx;
}
else
# endif /* _LIBC */
return 1;
}
#endif /* RE_ENABLE_I18N */
#ifndef __GNUC_PREREQ
# if defined __GNUC__ && defined __GNUC_MINOR__
# define __GNUC_PREREQ(maj, min) \
((__GNUC__ << 16) + __GNUC_MINOR__ >= ((maj) << 16) + (min))
# else
# define __GNUC_PREREQ(maj, min) 0
# endif
#endif
#if __GNUC_PREREQ (3,4)
# undef __attribute_warn_unused_result__
# define __attribute_warn_unused_result__ \
__attribute__ ((__warn_unused_result__))
#else
# define __attribute_warn_unused_result__ /* empty */
#endif
#ifndef FALLTHROUGH
# if __GNUC__ < 7
# define FALLTHROUGH ((void) 0)
# else
# define FALLTHROUGH __attribute__ ((__fallthrough__))
# endif
#endif
#endif /* _REGEX_INTERNAL_H */

4324
lib/regexec.c Normal file
View file

File diff suppressed because it is too large Load diff

49
m4/builtin-expect.m4 Normal file
View file

@ -0,0 +1,49 @@
dnl Check for __builtin_expect.
dnl Copyright 2016-2018 Free Software Foundation, Inc.
dnl This file is free software; the Free Software Foundation
dnl gives unlimited permission to copy and/or distribute it,
dnl with or without modifications, as long as this notice is preserved.
dnl Written by Paul Eggert.
AC_DEFUN([gl___BUILTIN_EXPECT],
[
AC_CACHE_CHECK([for __builtin_expect],
[gl_cv___builtin_expect],
[AC_LINK_IFELSE(
[AC_LANG_SOURCE([[
int
main (int argc, char **argv)
{
argc = __builtin_expect (argc, 100);
return argv[argc != 100][0];
}]])],
[gl_cv___builtin_expect=yes],
[AC_LINK_IFELSE(
[AC_LANG_SOURCE([[
#include <builtins.h>
int
main (int argc, char **argv)
{
argc = __builtin_expect (argc, 100);
return argv[argc != 100][0];
}]])],
[gl_cv___builtin_expect="in <builtins.h>"],
[gl_cv___builtin_expect=no])])])
if test "$gl_cv___builtin_expect" = yes; then
AC_DEFINE([HAVE___BUILTIN_EXPECT], [1])
elif test "$gl_cv___builtin_expect" = "in <builtins.h>"; then
AC_DEFINE([HAVE___BUILTIN_EXPECT], [2])
fi
AH_VERBATIM([HAVE___BUILTIN_EXPECT],
[/* Define to 1 if the compiler supports __builtin_expect,
and to 2 if <builtins.h> does. */
#undef HAVE___BUILTIN_EXPECT
#ifndef HAVE___BUILTIN_EXPECT
# define __builtin_expect(e, c) (e)
#elif HAVE___BUILTIN_EXPECT == 2
# include <builtins.h>
#endif
])
])

31
m4/eealloc.m4 Normal file
View file

@ -0,0 +1,31 @@
# eealloc.m4 serial 3
dnl Copyright (C) 2003, 2009-2018 Free Software Foundation, Inc.
dnl This file is free software; the Free Software Foundation
dnl gives unlimited permission to copy and/or distribute it,
dnl with or without modifications, as long as this notice is preserved.
AC_DEFUN([gl_EEALLOC],
[
AC_REQUIRE([gl_EEMALLOC])
AC_REQUIRE([gl_EEREALLOC])
])
AC_DEFUN([gl_EEMALLOC],
[
_AC_FUNC_MALLOC_IF(
[gl_cv_func_malloc_0_nonnull=1],
[gl_cv_func_malloc_0_nonnull=0])
AC_DEFINE_UNQUOTED([MALLOC_0_IS_NONNULL], [$gl_cv_func_malloc_0_nonnull],
[If malloc(0) is != NULL, define this to 1. Otherwise define this
to 0.])
])
AC_DEFUN([gl_EEREALLOC],
[
_AC_FUNC_REALLOC_IF(
[gl_cv_func_realloc_0_nonnull=1],
[gl_cv_func_realloc_0_nonnull=0])
AC_DEFINE_UNQUOTED([REALLOC_0_IS_NONNULL], [$gl_cv_func_realloc_0_nonnull],
[If realloc(NULL,0) is != NULL, define this to 1. Otherwise define this
to 0.])
])

