diff options
Diffstat (limited to 'lib')
43 files changed, 23553 insertions, 0 deletions
diff --git a/lib/Makefile.am b/lib/Makefile.am new file mode 100644 index 0000000..dd560a8 --- /dev/null +++ b/lib/Makefile.am @@ -0,0 +1,1045 @@ +## DO NOT EDIT! GENERATED AUTOMATICALLY! +## Process this file with automake to produce Makefile.in. +# Copyright (C) 2002-2012 Free Software Foundation, Inc. +# +# This file 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. +# +# This file 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 this file. If not, see <http://www.gnu.org/licenses/>. +# +# As a special exception to the GNU General Public License, +# this file may be distributed as part of a program that +# contains a configuration script generated by Autoconf, under +# the same distribution terms as the rest of that program. +# +# Generated by gnulib-tool. +# Reproduce by: gnulib-tool --import --dir=. --lib=libgnu --source-base=lib --m4-base=m4 --doc-base=doc --tests-base=tests --aux-dir=. --no-conditional-dependencies --no-libtool --macro-prefix=gl --no-vc-files flock regex + +AUTOMAKE_OPTIONS = 1.5 gnits + +SUBDIRS = +noinst_HEADERS = +noinst_LIBRARIES = +noinst_LTLIBRARIES = +EXTRA_DIST = +BUILT_SOURCES = +SUFFIXES = +MOSTLYCLEANFILES = core *.stackdump +MOSTLYCLEANDIRS = +CLEANFILES = +DISTCLEANFILES = +MAINTAINERCLEANFILES = + +AM_CPPFLAGS = +AM_CFLAGS = + +noinst_LIBRARIES += libgnu.a + +libgnu_a_SOURCES = +libgnu_a_LIBADD = $(gl_LIBOBJS) +libgnu_a_DEPENDENCIES = $(gl_LIBOBJS) +EXTRA_libgnu_a_SOURCES = + +## begin gnulib module alloca-opt + +BUILT_SOURCES += $(ALLOCA_H) + +# We need the following in order to create <alloca.h> when the system +# doesn't have one that works with the given compiler. +if GL_GENERATE_ALLOCA_H +alloca.h: alloca.in.h $(top_builddir)/config.status + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + cat $(srcdir)/alloca.in.h; \ + } > $@-t && \ + mv -f $@-t $@ +else +alloca.h: $(top_builddir)/config.status + rm -f $@ +endif +MOSTLYCLEANFILES += alloca.h alloca.h-t + +EXTRA_DIST += alloca.in.h + +## end gnulib module alloca-opt + +## begin gnulib module btowc + + +EXTRA_DIST += btowc.c + +EXTRA_libgnu_a_SOURCES += btowc.c + +## end gnulib module btowc + +## begin gnulib module configmake + +# Listed in the same order as the GNU makefile conventions, and +# provided by autoconf 2.59c+. +# The Automake-defined pkg* macros are appended, in the order +# listed in the Automake 1.10a+ documentation. +configmake.h: Makefile + $(AM_V_GEN)rm -f $@-t && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + echo '#define PREFIX "$(prefix)"'; \ + echo '#define EXEC_PREFIX "$(exec_prefix)"'; \ + echo '#define BINDIR "$(bindir)"'; \ + echo '#define SBINDIR "$(sbindir)"'; \ + echo '#define LIBEXECDIR "$(libexecdir)"'; \ + echo '#define DATAROOTDIR "$(datarootdir)"'; \ + echo '#define DATADIR "$(datadir)"'; \ + echo '#define SYSCONFDIR "$(sysconfdir)"'; \ + echo '#define SHAREDSTATEDIR "$(sharedstatedir)"'; \ + echo '#define LOCALSTATEDIR "$(localstatedir)"'; \ + echo '#define INCLUDEDIR "$(includedir)"'; \ + echo '#define OLDINCLUDEDIR "$(oldincludedir)"'; \ + echo '#define DOCDIR "$(docdir)"'; \ + echo '#define INFODIR "$(infodir)"'; \ + echo '#define HTMLDIR "$(htmldir)"'; \ + echo '#define DVIDIR "$(dvidir)"'; \ + echo '#define PDFDIR "$(pdfdir)"'; \ + echo '#define PSDIR "$(psdir)"'; \ + echo '#define LIBDIR "$(libdir)"'; \ + echo '#define LISPDIR "$(lispdir)"'; \ + echo '#define LOCALEDIR "$(localedir)"'; \ + echo '#define MANDIR "$(mandir)"'; \ + echo '#define MANEXT "$(manext)"'; \ + echo '#define PKGDATADIR "$(pkgdatadir)"'; \ + echo '#define PKGINCLUDEDIR "$(pkgincludedir)"'; \ + echo '#define PKGLIBDIR "$(pkglibdir)"'; \ + echo '#define PKGLIBEXECDIR "$(pkglibexecdir)"'; \ + } | sed '/""/d' > $@-t + mv -f $@-t $@ + +BUILT_SOURCES += configmake.h +CLEANFILES += configmake.h configmake.h-t + +## end gnulib module configmake + +## begin gnulib module flock + + +EXTRA_DIST += flock.c + +EXTRA_libgnu_a_SOURCES += flock.c + +## end gnulib module flock + +## begin gnulib module gettext-h + +libgnu_a_SOURCES += gettext.h + +## end gnulib module gettext-h + +## begin gnulib module langinfo + +BUILT_SOURCES += langinfo.h + +# We need the following in order to create an empty placeholder for +# <langinfo.h> when the system doesn't have one. +langinfo.h: langinfo.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''HAVE_LANGINFO_H''@|$(HAVE_LANGINFO_H)|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_LANGINFO_H''@|$(NEXT_LANGINFO_H)|g' \ + -e 's/@''GNULIB_NL_LANGINFO''@/$(GNULIB_NL_LANGINFO)/g' \ + -e 's|@''HAVE_LANGINFO_CODESET''@|$(HAVE_LANGINFO_CODESET)|g' \ + -e 's|@''HAVE_LANGINFO_T_FMT_AMPM''@|$(HAVE_LANGINFO_T_FMT_AMPM)|g' \ + -e 's|@''HAVE_LANGINFO_ERA''@|$(HAVE_LANGINFO_ERA)|g' \ + -e 's|@''HAVE_LANGINFO_YESEXPR''@|$(HAVE_LANGINFO_YESEXPR)|g' \ + -e 's|@''HAVE_NL_LANGINFO''@|$(HAVE_NL_LANGINFO)|g' \ + -e 's|@''REPLACE_NL_LANGINFO''@|$(REPLACE_NL_LANGINFO)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)' \ + < $(srcdir)/langinfo.in.h; \ + } > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += langinfo.h langinfo.h-t + +EXTRA_DIST += langinfo.in.h + +## end gnulib module langinfo + +## begin gnulib module localcharset + +libgnu_a_SOURCES += localcharset.h localcharset.c + +# We need the following in order to install a simple file in $(libdir) +# which is shared with other installed packages. We use a list of referencing +# packages so that "make uninstall" will remove the file if and only if it +# is not used by another installed package. +# On systems with glibc-2.1 or newer, the file is redundant, therefore we +# avoid installing it. + +all-local: charset.alias ref-add.sed ref-del.sed + +charset_alias = $(DESTDIR)$(libdir)/charset.alias +charset_tmp = $(DESTDIR)$(libdir)/charset.tmp +install-exec-local: install-exec-localcharset +install-exec-localcharset: all-local + if test $(GLIBC21) = no; then \ + case '$(host_os)' in \ + darwin[56]*) \ + need_charset_alias=true ;; \ + darwin* | cygwin* | mingw* | pw32* | cegcc*) \ + need_charset_alias=false ;; \ + *) \ + need_charset_alias=true ;; \ + esac ; \ + else \ + need_charset_alias=false ; \ + fi ; \ + if $$need_charset_alias; then \ + $(mkinstalldirs) $(DESTDIR)$(libdir) ; \ + fi ; \ + if test -f $(charset_alias); then \ + sed -f ref-add.sed $(charset_alias) > $(charset_tmp) ; \ + $(INSTALL_DATA) $(charset_tmp) $(charset_alias) ; \ + rm -f $(charset_tmp) ; \ + else \ + if $$need_charset_alias; then \ + sed -f ref-add.sed charset.alias > $(charset_tmp) ; \ + $(INSTALL_DATA) $(charset_tmp) $(charset_alias) ; \ + rm -f $(charset_tmp) ; \ + fi ; \ + fi + +uninstall-local: uninstall-localcharset +uninstall-localcharset: all-local + if test -f $(charset_alias); then \ + sed -f ref-del.sed $(charset_alias) > $(charset_tmp); \ + if grep '^# Packages using this file: $$' $(charset_tmp) \ + > /dev/null; then \ + rm -f $(charset_alias); \ + else \ + $(INSTALL_DATA) $(charset_tmp) $(charset_alias); \ + fi; \ + rm -f $(charset_tmp); \ + fi + +charset.alias: config.charset + $(AM_V_GEN)rm -f t-$@ $@ && \ + $(SHELL) $(srcdir)/config.charset '$(host)' > t-$@ && \ + mv t-$@ $@ + +SUFFIXES += .sed .sin +.sin.sed: + $(AM_V_GEN)rm -f t-$@ $@ && \ + sed -e '/^#/d' -e 's/@''PACKAGE''@/$(PACKAGE)/g' $< > t-$@ && \ + mv t-$@ $@ + +CLEANFILES += charset.alias ref-add.sed ref-del.sed + +EXTRA_DIST += config.charset ref-add.sin ref-del.sin + +## end gnulib module localcharset + +## begin gnulib module malloc-gnu + + +EXTRA_DIST += malloc.c + +EXTRA_libgnu_a_SOURCES += malloc.c + +## end gnulib module malloc-gnu + +## begin gnulib module malloc-posix + + +EXTRA_DIST += malloc.c + +EXTRA_libgnu_a_SOURCES += malloc.c + +## end gnulib module malloc-posix + +## begin gnulib module mbrtowc + + +EXTRA_DIST += mbrtowc.c + +EXTRA_libgnu_a_SOURCES += mbrtowc.c + +## end gnulib module mbrtowc + +## begin gnulib module mbsinit + + +EXTRA_DIST += mbsinit.c + +EXTRA_libgnu_a_SOURCES += mbsinit.c + +## end gnulib module mbsinit + +## begin gnulib module mbtowc + + +EXTRA_DIST += mbtowc-impl.h mbtowc.c + +EXTRA_libgnu_a_SOURCES += mbtowc.c + +## end gnulib module mbtowc + +## begin gnulib module msvc-inval + + +EXTRA_DIST += msvc-inval.c msvc-inval.h + +EXTRA_libgnu_a_SOURCES += msvc-inval.c + +## end gnulib module msvc-inval + +## begin gnulib module msvc-nothrow + + +EXTRA_DIST += msvc-nothrow.c msvc-nothrow.h + +EXTRA_libgnu_a_SOURCES += msvc-nothrow.c + +## end gnulib module msvc-nothrow + +## begin gnulib module nl_langinfo + + +EXTRA_DIST += nl_langinfo.c + +EXTRA_libgnu_a_SOURCES += nl_langinfo.c + +## end gnulib module nl_langinfo + +## begin 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 + +## end gnulib module regex + +## begin gnulib module snippet/_Noreturn + +# Because this Makefile snippet defines a variable used by other +# gnulib Makefile snippets, it must be present in all Makefile.am that +# need it. This is ensured by the applicability 'all' defined above. + +_NORETURN_H=$(top_srcdir)/./snippet/_Noreturn.h + +EXTRA_DIST += $(top_srcdir)/./snippet/_Noreturn.h + +## end gnulib module snippet/_Noreturn + +## begin gnulib module snippet/arg-nonnull + +# The BUILT_SOURCES created by this Makefile snippet are not used via #include +# statements but through direct file reference. Therefore this snippet must be +# present in all Makefile.am that need it. This is ensured by the applicability +# 'all' defined above. + +BUILT_SOURCES += arg-nonnull.h +# The arg-nonnull.h that gets inserted into generated .h files is the same as +# build-aux/snippet/arg-nonnull.h, except that it has the copyright header cut +# off. +arg-nonnull.h: $(top_srcdir)/./snippet/arg-nonnull.h + $(AM_V_GEN)rm -f $@-t $@ && \ + sed -n -e '/GL_ARG_NONNULL/,$$p' \ + < $(top_srcdir)/./snippet/arg-nonnull.h \ + > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += arg-nonnull.h arg-nonnull.h-t + +ARG_NONNULL_H=arg-nonnull.h + +EXTRA_DIST += $(top_srcdir)/./snippet/arg-nonnull.h + +## end gnulib module snippet/arg-nonnull + +## begin gnulib module snippet/c++defs + +# The BUILT_SOURCES created by this Makefile snippet are not used via #include +# statements but through direct file reference. Therefore this snippet must be +# present in all Makefile.am that need it. This is ensured by the applicability +# 'all' defined above. + +BUILT_SOURCES += c++defs.h +# The c++defs.h that gets inserted into generated .h files is the same as +# build-aux/snippet/c++defs.h, except that it has the copyright header cut off. +c++defs.h: $(top_srcdir)/./snippet/c++defs.h + $(AM_V_GEN)rm -f $@-t $@ && \ + sed -n -e '/_GL_CXXDEFS/,$$p' \ + < $(top_srcdir)/./snippet/c++defs.h \ + > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += c++defs.h c++defs.h-t + +CXXDEFS_H=c++defs.h + +EXTRA_DIST += $(top_srcdir)/./snippet/c++defs.h + +## end gnulib module snippet/c++defs + +## begin gnulib module snippet/warn-on-use + +BUILT_SOURCES += warn-on-use.h +# The warn-on-use.h that gets inserted into generated .h files is the same as +# build-aux/snippet/warn-on-use.h, except that it has the copyright header cut +# off. +warn-on-use.h: $(top_srcdir)/./snippet/warn-on-use.h + $(AM_V_GEN)rm -f $@-t $@ && \ + sed -n -e '/^.ifndef/,$$p' \ + < $(top_srcdir)/./snippet/warn-on-use.h \ + > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += warn-on-use.h warn-on-use.h-t + +WARN_ON_USE_H=warn-on-use.h + +EXTRA_DIST += $(top_srcdir)/./snippet/warn-on-use.h + +## end gnulib module snippet/warn-on-use + +## begin gnulib module stdbool + +BUILT_SOURCES += $(STDBOOL_H) + +# We need the following in order to create <stdbool.h> when the system +# doesn't have one that works. +if GL_GENERATE_STDBOOL_H +stdbool.h: stdbool.in.h $(top_builddir)/config.status + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's/@''HAVE__BOOL''@/$(HAVE__BOOL)/g' < $(srcdir)/stdbool.in.h; \ + } > $@-t && \ + mv $@-t $@ +else +stdbool.h: $(top_builddir)/config.status + rm -f $@ +endif +MOSTLYCLEANFILES += stdbool.h stdbool.h-t + +EXTRA_DIST += stdbool.in.h + +## end gnulib module stdbool + +## begin gnulib module stddef + +BUILT_SOURCES += $(STDDEF_H) + +# We need the following in order to create <stddef.h> when the system +# doesn't have one that works with the given compiler. +if GL_GENERATE_STDDEF_H +stddef.h: stddef.in.h $(top_builddir)/config.status + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */' && \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_STDDEF_H''@|$(NEXT_STDDEF_H)|g' \ + -e 's|@''HAVE_WCHAR_T''@|$(HAVE_WCHAR_T)|g' \ + -e 's|@''REPLACE_NULL''@|$(REPLACE_NULL)|g' \ + < $(srcdir)/stddef.in.h; \ + } > $@-t && \ + mv $@-t $@ +else +stddef.h: $(top_builddir)/config.status + rm -f $@ +endif +MOSTLYCLEANFILES += stddef.h stddef.h-t + +EXTRA_DIST += stddef.in.h + +## end gnulib module stddef + +## begin gnulib module stdint + +BUILT_SOURCES += $(STDINT_H) + +# We need the following in order to create <stdint.h> when the system +# doesn't have one that works with the given compiler. +if GL_GENERATE_STDINT_H +stdint.h: stdint.in.h $(top_builddir)/config.status + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's/@''HAVE_STDINT_H''@/$(HAVE_STDINT_H)/g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_STDINT_H''@|$(NEXT_STDINT_H)|g' \ + -e 's/@''HAVE_SYS_TYPES_H''@/$(HAVE_SYS_TYPES_H)/g' \ + -e 's/@''HAVE_INTTYPES_H''@/$(HAVE_INTTYPES_H)/g' \ + -e 's/@''HAVE_SYS_INTTYPES_H''@/$(HAVE_SYS_INTTYPES_H)/g' \ + -e 's/@''HAVE_SYS_BITYPES_H''@/$(HAVE_SYS_BITYPES_H)/g' \ + -e 's/@''HAVE_WCHAR_H''@/$(HAVE_WCHAR_H)/g' \ + -e 's/@''HAVE_LONG_LONG_INT''@/$(HAVE_LONG_LONG_INT)/g' \ + -e 's/@''HAVE_UNSIGNED_LONG_LONG_INT''@/$(HAVE_UNSIGNED_LONG_LONG_INT)/g' \ + -e 's/@''APPLE_UNIVERSAL_BUILD''@/$(APPLE_UNIVERSAL_BUILD)/g' \ + -e 's/@''BITSIZEOF_PTRDIFF_T''@/$(BITSIZEOF_PTRDIFF_T)/g' \ + -e 's/@''PTRDIFF_T_SUFFIX''@/$(PTRDIFF_T_SUFFIX)/g' \ + -e 's/@''BITSIZEOF_SIG_ATOMIC_T''@/$(BITSIZEOF_SIG_ATOMIC_T)/g' \ + -e 's/@''HAVE_SIGNED_SIG_ATOMIC_T''@/$(HAVE_SIGNED_SIG_ATOMIC_T)/g' \ + -e 's/@''SIG_ATOMIC_T_SUFFIX''@/$(SIG_ATOMIC_T_SUFFIX)/g' \ + -e 's/@''BITSIZEOF_SIZE_T''@/$(BITSIZEOF_SIZE_T)/g' \ + -e 's/@''SIZE_T_SUFFIX''@/$(SIZE_T_SUFFIX)/g' \ + -e 's/@''BITSIZEOF_WCHAR_T''@/$(BITSIZEOF_WCHAR_T)/g' \ + -e 's/@''HAVE_SIGNED_WCHAR_T''@/$(HAVE_SIGNED_WCHAR_T)/g' \ + -e 's/@''WCHAR_T_SUFFIX''@/$(WCHAR_T_SUFFIX)/g' \ + -e 's/@''BITSIZEOF_WINT_T''@/$(BITSIZEOF_WINT_T)/g' \ + -e 's/@''HAVE_SIGNED_WINT_T''@/$(HAVE_SIGNED_WINT_T)/g' \ + -e 's/@''WINT_T_SUFFIX''@/$(WINT_T_SUFFIX)/g' \ + < $(srcdir)/stdint.in.h; \ + } > $@-t && \ + mv $@-t $@ +else +stdint.h: $(top_builddir)/config.status + rm -f $@ +endif +MOSTLYCLEANFILES += stdint.h stdint.h-t + +EXTRA_DIST += stdint.in.h + +## end gnulib module stdint + +## begin gnulib module stdlib + +BUILT_SOURCES += stdlib.h + +# We need the following in order to create <stdlib.h> when the system +# doesn't have one that works with the given compiler. +stdlib.h: stdlib.in.h $(top_builddir)/config.status $(CXXDEFS_H) \ + $(_NORETURN_H) $(ARG_NONNULL_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */' && \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_STDLIB_H''@|$(NEXT_STDLIB_H)|g' \ + -e 's/@''GNULIB__EXIT''@/$(GNULIB__EXIT)/g' \ + -e 's/@''GNULIB_ATOLL''@/$(GNULIB_ATOLL)/g' \ + -e 's/@''GNULIB_CALLOC_POSIX''@/$(GNULIB_CALLOC_POSIX)/g' \ + -e 's/@''GNULIB_CANONICALIZE_FILE_NAME''@/$(GNULIB_CANONICALIZE_FILE_NAME)/g' \ + -e 's/@''GNULIB_GETLOADAVG''@/$(GNULIB_GETLOADAVG)/g' \ + -e 's/@''GNULIB_GETSUBOPT''@/$(GNULIB_GETSUBOPT)/g' \ + -e 's/@''GNULIB_GRANTPT''@/$(GNULIB_GRANTPT)/g' \ + -e 's/@''GNULIB_MALLOC_POSIX''@/$(GNULIB_MALLOC_POSIX)/g' \ + -e 's/@''GNULIB_MBTOWC''@/$(GNULIB_MBTOWC)/g' \ + -e 's/@''GNULIB_MKDTEMP''@/$(GNULIB_MKDTEMP)/g' \ + -e 's/@''GNULIB_MKOSTEMP''@/$(GNULIB_MKOSTEMP)/g' \ + -e 's/@''GNULIB_MKOSTEMPS''@/$(GNULIB_MKOSTEMPS)/g' \ + -e 's/@''GNULIB_MKSTEMP''@/$(GNULIB_MKSTEMP)/g' \ + -e 's/@''GNULIB_MKSTEMPS''@/$(GNULIB_MKSTEMPS)/g' \ + -e 's/@''GNULIB_POSIX_OPENPT''@/$(GNULIB_POSIX_OPENPT)/g' \ + -e 's/@''GNULIB_PTSNAME''@/$(GNULIB_PTSNAME)/g' \ + -e 's/@''GNULIB_PTSNAME_R''@/$(GNULIB_PTSNAME_R)/g' \ + -e 's/@''GNULIB_PUTENV''@/$(GNULIB_PUTENV)/g' \ + -e 's/@''GNULIB_RANDOM''@/$(GNULIB_RANDOM)/g' \ + -e 's/@''GNULIB_RANDOM_R''@/$(GNULIB_RANDOM_R)/g' \ + -e 's/@''GNULIB_REALLOC_POSIX''@/$(GNULIB_REALLOC_POSIX)/g' \ + -e 's/@''GNULIB_REALPATH''@/$(GNULIB_REALPATH)/g' \ + -e 's/@''GNULIB_RPMATCH''@/$(GNULIB_RPMATCH)/g' \ + -e 's/@''GNULIB_SETENV''@/$(GNULIB_SETENV)/g' \ + -e 's/@''GNULIB_STRTOD''@/$(GNULIB_STRTOD)/g' \ + -e 's/@''GNULIB_STRTOLL''@/$(GNULIB_STRTOLL)/g' \ + -e 's/@''GNULIB_STRTOULL''@/$(GNULIB_STRTOULL)/g' \ + -e 's/@''GNULIB_SYSTEM_POSIX''@/$(GNULIB_SYSTEM_POSIX)/g' \ + -e 's/@''GNULIB_UNLOCKPT''@/$(GNULIB_UNLOCKPT)/g' \ + -e 's/@''GNULIB_UNSETENV''@/$(GNULIB_UNSETENV)/g' \ + -e 's/@''GNULIB_WCTOMB''@/$(GNULIB_WCTOMB)/g' \ + < $(srcdir)/stdlib.in.h | \ + sed -e 's|@''HAVE__EXIT''@|$(HAVE__EXIT)|g' \ + -e 's|@''HAVE_ATOLL''@|$(HAVE_ATOLL)|g' \ + -e 's|@''HAVE_CANONICALIZE_FILE_NAME''@|$(HAVE_CANONICALIZE_FILE_NAME)|g' \ + -e 's|@''HAVE_DECL_GETLOADAVG''@|$(HAVE_DECL_GETLOADAVG)|g' \ + -e 's|@''HAVE_GETSUBOPT''@|$(HAVE_GETSUBOPT)|g' \ + -e 's|@''HAVE_GRANTPT''@|$(HAVE_GRANTPT)|g' \ + -e 's|@''HAVE_MKDTEMP''@|$(HAVE_MKDTEMP)|g' \ + -e 's|@''HAVE_MKOSTEMP''@|$(HAVE_MKOSTEMP)|g' \ + -e 's|@''HAVE_MKOSTEMPS''@|$(HAVE_MKOSTEMPS)|g' \ + -e 's|@''HAVE_MKSTEMP''@|$(HAVE_MKSTEMP)|g' \ + -e 's|@''HAVE_MKSTEMPS''@|$(HAVE_MKSTEMPS)|g' \ + -e 's|@''HAVE_POSIX_OPENPT''@|$(HAVE_POSIX_OPENPT)|g' \ + -e 's|@''HAVE_PTSNAME''@|$(HAVE_PTSNAME)|g' \ + -e 's|@''HAVE_PTSNAME_R''@|$(HAVE_PTSNAME_R)|g' \ + -e 's|@''HAVE_RANDOM''@|$(HAVE_RANDOM)|g' \ + -e 's|@''HAVE_RANDOM_H''@|$(HAVE_RANDOM_H)|g' \ + -e 's|@''HAVE_RANDOM_R''@|$(HAVE_RANDOM_R)|g' \ + -e 's|@''HAVE_REALPATH''@|$(HAVE_REALPATH)|g' \ + -e 's|@''HAVE_RPMATCH''@|$(HAVE_RPMATCH)|g' \ + -e 's|@''HAVE_DECL_SETENV''@|$(HAVE_DECL_SETENV)|g' \ + -e 's|@''HAVE_STRTOD''@|$(HAVE_STRTOD)|g' \ + -e 's|@''HAVE_STRTOLL''@|$(HAVE_STRTOLL)|g' \ + -e 's|@''HAVE_STRTOULL''@|$(HAVE_STRTOULL)|g' \ + -e 's|@''HAVE_STRUCT_RANDOM_DATA''@|$(HAVE_STRUCT_RANDOM_DATA)|g' \ + -e 's|@''HAVE_SYS_LOADAVG_H''@|$(HAVE_SYS_LOADAVG_H)|g' \ + -e 's|@''HAVE_UNLOCKPT''@|$(HAVE_UNLOCKPT)|g' \ + -e 's|@''HAVE_DECL_UNSETENV''@|$(HAVE_DECL_UNSETENV)|g' \ + -e 's|@''REPLACE_CALLOC''@|$(REPLACE_CALLOC)|g' \ + -e 's|@''REPLACE_CANONICALIZE_FILE_NAME''@|$(REPLACE_CANONICALIZE_FILE_NAME)|g' \ + -e 's|@''REPLACE_MALLOC''@|$(REPLACE_MALLOC)|g' \ + -e 's|@''REPLACE_MBTOWC''@|$(REPLACE_MBTOWC)|g' \ + -e 's|@''REPLACE_MKSTEMP''@|$(REPLACE_MKSTEMP)|g' \ + -e 's|@''REPLACE_PTSNAME_R''@|$(REPLACE_PTSNAME_R)|g' \ + -e 's|@''REPLACE_PUTENV''@|$(REPLACE_PUTENV)|g' \ + -e 's|@''REPLACE_RANDOM_R''@|$(REPLACE_RANDOM_R)|g' \ + -e 's|@''REPLACE_REALLOC''@|$(REPLACE_REALLOC)|g' \ + -e 's|@''REPLACE_REALPATH''@|$(REPLACE_REALPATH)|g' \ + -e 's|@''REPLACE_SETENV''@|$(REPLACE_SETENV)|g' \ + -e 's|@''REPLACE_STRTOD''@|$(REPLACE_STRTOD)|g' \ + -e 's|@''REPLACE_UNSETENV''@|$(REPLACE_UNSETENV)|g' \ + -e 's|@''REPLACE_WCTOMB''@|$(REPLACE_WCTOMB)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _Noreturn/r $(_NORETURN_H)' \ + -e '/definition of _GL_ARG_NONNULL/r $(ARG_NONNULL_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)'; \ + } > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += stdlib.h stdlib.h-t + +EXTRA_DIST += stdlib.in.h + +## end gnulib module stdlib + +## begin gnulib module strcase + + +EXTRA_DIST += strcasecmp.c strncasecmp.c + +EXTRA_libgnu_a_SOURCES += strcasecmp.c strncasecmp.c + +## end gnulib module strcase + +## begin gnulib module streq + + +EXTRA_DIST += streq.h + +## end gnulib module streq + +## begin gnulib module strings + +BUILT_SOURCES += strings.h + +# We need the following in order to create <strings.h> when the system +# doesn't have one that works with the given compiler. +strings.h: strings.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(WARN_ON_USE_H) $(ARG_NONNULL_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */' && \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''HAVE_STRINGS_H''@|$(HAVE_STRINGS_H)|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_STRINGS_H''@|$(NEXT_STRINGS_H)|g' \ + -e 's|@''GNULIB_FFS''@|$(GNULIB_FFS)|g' \ + -e 's|@''HAVE_FFS''@|$(HAVE_FFS)|g' \ + -e 's|@''HAVE_STRCASECMP''@|$(HAVE_STRCASECMP)|g' \ + -e 's|@''HAVE_DECL_STRNCASECMP''@|$(HAVE_DECL_STRNCASECMP)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_ARG_NONNULL/r $(ARG_NONNULL_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)' \ + < $(srcdir)/strings.in.h; \ + } > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += strings.h strings.h-t + +EXTRA_DIST += strings.in.h + +## end gnulib module strings + +## begin gnulib module sys_file + +BUILT_SOURCES += sys/file.h + +# We need the following in order to create <sys/file.h> when the system +# has one that is incomplete. +sys/file.h: sys_file.in.h $(top_builddir)/config.status $(WARN_ON_USE_H) + $(AM_V_at)$(MKDIR_P) sys + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's/@''HAVE_SYS_FILE_H''@/$(HAVE_SYS_FILE_H)/g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_SYS_FILE_H''@|$(NEXT_SYS_FILE_H)|g' \ + -e 's/@''HAVE_FLOCK''@/$(HAVE_FLOCK)/g' \ + -e 's/@''GNULIB_FLOCK''@/$(GNULIB_FLOCK)/g' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)' \ + < $(srcdir)/sys_file.in.h; \ + } > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += sys/file.h sys/file.h-t +MOSTLYCLEANDIRS += sys + +EXTRA_DIST += sys_file.in.h + +## end gnulib module sys_file + +## begin gnulib module sys_types + +BUILT_SOURCES += sys/types.h + +# We need the following in order to create <sys/types.h> when the system +# doesn't have one that works with the given compiler. +sys/types.h: sys_types.in.h $(top_builddir)/config.status + $(AM_V_at)$(MKDIR_P) sys + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_SYS_TYPES_H''@|$(NEXT_SYS_TYPES_H)|g' \ + < $(srcdir)/sys_types.in.h; \ + } > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += sys/types.h sys/types.h-t + +EXTRA_DIST += sys_types.in.h + +## end gnulib module sys_types + +## begin gnulib module unistd + +BUILT_SOURCES += unistd.h + +# We need the following in order to create an empty placeholder for +# <unistd.h> when the system doesn't have one. +unistd.h: unistd.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(ARG_NONNULL_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''HAVE_UNISTD_H''@|$(HAVE_UNISTD_H)|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_UNISTD_H''@|$(NEXT_UNISTD_H)|g' \ + -e 's/@''GNULIB_CHDIR''@/$(GNULIB_CHDIR)/g' \ + -e 's/@''GNULIB_CHOWN''@/$(GNULIB_CHOWN)/g' \ + -e 's/@''GNULIB_CLOSE''@/$(GNULIB_CLOSE)/g' \ + -e 's/@''GNULIB_DUP''@/$(GNULIB_DUP)/g' \ + -e 's/@''GNULIB_DUP2''@/$(GNULIB_DUP2)/g' \ + -e 's/@''GNULIB_DUP3''@/$(GNULIB_DUP3)/g' \ + -e 's/@''GNULIB_ENVIRON''@/$(GNULIB_ENVIRON)/g' \ + -e 's/@''GNULIB_EUIDACCESS''@/$(GNULIB_EUIDACCESS)/g' \ + -e 's/@''GNULIB_FACCESSAT''@/$(GNULIB_FACCESSAT)/g' \ + -e 's/@''GNULIB_FCHDIR''@/$(GNULIB_FCHDIR)/g' \ + -e 's/@''GNULIB_FCHOWNAT''@/$(GNULIB_FCHOWNAT)/g' \ + -e 's/@''GNULIB_FDATASYNC''@/$(GNULIB_FDATASYNC)/g' \ + -e 's/@''GNULIB_FSYNC''@/$(GNULIB_FSYNC)/g' \ + -e 's/@''GNULIB_FTRUNCATE''@/$(GNULIB_FTRUNCATE)/g' \ + -e 's/@''GNULIB_GETCWD''@/$(GNULIB_GETCWD)/g' \ + -e 's/@''GNULIB_GETDOMAINNAME''@/$(GNULIB_GETDOMAINNAME)/g' \ + -e 's/@''GNULIB_GETDTABLESIZE''@/$(GNULIB_GETDTABLESIZE)/g' \ + -e 's/@''GNULIB_GETGROUPS''@/$(GNULIB_GETGROUPS)/g' \ + -e 's/@''GNULIB_GETHOSTNAME''@/$(GNULIB_GETHOSTNAME)/g' \ + -e 's/@''GNULIB_GETLOGIN''@/$(GNULIB_GETLOGIN)/g' \ + -e 's/@''GNULIB_GETLOGIN_R''@/$(GNULIB_GETLOGIN_R)/g' \ + -e 's/@''GNULIB_GETPAGESIZE''@/$(GNULIB_GETPAGESIZE)/g' \ + -e 's/@''GNULIB_GETUSERSHELL''@/$(GNULIB_GETUSERSHELL)/g' \ + -e 's/@''GNULIB_GROUP_MEMBER''@/$(GNULIB_GROUP_MEMBER)/g' \ + -e 's/@''GNULIB_ISATTY''@/$(GNULIB_ISATTY)/g' \ + -e 's/@''GNULIB_LCHOWN''@/$(GNULIB_LCHOWN)/g' \ + -e 's/@''GNULIB_LINK''@/$(GNULIB_LINK)/g' \ + -e 's/@''GNULIB_LINKAT''@/$(GNULIB_LINKAT)/g' \ + -e 's/@''GNULIB_LSEEK''@/$(GNULIB_LSEEK)/g' \ + -e 's/@''GNULIB_PIPE''@/$(GNULIB_PIPE)/g' \ + -e 's/@''GNULIB_PIPE2''@/$(GNULIB_PIPE2)/g' \ + -e 's/@''GNULIB_PREAD''@/$(GNULIB_PREAD)/g' \ + -e 's/@''GNULIB_PWRITE''@/$(GNULIB_PWRITE)/g' \ + -e 's/@''GNULIB_READ''@/$(GNULIB_READ)/g' \ + -e 's/@''GNULIB_READLINK''@/$(GNULIB_READLINK)/g' \ + -e 's/@''GNULIB_READLINKAT''@/$(GNULIB_READLINKAT)/g' \ + -e 's/@''GNULIB_RMDIR''@/$(GNULIB_RMDIR)/g' \ + -e 's/@''GNULIB_SETHOSTNAME''@/$(GNULIB_SETHOSTNAME)/g' \ + -e 's/@''GNULIB_SLEEP''@/$(GNULIB_SLEEP)/g' \ + -e 's/@''GNULIB_SYMLINK''@/$(GNULIB_SYMLINK)/g' \ + -e 's/@''GNULIB_SYMLINKAT''@/$(GNULIB_SYMLINKAT)/g' \ + -e 's/@''GNULIB_TTYNAME_R''@/$(GNULIB_TTYNAME_R)/g' \ + -e 's/@''GNULIB_UNISTD_H_GETOPT''@/0$(GNULIB_GL_UNISTD_H_GETOPT)/g' \ + -e 's/@''GNULIB_UNISTD_H_NONBLOCKING''@/$(GNULIB_UNISTD_H_NONBLOCKING)/g' \ + -e 's/@''GNULIB_UNISTD_H_SIGPIPE''@/$(GNULIB_UNISTD_H_SIGPIPE)/g' \ + -e 's/@''GNULIB_UNLINK''@/$(GNULIB_UNLINK)/g' \ + -e 's/@''GNULIB_UNLINKAT''@/$(GNULIB_UNLINKAT)/g' \ + -e 's/@''GNULIB_USLEEP''@/$(GNULIB_USLEEP)/g' \ + -e 's/@''GNULIB_WRITE''@/$(GNULIB_WRITE)/g' \ + < $(srcdir)/unistd.in.h | \ + sed -e 's|@''HAVE_CHOWN''@|$(HAVE_CHOWN)|g' \ + -e 's|@''HAVE_DUP2''@|$(HAVE_DUP2)|g' \ + -e 's|@''HAVE_DUP3''@|$(HAVE_DUP3)|g' \ + -e 's|@''HAVE_EUIDACCESS''@|$(HAVE_EUIDACCESS)|g' \ + -e 's|@''HAVE_FACCESSAT''@|$(HAVE_FACCESSAT)|g' \ + -e 's|@''HAVE_FCHDIR''@|$(HAVE_FCHDIR)|g' \ + -e 's|@''HAVE_FCHOWNAT''@|$(HAVE_FCHOWNAT)|g' \ + -e 's|@''HAVE_FDATASYNC''@|$(HAVE_FDATASYNC)|g' \ + -e 's|@''HAVE_FSYNC''@|$(HAVE_FSYNC)|g' \ + -e 's|@''HAVE_FTRUNCATE''@|$(HAVE_FTRUNCATE)|g' \ + -e 's|@''HAVE_GETDTABLESIZE''@|$(HAVE_GETDTABLESIZE)|g' \ + -e 's|@''HAVE_GETGROUPS''@|$(HAVE_GETGROUPS)|g' \ + -e 's|@''HAVE_GETHOSTNAME''@|$(HAVE_GETHOSTNAME)|g' \ + -e 's|@''HAVE_GETLOGIN''@|$(HAVE_GETLOGIN)|g' \ + -e 's|@''HAVE_GETPAGESIZE''@|$(HAVE_GETPAGESIZE)|g' \ + -e 's|@''HAVE_GROUP_MEMBER''@|$(HAVE_GROUP_MEMBER)|g' \ + -e 's|@''HAVE_LCHOWN''@|$(HAVE_LCHOWN)|g' \ + -e 's|@''HAVE_LINK''@|$(HAVE_LINK)|g' \ + -e 's|@''HAVE_LINKAT''@|$(HAVE_LINKAT)|g' \ + -e 's|@''HAVE_PIPE''@|$(HAVE_PIPE)|g' \ + -e 's|@''HAVE_PIPE2''@|$(HAVE_PIPE2)|g' \ + -e 's|@''HAVE_PREAD''@|$(HAVE_PREAD)|g' \ + -e 's|@''HAVE_PWRITE''@|$(HAVE_PWRITE)|g' \ + -e 's|@''HAVE_READLINK''@|$(HAVE_READLINK)|g' \ + -e 's|@''HAVE_READLINKAT''@|$(HAVE_READLINKAT)|g' \ + -e 's|@''HAVE_SETHOSTNAME''@|$(HAVE_SETHOSTNAME)|g' \ + -e 's|@''HAVE_SLEEP''@|$(HAVE_SLEEP)|g' \ + -e 's|@''HAVE_SYMLINK''@|$(HAVE_SYMLINK)|g' \ + -e 's|@''HAVE_SYMLINKAT''@|$(HAVE_SYMLINKAT)|g' \ + -e 's|@''HAVE_UNLINKAT''@|$(HAVE_UNLINKAT)|g' \ + -e 's|@''HAVE_USLEEP''@|$(HAVE_USLEEP)|g' \ + -e 's|@''HAVE_DECL_ENVIRON''@|$(HAVE_DECL_ENVIRON)|g' \ + -e 's|@''HAVE_DECL_FCHDIR''@|$(HAVE_DECL_FCHDIR)|g' \ + -e 's|@''HAVE_DECL_FDATASYNC''@|$(HAVE_DECL_FDATASYNC)|g' \ + -e 's|@''HAVE_DECL_GETDOMAINNAME''@|$(HAVE_DECL_GETDOMAINNAME)|g' \ + -e 's|@''HAVE_DECL_GETLOGIN_R''@|$(HAVE_DECL_GETLOGIN_R)|g' \ + -e 's|@''HAVE_DECL_GETPAGESIZE''@|$(HAVE_DECL_GETPAGESIZE)|g' \ + -e 's|@''HAVE_DECL_GETUSERSHELL''@|$(HAVE_DECL_GETUSERSHELL)|g' \ + -e 's|@''HAVE_DECL_SETHOSTNAME''@|$(HAVE_DECL_SETHOSTNAME)|g' \ + -e 's|@''HAVE_DECL_TTYNAME_R''@|$(HAVE_DECL_TTYNAME_R)|g' \ + -e 's|@''HAVE_OS_H''@|$(HAVE_OS_H)|g' \ + -e 's|@''HAVE_SYS_PARAM_H''@|$(HAVE_SYS_PARAM_H)|g' \ + | \ + sed -e 's|@''REPLACE_CHOWN''@|$(REPLACE_CHOWN)|g' \ + -e 's|@''REPLACE_CLOSE''@|$(REPLACE_CLOSE)|g' \ + -e 's|@''REPLACE_DUP''@|$(REPLACE_DUP)|g' \ + -e 's|@''REPLACE_DUP2''@|$(REPLACE_DUP2)|g' \ + -e 's|@''REPLACE_FCHOWNAT''@|$(REPLACE_FCHOWNAT)|g' \ + -e 's|@''REPLACE_GETCWD''@|$(REPLACE_GETCWD)|g' \ + -e 's|@''REPLACE_GETDOMAINNAME''@|$(REPLACE_GETDOMAINNAME)|g' \ + -e 's|@''REPLACE_GETLOGIN_R''@|$(REPLACE_GETLOGIN_R)|g' \ + -e 's|@''REPLACE_GETGROUPS''@|$(REPLACE_GETGROUPS)|g' \ + -e 's|@''REPLACE_GETPAGESIZE''@|$(REPLACE_GETPAGESIZE)|g' \ + -e 's|@''REPLACE_ISATTY''@|$(REPLACE_ISATTY)|g' \ + -e 's|@''REPLACE_LCHOWN''@|$(REPLACE_LCHOWN)|g' \ + -e 's|@''REPLACE_LINK''@|$(REPLACE_LINK)|g' \ + -e 's|@''REPLACE_LINKAT''@|$(REPLACE_LINKAT)|g' \ + -e 's|@''REPLACE_LSEEK''@|$(REPLACE_LSEEK)|g' \ + -e 's|@''REPLACE_PREAD''@|$(REPLACE_PREAD)|g' \ + -e 's|@''REPLACE_PWRITE''@|$(REPLACE_PWRITE)|g' \ + -e 's|@''REPLACE_READ''@|$(REPLACE_READ)|g' \ + -e 's|@''REPLACE_READLINK''@|$(REPLACE_READLINK)|g' \ + -e 's|@''REPLACE_RMDIR''@|$(REPLACE_RMDIR)|g' \ + -e 's|@''REPLACE_SLEEP''@|$(REPLACE_SLEEP)|g' \ + -e 's|@''REPLACE_SYMLINK''@|$(REPLACE_SYMLINK)|g' \ + -e 's|@''REPLACE_TTYNAME_R''@|$(REPLACE_TTYNAME_R)|g' \ + -e 's|@''REPLACE_UNLINK''@|$(REPLACE_UNLINK)|g' \ + -e 's|@''REPLACE_UNLINKAT''@|$(REPLACE_UNLINKAT)|g' \ + -e 's|@''REPLACE_USLEEP''@|$(REPLACE_USLEEP)|g' \ + -e 's|@''REPLACE_WRITE''@|$(REPLACE_WRITE)|g' \ + -e 's|@''UNISTD_H_HAVE_WINSOCK2_H''@|$(UNISTD_H_HAVE_WINSOCK2_H)|g' \ + -e 's|@''UNISTD_H_HAVE_WINSOCK2_H_AND_USE_SOCKETS''@|$(UNISTD_H_HAVE_WINSOCK2_H_AND_USE_SOCKETS)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_ARG_NONNULL/r $(ARG_NONNULL_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)'; \ + } > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += unistd.h unistd.h-t + +EXTRA_DIST += unistd.in.h + +## end gnulib module unistd + +## begin gnulib module verify + + +EXTRA_DIST += verify.h + +## end gnulib module verify + +## begin gnulib module wchar + +BUILT_SOURCES += wchar.h + +# We need the following in order to create <wchar.h> when the system +# version does not work standalone. +wchar.h: wchar.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(ARG_NONNULL_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''HAVE_FEATURES_H''@|$(HAVE_FEATURES_H)|g' \ + -e 's|@''NEXT_WCHAR_H''@|$(NEXT_WCHAR_H)|g' \ + -e 's|@''HAVE_WCHAR_H''@|$(HAVE_WCHAR_H)|g' \ + -e 's/@''GNULIB_BTOWC''@/$(GNULIB_BTOWC)/g' \ + -e 's/@''GNULIB_WCTOB''@/$(GNULIB_WCTOB)/g' \ + -e 's/@''GNULIB_MBSINIT''@/$(GNULIB_MBSINIT)/g' \ + -e 's/@''GNULIB_MBRTOWC''@/$(GNULIB_MBRTOWC)/g' \ + -e 's/@''GNULIB_MBRLEN''@/$(GNULIB_MBRLEN)/g' \ + -e 's/@''GNULIB_MBSRTOWCS''@/$(GNULIB_MBSRTOWCS)/g' \ + -e 's/@''GNULIB_MBSNRTOWCS''@/$(GNULIB_MBSNRTOWCS)/g' \ + -e 's/@''GNULIB_WCRTOMB''@/$(GNULIB_WCRTOMB)/g' \ + -e 's/@''GNULIB_WCSRTOMBS''@/$(GNULIB_WCSRTOMBS)/g' \ + -e 's/@''GNULIB_WCSNRTOMBS''@/$(GNULIB_WCSNRTOMBS)/g' \ + -e 's/@''GNULIB_WCWIDTH''@/$(GNULIB_WCWIDTH)/g' \ + -e 's/@''GNULIB_WMEMCHR''@/$(GNULIB_WMEMCHR)/g' \ + -e 's/@''GNULIB_WMEMCMP''@/$(GNULIB_WMEMCMP)/g' \ + -e 's/@''GNULIB_WMEMCPY''@/$(GNULIB_WMEMCPY)/g' \ + -e 's/@''GNULIB_WMEMMOVE''@/$(GNULIB_WMEMMOVE)/g' \ + -e 's/@''GNULIB_WMEMSET''@/$(GNULIB_WMEMSET)/g' \ + -e 's/@''GNULIB_WCSLEN''@/$(GNULIB_WCSLEN)/g' \ + -e 's/@''GNULIB_WCSNLEN''@/$(GNULIB_WCSNLEN)/g' \ + -e 's/@''GNULIB_WCSCPY''@/$(GNULIB_WCSCPY)/g' \ + -e 's/@''GNULIB_WCPCPY''@/$(GNULIB_WCPCPY)/g' \ + -e 's/@''GNULIB_WCSNCPY''@/$(GNULIB_WCSNCPY)/g' \ + -e 's/@''GNULIB_WCPNCPY''@/$(GNULIB_WCPNCPY)/g' \ + -e 's/@''GNULIB_WCSCAT''@/$(GNULIB_WCSCAT)/g' \ + -e 's/@''GNULIB_WCSNCAT''@/$(GNULIB_WCSNCAT)/g' \ + -e 's/@''GNULIB_WCSCMP''@/$(GNULIB_WCSCMP)/g' \ + -e 's/@''GNULIB_WCSNCMP''@/$(GNULIB_WCSNCMP)/g' \ + -e 's/@''GNULIB_WCSCASECMP''@/$(GNULIB_WCSCASECMP)/g' \ + -e 's/@''GNULIB_WCSNCASECMP''@/$(GNULIB_WCSNCASECMP)/g' \ + -e 's/@''GNULIB_WCSCOLL''@/$(GNULIB_WCSCOLL)/g' \ + -e 's/@''GNULIB_WCSXFRM''@/$(GNULIB_WCSXFRM)/g' \ + -e 's/@''GNULIB_WCSDUP''@/$(GNULIB_WCSDUP)/g' \ + -e 's/@''GNULIB_WCSCHR''@/$(GNULIB_WCSCHR)/g' \ + -e 's/@''GNULIB_WCSRCHR''@/$(GNULIB_WCSRCHR)/g' \ + -e 's/@''GNULIB_WCSCSPN''@/$(GNULIB_WCSCSPN)/g' \ + -e 's/@''GNULIB_WCSSPN''@/$(GNULIB_WCSSPN)/g' \ + -e 's/@''GNULIB_WCSPBRK''@/$(GNULIB_WCSPBRK)/g' \ + -e 's/@''GNULIB_WCSSTR''@/$(GNULIB_WCSSTR)/g' \ + -e 's/@''GNULIB_WCSTOK''@/$(GNULIB_WCSTOK)/g' \ + -e 's/@''GNULIB_WCSWIDTH''@/$(GNULIB_WCSWIDTH)/g' \ + < $(srcdir)/wchar.in.h | \ + sed -e 's|@''HAVE_WINT_T''@|$(HAVE_WINT_T)|g' \ + -e 's|@''HAVE_BTOWC''@|$(HAVE_BTOWC)|g' \ + -e 's|@''HAVE_MBSINIT''@|$(HAVE_MBSINIT)|g' \ + -e 's|@''HAVE_MBRTOWC''@|$(HAVE_MBRTOWC)|g' \ + -e 's|@''HAVE_MBRLEN''@|$(HAVE_MBRLEN)|g' \ + -e 's|@''HAVE_MBSRTOWCS''@|$(HAVE_MBSRTOWCS)|g' \ + -e 's|@''HAVE_MBSNRTOWCS''@|$(HAVE_MBSNRTOWCS)|g' \ + -e 's|@''HAVE_WCRTOMB''@|$(HAVE_WCRTOMB)|g' \ + -e 's|@''HAVE_WCSRTOMBS''@|$(HAVE_WCSRTOMBS)|g' \ + -e 's|@''HAVE_WCSNRTOMBS''@|$(HAVE_WCSNRTOMBS)|g' \ + -e 's|@''HAVE_WMEMCHR''@|$(HAVE_WMEMCHR)|g' \ + -e 's|@''HAVE_WMEMCMP''@|$(HAVE_WMEMCMP)|g' \ + -e 's|@''HAVE_WMEMCPY''@|$(HAVE_WMEMCPY)|g' \ + -e 's|@''HAVE_WMEMMOVE''@|$(HAVE_WMEMMOVE)|g' \ + -e 's|@''HAVE_WMEMSET''@|$(HAVE_WMEMSET)|g' \ + -e 's|@''HAVE_WCSLEN''@|$(HAVE_WCSLEN)|g' \ + -e 's|@''HAVE_WCSNLEN''@|$(HAVE_WCSNLEN)|g' \ + -e 's|@''HAVE_WCSCPY''@|$(HAVE_WCSCPY)|g' \ + -e 's|@''HAVE_WCPCPY''@|$(HAVE_WCPCPY)|g' \ + -e 's|@''HAVE_WCSNCPY''@|$(HAVE_WCSNCPY)|g' \ + -e 's|@''HAVE_WCPNCPY''@|$(HAVE_WCPNCPY)|g' \ + -e 's|@''HAVE_WCSCAT''@|$(HAVE_WCSCAT)|g' \ + -e 's|@''HAVE_WCSNCAT''@|$(HAVE_WCSNCAT)|g' \ + -e 's|@''HAVE_WCSCMP''@|$(HAVE_WCSCMP)|g' \ + -e 's|@''HAVE_WCSNCMP''@|$(HAVE_WCSNCMP)|g' \ + -e 's|@''HAVE_WCSCASECMP''@|$(HAVE_WCSCASECMP)|g' \ + -e 's|@''HAVE_WCSNCASECMP''@|$(HAVE_WCSNCASECMP)|g' \ + -e 's|@''HAVE_WCSCOLL''@|$(HAVE_WCSCOLL)|g' \ + -e 's|@''HAVE_WCSXFRM''@|$(HAVE_WCSXFRM)|g' \ + -e 's|@''HAVE_WCSDUP''@|$(HAVE_WCSDUP)|g' \ + -e 's|@''HAVE_WCSCHR''@|$(HAVE_WCSCHR)|g' \ + -e 's|@''HAVE_WCSRCHR''@|$(HAVE_WCSRCHR)|g' \ + -e 's|@''HAVE_WCSCSPN''@|$(HAVE_WCSCSPN)|g' \ + -e 's|@''HAVE_WCSSPN''@|$(HAVE_WCSSPN)|g' \ + -e 's|@''HAVE_WCSPBRK''@|$(HAVE_WCSPBRK)|g' \ + -e 's|@''HAVE_WCSSTR''@|$(HAVE_WCSSTR)|g' \ + -e 's|@''HAVE_WCSTOK''@|$(HAVE_WCSTOK)|g' \ + -e 's|@''HAVE_WCSWIDTH''@|$(HAVE_WCSWIDTH)|g' \ + -e 's|@''HAVE_DECL_WCTOB''@|$(HAVE_DECL_WCTOB)|g' \ + -e 's|@''HAVE_DECL_WCWIDTH''@|$(HAVE_DECL_WCWIDTH)|g' \ + | \ + sed -e 's|@''REPLACE_MBSTATE_T''@|$(REPLACE_MBSTATE_T)|g' \ + -e 's|@''REPLACE_BTOWC''@|$(REPLACE_BTOWC)|g' \ + -e 's|@''REPLACE_WCTOB''@|$(REPLACE_WCTOB)|g' \ + -e 's|@''REPLACE_MBSINIT''@|$(REPLACE_MBSINIT)|g' \ + -e 's|@''REPLACE_MBRTOWC''@|$(REPLACE_MBRTOWC)|g' \ + -e 's|@''REPLACE_MBRLEN''@|$(REPLACE_MBRLEN)|g' \ + -e 's|@''REPLACE_MBSRTOWCS''@|$(REPLACE_MBSRTOWCS)|g' \ + -e 's|@''REPLACE_MBSNRTOWCS''@|$(REPLACE_MBSNRTOWCS)|g' \ + -e 's|@''REPLACE_WCRTOMB''@|$(REPLACE_WCRTOMB)|g' \ + -e 's|@''REPLACE_WCSRTOMBS''@|$(REPLACE_WCSRTOMBS)|g' \ + -e 's|@''REPLACE_WCSNRTOMBS''@|$(REPLACE_WCSNRTOMBS)|g' \ + -e 's|@''REPLACE_WCWIDTH''@|$(REPLACE_WCWIDTH)|g' \ + -e 's|@''REPLACE_WCSWIDTH''@|$(REPLACE_WCSWIDTH)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_ARG_NONNULL/r $(ARG_NONNULL_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)'; \ + } > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += wchar.h wchar.h-t + +EXTRA_DIST += wchar.in.h + +## end gnulib module wchar + +## begin gnulib module wcrtomb + + +EXTRA_DIST += wcrtomb.c + +EXTRA_libgnu_a_SOURCES += wcrtomb.c + +## end gnulib module wcrtomb + +## begin gnulib module wctype-h + +BUILT_SOURCES += wctype.h + +# We need the following in order to create <wctype.h> when the system +# doesn't have one that works with the given compiler. +wctype.h: wctype.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's/@''HAVE_WCTYPE_H''@/$(HAVE_WCTYPE_H)/g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_WCTYPE_H''@|$(NEXT_WCTYPE_H)|g' \ + -e 's/@''GNULIB_ISWBLANK''@/$(GNULIB_ISWBLANK)/g' \ + -e 's/@''GNULIB_WCTYPE''@/$(GNULIB_WCTYPE)/g' \ + -e 's/@''GNULIB_ISWCTYPE''@/$(GNULIB_ISWCTYPE)/g' \ + -e 's/@''GNULIB_WCTRANS''@/$(GNULIB_WCTRANS)/g' \ + -e 's/@''GNULIB_TOWCTRANS''@/$(GNULIB_TOWCTRANS)/g' \ + -e 's/@''HAVE_ISWBLANK''@/$(HAVE_ISWBLANK)/g' \ + -e 's/@''HAVE_ISWCNTRL''@/$(HAVE_ISWCNTRL)/g' \ + -e 's/@''HAVE_WCTYPE_T''@/$(HAVE_WCTYPE_T)/g' \ + -e 's/@''HAVE_WCTRANS_T''@/$(HAVE_WCTRANS_T)/g' \ + -e 's/@''HAVE_WINT_T''@/$(HAVE_WINT_T)/g' \ + -e 's/@''REPLACE_ISWBLANK''@/$(REPLACE_ISWBLANK)/g' \ + -e 's/@''REPLACE_ISWCNTRL''@/$(REPLACE_ISWCNTRL)/g' \ + -e 's/@''REPLACE_TOWLOWER''@/$(REPLACE_TOWLOWER)/g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)' \ + < $(srcdir)/wctype.in.h; \ + } > $@-t && \ + mv $@-t $@ +MOSTLYCLEANFILES += wctype.h wctype.h-t + +EXTRA_DIST += wctype.in.h + +## end gnulib module wctype-h + + +mostlyclean-local: mostlyclean-generic + @for dir in '' $(MOSTLYCLEANDIRS); do \ + if test -n "$$dir" && test -d $$dir; then \ + echo "rmdir $$dir"; rmdir $$dir; \ + fi; \ + done; \ + : diff --git a/lib/Makefile.in b/lib/Makefile.in new file mode 100644 index 0000000..4102ac5 --- /dev/null +++ b/lib/Makefile.in @@ -0,0 +1,1812 @@ +# Makefile.in generated by automake 1.11.6 from Makefile.am. +# @configure_input@ + +# Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, +# 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011 Free Software +# Foundation, Inc. +# This Makefile.in 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. + +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY, to the extent permitted by law; without +# even the implied warranty of MERCHANTABILITY or FITNESS FOR A +# PARTICULAR PURPOSE. + +@SET_MAKE@ + +# Copyright (C) 2002-2012 Free Software Foundation, Inc. +# +# This file 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. +# +# This file 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 this file. If not, see <http://www.gnu.org/licenses/>. +# +# As a special exception to the GNU General Public License, +# this file may be distributed as part of a program that +# contains a configuration script generated by Autoconf, under +# the same distribution terms as the rest of that program. +# +# Generated by gnulib-tool. +# Reproduce by: gnulib-tool --import --dir=. --lib=libgnu --source-base=lib --m4-base=m4 --doc-base=doc --tests-base=tests --aux-dir=. --no-conditional-dependencies --no-libtool --macro-prefix=gl --no-vc-files flock regex + + + +VPATH = @srcdir@ +am__make_dryrun = \ + { \ + am__dry=no; \ + case $$MAKEFLAGS in \ + *\\[\ \ ]*) \ + echo 'am--echo: ; @echo "AM" OK' | $(MAKE) -f - 2>/dev/null \ + | grep '^AM OK$$' >/dev/null || am__dry=yes;; \ + *) \ + for am__flg in $$MAKEFLAGS; do \ + case $$am__flg in \ + *=*|--*) ;; \ + *n*) am__dry=yes; break;; \ + esac; \ + done;; \ + esac; \ + test $$am__dry = yes; \ + } +pkgdatadir = $(datadir)/@PACKAGE@ +pkgincludedir = $(includedir)/@PACKAGE@ +pkglibdir = $(libdir)/@PACKAGE@ +am__cd = CDPATH="$${ZSH_VERSION+.