34
m4/glibc21.m4 Normal file
View file

@ -0,0 +1,34 @@
# glibc21.m4 serial 5
dnl Copyright (C) 2000-2002, 2004, 2008, 2010-2018 Free Software Foundation,
dnl Inc.
dnl This file is free software; the Free Software Foundation
dnl gives unlimited permission to copy and/or distribute it,
dnl with or without modifications, as long as this notice is preserved.
# Test for the GNU C Library, version 2.1 or newer, or uClibc.
# From Bruno Haible.
AC_DEFUN([gl_GLIBC21],
[
AC_CACHE_CHECK([whether we are using the GNU C Library >= 2.1 or uClibc],
[ac_cv_gnu_library_2_1],
[AC_EGREP_CPP([Lucky],
[
#include <features.h>
#ifdef __GNU_LIBRARY__
#if (__GLIBC__ == 2 && __GLIBC_MINOR__ >= 1) || (__GLIBC__ > 2)
Lucky GNU user
#endif
#endif
#ifdef __UCLIBC__
Lucky user
#endif
],
[ac_cv_gnu_library_2_1=yes],
[ac_cv_gnu_library_2_1=no])
]
)
AC_SUBST([GLIBC21])
GLIBC21="$ac_cv_gnu_library_2_1"
]
)

30
m4/gnulib-comp.m4
View file

@ -48,6 +48,7 @@ AC_DEFUN([gl_EARLY],
# Code from module allocator:
# Code from module at-internal:
# Code from module binary-io:
# Code from module builtin-expect:
# Code from module byteswap:
# Code from module c-ctype:
# Code from module c-strcase:
@ -129,6 +130,7 @@ AC_DEFUN([gl_EARLY],
# Code from module qcopy-acl:
# Code from module readlink:
# Code from module readlinkat:
# Code from module regex:
# Code from module root-uid:
# Code from module sig2str:
# Code from module signal-h:
@ -358,6 +360,11 @@ AC_DEFUN([gl_INIT],
AC_LIBOBJ([readlinkat])
fi
gl_UNISTD_MODULE_INDICATOR([readlinkat])
gl_REGEX
if test $ac_use_included_regex = yes; then
AC_LIBOBJ([regex])
gl_PREREQ_REGEX
fi
gl_FUNC_SIG2STR
if test $ac_cv_func_sig2str = no; then
AC_LIBOBJ([sig2str])
@ -425,6 +432,7 @@ AC_DEFUN([gl_INIT],
gl_UTIMENS
AC_C_VARARRAYS
gl_gnulib_enabled_260941c0e5dc67ec9e87d1fb321c300b=false
gl_gnulib_enabled_37f71b604aa9c54446783d80f42fe547=false
gl_gnulib_enabled_cloexec=false
gl_gnulib_enabled_dirfd=false
gl_gnulib_enabled_dosname=false
@ -448,6 +456,13 @@ AC_DEFUN([gl_INIT],
func_gl_gnulib_m4code_open
fi
}
func_gl_gnulib_m4code_37f71b604aa9c54446783d80f42fe547 ()
{
if ! $gl_gnulib_enabled_37f71b604aa9c54446783d80f42fe547; then
gl___BUILTIN_EXPECT
gl_gnulib_enabled_37f71b604aa9c54446783d80f42fe547=true
fi
}
func_gl_gnulib_m4code_cloexec ()
{
if ! $gl_gnulib_enabled_cloexec; then
@ -651,6 +666,9 @@ AC_DEFUN([gl_INIT],
if test $HAVE_READLINKAT = 0; then
func_gl_gnulib_m4code_03e0aaad4cb89ca757653bd367a6ccb7
fi
if test $ac_use_included_regex = yes; then
func_gl_gnulib_m4code_37f71b604aa9c54446783d80f42fe547
fi
if { test $HAVE_DECL_STRTOIMAX = 0 || test $REPLACE_STRTOIMAX = 1; } && test $ac_cv_type_long_long_int = yes; then
func_gl_gnulib_m4code_strtoll
fi
@ -659,6 +677,7 @@ AC_DEFUN([gl_INIT],
fi
m4_pattern_allow([^gl_GNULIB_ENABLED_])
AM_CONDITIONAL([gl_GNULIB_ENABLED_260941c0e5dc67ec9e87d1fb321c300b], [$gl_gnulib_enabled_260941c0e5dc67ec9e87d1fb321c300b])
AM_CONDITIONAL([gl_GNULIB_ENABLED_37f71b604aa9c54446783d80f42fe547], [$gl_gnulib_enabled_37f71b604aa9c54446783d80f42fe547])
AM_CONDITIONAL([gl_GNULIB_ENABLED_cloexec], [$gl_gnulib_enabled_cloexec])
AM_CONDITIONAL([gl_GNULIB_ENABLED_dirfd], [$gl_gnulib_enabled_dirfd])
AM_CONDITIONAL([gl_GNULIB_ENABLED_dosname], [$gl_gnulib_enabled_dosname])
@ -924,6 +943,12 @@ AC_DEFUN([gl_FILE_LIST], [
lib/qcopy-acl.c
lib/readlink.c
lib/readlinkat.c
lib/regcomp.c
lib/regex.c
lib/regex.h
lib/regex_internal.c
lib/regex_internal.h
lib/regexec.c
lib/root-uid.h
lib/set-permissions.c
lib/sha1.c
@ -980,6 +1005,7 @@ AC_DEFUN([gl_FILE_LIST], [
m4/absolute-header.m4
m4/acl.m4
m4/alloca.m4
m4/builtin-expect.m4
m4/byteswap.m4
m4/c-strtod.m4
m4/clock_time.m4
@ -991,6 +1017,7 @@ AC_DEFUN([gl_FILE_LIST], [
m4/dirent_h.m4
m4/dirfd.m4
m4/dup2.m4
m4/eealloc.m4
m4/environ.m4
m4/errno_h.m4
m4/euidaccess.m4
@ -1018,6 +1045,7 @@ AC_DEFUN([gl_FILE_LIST], [
m4/gettime.m4
m4/gettimeofday.m4
m4/gl-openssl.m4
m4/glibc21.m4
m4/gnulib-common.m4
m4/group-member.m4
m4/ieee754-h.m4
@ -1030,6 +1058,7 @@ AC_DEFUN([gl_FILE_LIST], [
m4/lstat.m4
m4/manywarnings-c++.m4
m4/manywarnings.m4
m4/mbstate_t.m4
m4/md5.m4
m4/memrchr.m4
m4/minmax.m4
@ -1048,6 +1077,7 @@ AC_DEFUN([gl_FILE_LIST], [
m4/putenv.m4
m4/readlink.m4
m4/readlinkat.m4
m4/regex.m4
m4/sha1.m4
m4/sha256.m4
m4/sha512.m4

41
m4/mbstate_t.m4 Normal file
View file

@ -0,0 +1,41 @@
# mbstate_t.m4 serial 13
dnl Copyright (C) 2000-2002, 2008-2018 Free Software Foundation, Inc.
dnl This file is free software; the Free Software Foundation
dnl gives unlimited permission to copy and/or distribute it,
dnl with or without modifications, as long as this notice is preserved.
# From Paul Eggert.
# BeOS 5 has <wchar.h> but does not define mbstate_t,
# so you can't declare an object of that type.
# Check for this incompatibility with Standard C.
# AC_TYPE_MBSTATE_T
# -----------------
AC_DEFUN([AC_TYPE_MBSTATE_T],
[
AC_REQUIRE([AC_USE_SYSTEM_EXTENSIONS]) dnl for HP-UX 11.11
AC_CACHE_CHECK([for mbstate_t], [ac_cv_type_mbstate_t],
[AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
[AC_INCLUDES_DEFAULT[
/* Tru64 with Desktop Toolkit C has a bug: <stdio.h> must be included before
<wchar.h>.
BSD/OS 4.0.1 has a bug: <stddef.h>, <stdio.h> and <time.h> must be
included before <wchar.h>. */
#include <stddef.h>
#include <stdio.h>
#include <time.h>
#include <wchar.h>]],
[[mbstate_t x; return sizeof x;]])],
[ac_cv_type_mbstate_t=yes],
[ac_cv_type_mbstate_t=no])])
if test $ac_cv_type_mbstate_t = yes; then
AC_DEFINE([HAVE_MBSTATE_T], [1],
[Define to 1 if <wchar.h> declares mbstate_t.])
else
AC_DEFINE([mbstate_t], [int],
[Define to a type if <wchar.h> does not define.])
fi
])

300
m4/regex.m4 Normal file
View file

@ -0,0 +1,300 @@
# serial 67
# Copyright (C) 1996-2001, 2003-2018 Free Software Foundation, Inc.
#
# This file is free software; the Free Software Foundation
# gives unlimited permission to copy and/or distribute it,
# with or without modifications, as long as this notice is preserved.
dnl Initially derived from code in GNU grep.
dnl Mostly written by Jim Meyering.
AC_PREREQ([2.50])
AC_DEFUN([gl_REGEX],
[
AC_REQUIRE([AC_CANONICAL_HOST]) dnl for cross-compiles
AC_ARG_WITH([included-regex],
[AS_HELP_STRING([--without-included-regex],
[don't compile regex; this is the default on systems
with recent-enough versions of the GNU C Library
(use with caution on other systems).])])
case $with_included_regex in #(
yes|no) ac_use_included_regex=$with_included_regex
;;
'')
# If the system regex support is good enough that it passes the
# following run test, then default to *not* using the included regex.c.
# If cross compiling, assume the test would fail and use the included
# regex.c.
AC_CHECK_DECLS_ONCE([alarm])
AC_CHECK_HEADERS_ONCE([malloc.h])
AC_CACHE_CHECK([for working re_compile_pattern],
[gl_cv_func_re_compile_pattern_working],
[AC_RUN_IFELSE(
[AC_LANG_PROGRAM(
[[#include <regex.h>
#include <locale.h>
#include <limits.h>
#include <string.h>
#if defined M_CHECK_ACTION || HAVE_DECL_ALARM
# include <signal.h>
# include <unistd.h>
#endif
#if HAVE_MALLOC_H
# include <malloc.h>
#endif
#ifdef M_CHECK_ACTION
/* Exit with distinguishable exit code. */
static void sigabrt_no_core (int sig) { raise (SIGTERM); }
#endif
]],
[[int result = 0;
static struct re_pattern_buffer regex;
unsigned char folded_chars[UCHAR_MAX + 1];
int i;
const char *s;
struct re_registers regs;
/* Some builds of glibc go into an infinite loop on this
test. Use alarm to force death, and mallopt to avoid
malloc recursion in diagnosing the corrupted heap. */
#if HAVE_DECL_ALARM
signal (SIGALRM, SIG_DFL);
alarm (2);
#endif
#ifdef M_CHECK_ACTION
signal (SIGABRT, sigabrt_no_core);
mallopt (M_CHECK_ACTION, 2);
#endif
if (setlocale (LC_ALL, "en_US.UTF-8"))
{
{
/* https://sourceware.org/ml/libc-hacker/2006-09/msg00008.html
This test needs valgrind to catch the bug on Debian
GNU/Linux 3.1 x86, but it might catch the bug better
on other platforms and it shouldn't hurt to try the
test here. */
static char const pat[] = "insert into";
static char const data[] =
"\xFF\0\x12\xA2\xAA\xC4\xB1,K\x12\xC4\xB1*\xACK";
re_set_syntax (RE_SYNTAX_GREP | RE_HAT_LISTS_NOT_NEWLINE
| RE_ICASE);
memset (&regex, 0, sizeof regex);
s = re_compile_pattern (pat, sizeof pat - 1, &regex);
if (s)
result |= 1;
else if (re_search (&regex, data, sizeof data - 1,
0, sizeof data - 1, &regs)
!= -1)
result |= 1;
regfree (&regex);
}
{
/* This test is from glibc bug 15078.
The test case is from Andreas Schwab in
<https://sourceware.org/ml/libc-alpha/2013-01/msg00967.html>.
*/
static char const pat[] = "[^x]x";
static char const data[] =
/* <U1000><U103B><U103D><U1014><U103A><U102F><U1015><U103A> */
"\xe1\x80\x80"
"\xe1\x80\xbb"
"\xe1\x80\xbd"
"\xe1\x80\x94"
"\xe1\x80\xba"
"\xe1\x80\xaf"
"\xe1\x80\x95"
"\xe1\x80\xba"
"x";
re_set_syntax (0);
memset (&regex, 0, sizeof regex);
s = re_compile_pattern (pat, sizeof pat - 1, &regex);
if (s)
result |= 1;
else
{
i = re_search (&regex, data, sizeof data - 1,
0, sizeof data - 1, 0);
if (i != 0 && i != 21)
result |= 1;
}
regfree (&regex);
}
if (! setlocale (LC_ALL, "C"))
return 1;
}
/* This test is from glibc bug 3957, reported by Andrew Mackey. */
re_set_syntax (RE_SYNTAX_EGREP | RE_HAT_LISTS_NOT_NEWLINE);
memset (&regex, 0, sizeof regex);
s = re_compile_pattern ("a[^x]b", 6, &regex);
if (s)
result |= 2;
/* This should fail, but succeeds for glibc-2.5. */
else if (re_search (&regex, "a\nb", 3, 0, 3, &regs) != -1)
result |= 2;
/* This regular expression is from Spencer ere test number 75
in grep-2.3. */
re_set_syntax (RE_SYNTAX_POSIX_EGREP);
memset (&regex, 0, sizeof regex);
for (i = 0; i <= UCHAR_MAX; i++)
folded_chars[i] = i;
regex.translate = folded_chars;
s = re_compile_pattern ("a[[:@:>@:]]b\n", 11, &regex);
/* This should fail with _Invalid character class name_ error. */
if (!s)
result |= 4;
/* Ensure that [b-a] is diagnosed as invalid, when
using RE_NO_EMPTY_RANGES. */
re_set_syntax (RE_SYNTAX_POSIX_EGREP | RE_NO_EMPTY_RANGES);
memset (&regex, 0, sizeof regex);
s = re_compile_pattern ("a[b-a]", 6, &regex);
if (s == 0)
result |= 8;
/* This should succeed, but does not for glibc-2.1.3. */
memset (&regex, 0, sizeof regex);
s = re_compile_pattern ("{1", 2, &regex);
if (s)
result |= 8;
/* The following example is derived from a problem report
against gawk from Jorge Stolfi <stolfi@ic.unicamp.br>. */
memset (&regex, 0, sizeof regex);
s = re_compile_pattern ("[an\371]*n", 7, &regex);
if (s)
result |= 8;
/* This should match, but does not for glibc-2.2.1. */
else if (re_match (&regex, "an", 2, 0, &regs) != 2)
result |= 8;
memset (&regex, 0, sizeof regex);
s = re_compile_pattern ("x", 1, &regex);
if (s)
result |= 8;
/* glibc-2.2.93 does not work with a negative RANGE argument. */
else if (re_search (&regex, "wxy", 3, 2, -2, &regs) != 1)
result |= 8;
/* The version of regex.c in older versions of gnulib
ignored RE_ICASE. Detect that problem too. */
re_set_syntax (RE_SYNTAX_EMACS | RE_ICASE);
memset (&regex, 0, sizeof regex);
s = re_compile_pattern ("x", 1, &regex);
if (s)
result |= 16;
else if (re_search (&regex, "WXY", 3, 0, 3, &regs) < 0)
result |= 16;
/* Catch a bug reported by Vin Shelton in
https://lists.gnu.org/r/bug-coreutils/2007-06/msg00089.html
*/
re_set_syntax (RE_SYNTAX_POSIX_BASIC
& ~RE_CONTEXT_INVALID_DUP
& ~RE_NO_EMPTY_RANGES);
memset (&regex, 0, sizeof regex);
s = re_compile_pattern ("[[:alnum:]_-]\\\\+$", 16, &regex);
if (s)
result |= 32;
/* REG_STARTEND was added to glibc on 2004-01-15.
Reject older versions. */
if (! REG_STARTEND)
result |= 64;
#if 0
/* It would be nice to reject hosts whose regoff_t values are too
narrow (including glibc on hosts with 64-bit ptrdiff_t and
32-bit int), but we should wait until glibc implements this
feature. Otherwise, support for equivalence classes and
multibyte collation symbols would always be broken except
when compiling --without-included-regex. */
if (sizeof (regoff_t) < sizeof (ptrdiff_t)
|| sizeof (regoff_t) < sizeof (ssize_t))
result |= 64;
#endif
return result;
]])],
[gl_cv_func_re_compile_pattern_working=yes],
[gl_cv_func_re_compile_pattern_working=no],
[case "$host_os" in
# Guess no on native Windows.
mingw*) gl_cv_func_re_compile_pattern_working="guessing no" ;;
# Otherwise, assume it is not working.
*) gl_cv_func_re_compile_pattern_working="guessing no" ;;
esac
])
])
case "$gl_cv_func_re_compile_pattern_working" in #(
*yes) ac_use_included_regex=no;; #(
*no) ac_use_included_regex=yes;;
esac
;;
*) AC_MSG_ERROR([Invalid value for --with-included-regex: $with_included_regex])
;;
esac
if test $ac_use_included_regex = yes; then
AC_DEFINE([_REGEX_INCLUDE_LIMITS_H], [1],
[Define if you want <regex.h> to include <limits.h>, so that it
consistently overrides <limits.h>'s RE_DUP_MAX.])
AC_DEFINE([_REGEX_LARGE_OFFSETS], [1],
[Define if you want regoff_t to be at least as wide POSIX requires.])
AC_DEFINE([re_syntax_options], [rpl_re_syntax_options],
[Define to rpl_re_syntax_options if the replacement should be used.])
AC_DEFINE([re_set_syntax], [rpl_re_set_syntax],
[Define to rpl_re_set_syntax if the replacement should be used.])
AC_DEFINE([re_compile_pattern], [rpl_re_compile_pattern],
[Define to rpl_re_compile_pattern if the replacement should be used.])
AC_DEFINE([re_compile_fastmap], [rpl_re_compile_fastmap],
[Define to rpl_re_compile_fastmap if the replacement should be used.])
AC_DEFINE([re_search], [rpl_re_search],
[Define to rpl_re_search if the replacement should be used.])
AC_DEFINE([re_search_2], [rpl_re_search_2],
[Define to rpl_re_search_2 if the replacement should be used.])
AC_DEFINE([re_match], [rpl_re_match],
[Define to rpl_re_match if the replacement should be used.])
AC_DEFINE([re_match_2], [rpl_re_match_2],
[Define to rpl_re_match_2 if the replacement should be used.])
AC_DEFINE([re_set_registers], [rpl_re_set_registers],
[Define to rpl_re_set_registers if the replacement should be used.])
AC_DEFINE([re_comp], [rpl_re_comp],
[Define to rpl_re_comp if the replacement should be used.])
AC_DEFINE([re_exec], [rpl_re_exec],
[Define to rpl_re_exec if the replacement should be used.])
AC_DEFINE([regcomp], [rpl_regcomp],
[Define to rpl_regcomp if the replacement should be used.])
AC_DEFINE([regexec], [rpl_regexec],
[Define to rpl_regexec if the replacement should be used.])
AC_DEFINE([regerror], [rpl_regerror],
[Define to rpl_regerror if the replacement should be used.])
AC_DEFINE([regfree], [rpl_regfree],
[Define to rpl_regfree if the replacement should be used.])
fi
])
# Prerequisites of lib/regex.c and lib/regex_internal.c.
AC_DEFUN([gl_PREREQ_REGEX],
[
AC_REQUIRE([AC_USE_SYSTEM_EXTENSIONS])
AC_REQUIRE([AC_C_INLINE])
AC_REQUIRE([AC_C_RESTRICT])
AC_REQUIRE([AC_TYPE_MBSTATE_T])
AC_REQUIRE([gl_EEMALLOC])
AC_REQUIRE([gl_GLIBC21])
AC_CHECK_HEADERS([libintl.h])
AC_CHECK_FUNCS_ONCE([isblank iswctype])
AC_CHECK_DECLS([isblank], [], [], [[#include <ctype.h>]])
])

7
src/conf_post.h
View file

@ -202,13 +202,6 @@ extern void _DebPrint (const char *fmt, ...);
#endif
#endif
#ifdef emacs /* Don't do this for lib-src. */
/* Tell regex-emacs.c to use a type compatible with Emacs. */
#define RE_TRANSLATE_TYPE Lisp_Object
#define RE_TRANSLATE(TBL, C) char_table_translate (TBL, C)
#define RE_TRANSLATE_P(TBL) (!EQ (TBL, make_number (0)))
#endif
/* Tell time_rz.c to use Emacs's getter and setter for TZ.
Only Emacs uses time_rz so this is OK. */
#define getenv_TZ emacs_getenv_TZ

7
src/regex-emacs.h
View file

@ -219,7 +219,7 @@ extern ptrdiff_t emacs_re_safe_alloca;
((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
& ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
#define RE_SYNTAX_POSIX_AWK \
#define RE_SYNTAX_POSIX_AWK \
(RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
| RE_INTERVALS | RE_NO_GNU_OPS)
@ -350,6 +350,11 @@ typedef enum
REG_ESIZEBR /* n or m too big in \{n,m\} */
} reg_errcode_t;
/* Use a type compatible with Emacs. */
#define RE_TRANSLATE_TYPE Lisp_Object
#define RE_TRANSLATE(TBL, C) char_table_translate (TBL, C)
#define RE_TRANSLATE_P(TBL) (!EQ (TBL, make_number (0)))
/* 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