}$(PATH_SEPARATOR)" && cd +install_sh_DATA = $(install_sh) -c -m 644 +install_sh_PROGRAM = $(install_sh) -c +install_sh_SCRIPT = $(install_sh) -c +INSTALL_HEADER = $(INSTALL_DATA) +transform = $(program_transform_name) +NORMAL_INSTALL = : +PRE_INSTALL = : +POST_INSTALL = : +NORMAL_UNINSTALL = : +PRE_UNINSTALL = : +POST_UNINSTALL = : +build_triplet = @build@ +host_triplet = @host@ +subdir = lib +DIST_COMMON = $(noinst_HEADERS) $(srcdir)/Makefile.am \ + $(srcdir)/Makefile.in +ACLOCAL_M4 = $(top_srcdir)/aclocal.m4 +am__aclocal_m4_deps = $(top_srcdir)/m4/00gnulib.m4 \ + $(top_srcdir)/m4/alloca.m4 $(top_srcdir)/m4/btowc.m4 \ + $(top_srcdir)/m4/codeset.m4 $(top_srcdir)/m4/configmake.m4 \ + $(top_srcdir)/m4/extensions.m4 $(top_srcdir)/m4/fcntl-o.m4 \ + $(top_srcdir)/m4/flock.m4 $(top_srcdir)/m4/glibc21.m4 \ + $(top_srcdir)/m4/gnulib-common.m4 \ + $(top_srcdir)/m4/gnulib-comp.m4 \ + $(top_srcdir)/m4/include_next.m4 \ + $(top_srcdir)/m4/langinfo_h.m4 \ + $(top_srcdir)/m4/localcharset.m4 $(top_srcdir)/m4/locale-fr.m4 \ + $(top_srcdir)/m4/locale-ja.m4 $(top_srcdir)/m4/locale-zh.m4 \ + $(top_srcdir)/m4/longlong.m4 $(top_srcdir)/m4/malloc.m4 \ + $(top_srcdir)/m4/mbrtowc.m4 $(top_srcdir)/m4/mbsinit.m4 \ + $(top_srcdir)/m4/mbstate_t.m4 $(top_srcdir)/m4/mbtowc.m4 \ + $(top_srcdir)/m4/msvc-inval.m4 \ + $(top_srcdir)/m4/msvc-nothrow.m4 $(top_srcdir)/m4/multiarch.m4 \ + $(top_srcdir)/m4/nl_langinfo.m4 $(top_srcdir)/m4/onceonly.m4 \ + $(top_srcdir)/m4/regex.m4 $(top_srcdir)/m4/ssize_t.m4 \ + $(top_srcdir)/m4/stdbool.m4 $(top_srcdir)/m4/stddef_h.m4 \ + $(top_srcdir)/m4/stdint.m4 $(top_srcdir)/m4/stdlib_h.m4 \ + $(top_srcdir)/m4/strcase.m4 $(top_srcdir)/m4/strings_h.m4 \ + $(top_srcdir)/m4/sys_file_h.m4 $(top_srcdir)/m4/sys_types_h.m4 \ + $(top_srcdir)/m4/unistd_h.m4 $(top_srcdir)/m4/warn-on-use.m4 \ + $(top_srcdir)/m4/wchar_h.m4 $(top_srcdir)/m4/wchar_t.m4 \ + $(top_srcdir)/m4/wcrtomb.m4 $(top_srcdir)/m4/wctype_h.m4 \ + $(top_srcdir)/m4/wint_t.m4 $(top_srcdir)/configure.ac +am__configure_deps = $(am__aclocal_m4_deps) $(CONFIGURE_DEPENDENCIES) \ + $(ACLOCAL_M4) +mkinstalldirs = $(SHELL) $(top_srcdir)/mkinstalldirs +CONFIG_HEADER = $(top_builddir)/config.h +CONFIG_CLEAN_FILES = +CONFIG_CLEAN_VPATH_FILES = +LIBRARIES = $(noinst_LIBRARIES) +libgnu_a_AR = $(AR) $(ARFLAGS) +am__DEPENDENCIES_1 = +am_libgnu_a_OBJECTS = localcharset.$(OBJEXT) +libgnu_a_OBJECTS = $(am_libgnu_a_OBJECTS) +LTLIBRARIES = $(noinst_LTLIBRARIES) +DEFAULT_INCLUDES = -I.@am__isrc@ -I$(top_builddir) +depcomp = $(SHELL) $(top_srcdir)/depcomp +am__depfiles_maybe = depfiles +am__mv = mv -f +COMPILE = $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) \ + $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) +CCLD = $(CC) +LINK = $(CCLD) $(AM_CFLAGS) $(CFLAGS) $(AM_LDFLAGS) $(LDFLAGS) -o $@ +SOURCES = $(libgnu_a_SOURCES) $(EXTRA_libgnu_a_SOURCES) +DIST_SOURCES = $(libgnu_a_SOURCES) $(EXTRA_libgnu_a_SOURCES) +RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \ + html-recursive info-recursive install-data-recursive \ + install-dvi-recursive install-exec-recursive \ + install-html-recursive install-info-recursive \ + install-pdf-recursive install-ps-recursive install-recursive \ + installcheck-recursive installdirs-recursive pdf-recursive \ + ps-recursive uninstall-recursive +am__can_run_installinfo = \ + case $$AM_UPDATE_INFO_DIR in \ + n|no|NO) false;; \ + *) (install-info --version) >/dev/null 2>&1;; \ + esac +HEADERS = $(noinst_HEADERS) +RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive \ + distclean-recursive maintainer-clean-recursive +AM_RECURSIVE_TARGETS = $(RECURSIVE_TARGETS:-recursive=) \ + $(RECURSIVE_CLEAN_TARGETS:-recursive=) tags TAGS ctags CTAGS \ + distdir +ETAGS = etags +CTAGS = ctags +DIST_SUBDIRS = $(SUBDIRS) +DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST) +am__relativize = \ + dir0=`pwd`; \ + sed_first='s,^\([^/]*\)/.*$$,\1,'; \ + sed_rest='s,^[^/]*/*,,'; \ + sed_last='s,^.*/\([^/]*\)$$,\1,'; \ + sed_butlast='s,/*[^/]*$$,,'; \ + while test -n "$$dir1"; do \ + first=`echo "$$dir1" | sed -e "$$sed_first"`; \ + if test "$$first" != "."; then \ + if test "$$first" = ".."; then \ + dir2=`echo "$$dir0" | sed -e "$$sed_last"`/"$$dir2"; \ + dir0=`echo "$$dir0" | sed -e "$$sed_butlast"`; \ + else \ + first2=`echo "$$dir2" | sed -e "$$sed_first"`; \ + if test "$$first2" = "$$first"; then \ + dir2=`echo "$$dir2" | sed -e "$$sed_rest"`; \ + else \ + dir2="../$$dir2"; \ + fi; \ + dir0="$$dir0"/"$$first"; \ + fi; \ + fi; \ + dir1=`echo "$$dir1" | sed -e "$$sed_rest"`; \ + done; \ + reldir="$$dir2" +pkglibexecdir = @pkglibexecdir@ +ACLOCAL = @ACLOCAL@ +ALLOCA = @ALLOCA@ +ALLOCA_H = @ALLOCA_H@ +AMTAR = @AMTAR@ +APPLE_UNIVERSAL_BUILD = @APPLE_UNIVERSAL_BUILD@ +AR = @AR@ +ARFLAGS = @ARFLAGS@ +AUTOCONF = @AUTOCONF@ +AUTOHEADER = @AUTOHEADER@ +AUTOMAKE = @AUTOMAKE@ +AWK = @AWK@ +BITSIZEOF_PTRDIFF_T = @BITSIZEOF_PTRDIFF_T@ +BITSIZEOF_SIG_ATOMIC_T = @BITSIZEOF_SIG_ATOMIC_T@ +BITSIZEOF_SIZE_T = @BITSIZEOF_SIZE_T@ +BITSIZEOF_WCHAR_T = @BITSIZEOF_WCHAR_T@ +BITSIZEOF_WINT_T = @BITSIZEOF_WINT_T@ +CC = @CC@ +CCDEPMODE = @CCDEPMODE@ +CFLAGS = @CFLAGS@ +CPP = @CPP@ +CPPFLAGS = @CPPFLAGS@ +CXX = @CXX@ +CXXDEPMODE = @CXXDEPMODE@ +CXXFLAGS = @CXXFLAGS@ +CYGPATH_W = @CYGPATH_W@ +DEFS = @DEFS@ +DEPDIR = @DEPDIR@ +ECHO_C = @ECHO_C@ +ECHO_N = @ECHO_N@ +ECHO_T = @ECHO_T@ +EGREP = @EGREP@ +EXEEXT = @EXEEXT@ +GLIBC21 = @GLIBC21@ +GNULIB_ATOLL = @GNULIB_ATOLL@ +GNULIB_BTOWC = @GNULIB_BTOWC@ +GNULIB_CALLOC_POSIX = @GNULIB_CALLOC_POSIX@ +GNULIB_CANONICALIZE_FILE_NAME = @GNULIB_CANONICALIZE_FILE_NAME@ +GNULIB_CHDIR = @GNULIB_CHDIR@ +GNULIB_CHOWN = @GNULIB_CHOWN@ +GNULIB_CLOSE = @GNULIB_CLOSE@ +GNULIB_DUP = @GNULIB_DUP@ +GNULIB_DUP2 = @GNULIB_DUP2@ +GNULIB_DUP3 = @GNULIB_DUP3@ +GNULIB_ENVIRON = @GNULIB_ENVIRON@ +GNULIB_EUIDACCESS = @GNULIB_EUIDACCESS@ +GNULIB_FACCESSAT = @GNULIB_FACCESSAT@ +GNULIB_FCHDIR = @GNULIB_FCHDIR@ +GNULIB_FCHOWNAT = @GNULIB_FCHOWNAT@ +GNULIB_FDATASYNC = @GNULIB_FDATASYNC@ +GNULIB_FFS = @GNULIB_FFS@ +GNULIB_FLOCK = @GNULIB_FLOCK@ +GNULIB_FSYNC = @GNULIB_FSYNC@ +GNULIB_FTRUNCATE = @GNULIB_FTRUNCATE@ +GNULIB_GETCWD = @GNULIB_GETCWD@ +GNULIB_GETDOMAINNAME = @GNULIB_GETDOMAINNAME@ +GNULIB_GETDTABLESIZE = @GNULIB_GETDTABLESIZE@ +GNULIB_GETGROUPS = @GNULIB_GETGROUPS@ +GNULIB_GETHOSTNAME = @GNULIB_GETHOSTNAME@ +GNULIB_GETLOADAVG = @GNULIB_GETLOADAVG@ +GNULIB_GETLOGIN = @GNULIB_GETLOGIN@ +GNULIB_GETLOGIN_R = @GNULIB_GETLOGIN_R@ +GNULIB_GETPAGESIZE = @GNULIB_GETPAGESIZE@ +GNULIB_GETSUBOPT = @GNULIB_GETSUBOPT@ +GNULIB_GETUSERSHELL = @GNULIB_GETUSERSHELL@ +GNULIB_GRANTPT = @GNULIB_GRANTPT@ +GNULIB_GROUP_MEMBER = @GNULIB_GROUP_MEMBER@ +GNULIB_ISATTY = @GNULIB_ISATTY@ +GNULIB_ISWBLANK = @GNULIB_ISWBLANK@ +GNULIB_ISWCTYPE = @GNULIB_ISWCTYPE@ +GNULIB_LCHOWN = @GNULIB_LCHOWN@ +GNULIB_LINK = @GNULIB_LINK@ +GNULIB_LINKAT = @GNULIB_LINKAT@ +GNULIB_LSEEK = @GNULIB_LSEEK@ +GNULIB_MALLOC_POSIX = @GNULIB_MALLOC_POSIX@ +GNULIB_MBRLEN = @GNULIB_MBRLEN@ +GNULIB_MBRTOWC = @GNULIB_MBRTOWC@ +GNULIB_MBSINIT = @GNULIB_MBSINIT@ +GNULIB_MBSNRTOWCS = @GNULIB_MBSNRTOWCS@ +GNULIB_MBSRTOWCS = @GNULIB_MBSRTOWCS@ +GNULIB_MBTOWC = @GNULIB_MBTOWC@ +GNULIB_MKDTEMP = @GNULIB_MKDTEMP@ +GNULIB_MKOSTEMP = @GNULIB_MKOSTEMP@ +GNULIB_MKOSTEMPS = @GNULIB_MKOSTEMPS@ +GNULIB_MKSTEMP = @GNULIB_MKSTEMP@ +GNULIB_MKSTEMPS = @GNULIB_MKSTEMPS@ +GNULIB_NL_LANGINFO = @GNULIB_NL_LANGINFO@ +GNULIB_PIPE = @GNULIB_PIPE@ +GNULIB_PIPE2 = @GNULIB_PIPE2@ +GNULIB_POSIX_OPENPT = @GNULIB_POSIX_OPENPT@ +GNULIB_PREAD = @GNULIB_PREAD@ +GNULIB_PTSNAME = @GNULIB_PTSNAME@ +GNULIB_PTSNAME_R = @GNULIB_PTSNAME_R@ +GNULIB_PUTENV = @GNULIB_PUTENV@ +GNULIB_PWRITE = @GNULIB_PWRITE@ +GNULIB_RANDOM = @GNULIB_RANDOM@ +GNULIB_RANDOM_R = @GNULIB_RANDOM_R@ +GNULIB_READ = @GNULIB_READ@ +GNULIB_READLINK = @GNULIB_READLINK@ +GNULIB_READLINKAT = @GNULIB_READLINKAT@ +GNULIB_REALLOC_POSIX = @GNULIB_REALLOC_POSIX@ +GNULIB_REALPATH = @GNULIB_REALPATH@ +GNULIB_RMDIR = @GNULIB_RMDIR@ +GNULIB_RPMATCH = @GNULIB_RPMATCH@ +GNULIB_SETENV = @GNULIB_SETENV@ +GNULIB_SETHOSTNAME = @GNULIB_SETHOSTNAME@ +GNULIB_SLEEP = @GNULIB_SLEEP@ +GNULIB_STRTOD = @GNULIB_STRTOD@ +GNULIB_STRTOLL = @GNULIB_STRTOLL@ +GNULIB_STRTOULL = @GNULIB_STRTOULL@ +GNULIB_SYMLINK = @GNULIB_SYMLINK@ +GNULIB_SYMLINKAT = @GNULIB_SYMLINKAT@ +GNULIB_SYSTEM_POSIX = @GNULIB_SYSTEM_POSIX@ +GNULIB_TOWCTRANS = @GNULIB_TOWCTRANS@ +GNULIB_TTYNAME_R = @GNULIB_TTYNAME_R@ +GNULIB_UNISTD_H_NONBLOCKING = @GNULIB_UNISTD_H_NONBLOCKING@ +GNULIB_UNISTD_H_SIGPIPE = @GNULIB_UNISTD_H_SIGPIPE@ +GNULIB_UNLINK = @GNULIB_UNLINK@ +GNULIB_UNLINKAT = @GNULIB_UNLINKAT@ +GNULIB_UNLOCKPT = @GNULIB_UNLOCKPT@ +GNULIB_UNSETENV = @GNULIB_UNSETENV@ +GNULIB_USLEEP = @GNULIB_USLEEP@ +GNULIB_WCPCPY = @GNULIB_WCPCPY@ +GNULIB_WCPNCPY = @GNULIB_WCPNCPY@ +GNULIB_WCRTOMB = @GNULIB_WCRTOMB@ +GNULIB_WCSCASECMP = @GNULIB_WCSCASECMP@ +GNULIB_WCSCAT = @GNULIB_WCSCAT@ +GNULIB_WCSCHR = @GNULIB_WCSCHR@ +GNULIB_WCSCMP = @GNULIB_WCSCMP@ +GNULIB_WCSCOLL = @GNULIB_WCSCOLL@ +GNULIB_WCSCPY = @GNULIB_WCSCPY@ +GNULIB_WCSCSPN = @GNULIB_WCSCSPN@ +GNULIB_WCSDUP = @GNULIB_WCSDUP@ +GNULIB_WCSLEN = @GNULIB_WCSLEN@ +GNULIB_WCSNCASECMP = @GNULIB_WCSNCASECMP@ +GNULIB_WCSNCAT = @GNULIB_WCSNCAT@ +GNULIB_WCSNCMP = @GNULIB_WCSNCMP@ +GNULIB_WCSNCPY = @GNULIB_WCSNCPY@ +GNULIB_WCSNLEN = @GNULIB_WCSNLEN@ +GNULIB_WCSNRTOMBS = @GNULIB_WCSNRTOMBS@ +GNULIB_WCSPBRK = @GNULIB_WCSPBRK@ +GNULIB_WCSRCHR = @GNULIB_WCSRCHR@ +GNULIB_WCSRTOMBS = @GNULIB_WCSRTOMBS@ +GNULIB_WCSSPN = @GNULIB_WCSSPN@ +GNULIB_WCSSTR = @GNULIB_WCSSTR@ +GNULIB_WCSTOK = @GNULIB_WCSTOK@ +GNULIB_WCSWIDTH = @GNULIB_WCSWIDTH@ +GNULIB_WCSXFRM = @GNULIB_WCSXFRM@ +GNULIB_WCTOB = @GNULIB_WCTOB@ +GNULIB_WCTOMB = @GNULIB_WCTOMB@ +GNULIB_WCTRANS = @GNULIB_WCTRANS@ +GNULIB_WCTYPE = @GNULIB_WCTYPE@ +GNULIB_WCWIDTH = @GNULIB_WCWIDTH@ +GNULIB_WMEMCHR = @GNULIB_WMEMCHR@ +GNULIB_WMEMCMP = @GNULIB_WMEMCMP@ +GNULIB_WMEMCPY = @GNULIB_WMEMCPY@ +GNULIB_WMEMMOVE = @GNULIB_WMEMMOVE@ +GNULIB_WMEMSET = @GNULIB_WMEMSET@ +GNULIB_WRITE = @GNULIB_WRITE@ +GNULIB__EXIT = @GNULIB__EXIT@ +GREP = @GREP@ +HAVE_ATOLL = @HAVE_ATOLL@ +HAVE_BTOWC = @HAVE_BTOWC@ +HAVE_CANONICALIZE_FILE_NAME = @HAVE_CANONICALIZE_FILE_NAME@ +HAVE_CHOWN = @HAVE_CHOWN@ +HAVE_DECL_ENVIRON = @HAVE_DECL_ENVIRON@ +HAVE_DECL_FCHDIR = @HAVE_DECL_FCHDIR@ +HAVE_DECL_FDATASYNC = @HAVE_DECL_FDATASYNC@ +HAVE_DECL_GETDOMAINNAME = @HAVE_DECL_GETDOMAINNAME@ +HAVE_DECL_GETLOADAVG = @HAVE_DECL_GETLOADAVG@ +HAVE_DECL_GETLOGIN_R = @HAVE_DECL_GETLOGIN_R@ +HAVE_DECL_GETPAGESIZE = @HAVE_DECL_GETPAGESIZE@ +HAVE_DECL_GETUSERSHELL = @HAVE_DECL_GETUSERSHELL@ +HAVE_DECL_SETENV = @HAVE_DECL_SETENV@ +HAVE_DECL_SETHOSTNAME = @HAVE_DECL_SETHOSTNAME@ +HAVE_DECL_STRNCASECMP = @HAVE_DECL_STRNCASECMP@ +HAVE_DECL_TTYNAME_R = @HAVE_DECL_TTYNAME_R@ +HAVE_DECL_UNSETENV = @HAVE_DECL_UNSETENV@ +HAVE_DECL_WCTOB = @HAVE_DECL_WCTOB@ +HAVE_DECL_WCWIDTH = @HAVE_DECL_WCWIDTH@ +HAVE_DUP2 = @HAVE_DUP2@ +HAVE_DUP3 = @HAVE_DUP3@ +HAVE_EUIDACCESS = @HAVE_EUIDACCESS@ +HAVE_FACCESSAT = @HAVE_FACCESSAT@ +HAVE_FCHDIR = @HAVE_FCHDIR@ +HAVE_FCHOWNAT = @HAVE_FCHOWNAT@ +HAVE_FDATASYNC = @HAVE_FDATASYNC@ +HAVE_FEATURES_H = @HAVE_FEATURES_H@ +HAVE_FFS = @HAVE_FFS@ +HAVE_FLOCK = @HAVE_FLOCK@ +HAVE_FSYNC = @HAVE_FSYNC@ +HAVE_FTRUNCATE = @HAVE_FTRUNCATE@ +HAVE_GETDTABLESIZE = @HAVE_GETDTABLESIZE@ +HAVE_GETGROUPS = @HAVE_GETGROUPS@ +HAVE_GETHOSTNAME = @HAVE_GETHOSTNAME@ +HAVE_GETLOGIN = @HAVE_GETLOGIN@ +HAVE_GETPAGESIZE = @HAVE_GETPAGESIZE@ +HAVE_GETSUBOPT = @HAVE_GETSUBOPT@ +HAVE_GRANTPT = @HAVE_GRANTPT@ +HAVE_GROUP_MEMBER = @HAVE_GROUP_MEMBER@ +HAVE_INTTYPES_H = @HAVE_INTTYPES_H@ +HAVE_ISWBLANK = @HAVE_ISWBLANK@ +HAVE_ISWCNTRL = @HAVE_ISWCNTRL@ +HAVE_LANGINFO_CODESET = @HAVE_LANGINFO_CODESET@ +HAVE_LANGINFO_ERA = @HAVE_LANGINFO_ERA@ +HAVE_LANGINFO_H = @HAVE_LANGINFO_H@ +HAVE_LANGINFO_T_FMT_AMPM = @HAVE_LANGINFO_T_FMT_AMPM@ +HAVE_LANGINFO_YESEXPR = @HAVE_LANGINFO_YESEXPR@ +HAVE_LCHOWN = @HAVE_LCHOWN@ +HAVE_LINK = @HAVE_LINK@ +HAVE_LINKAT = @HAVE_LINKAT@ +HAVE_LONG_LONG_INT = @HAVE_LONG_LONG_INT@ +HAVE_MBRLEN = @HAVE_MBRLEN@ +HAVE_MBRTOWC = @HAVE_MBRTOWC@ +HAVE_MBSINIT = @HAVE_MBSINIT@ +HAVE_MBSNRTOWCS = @HAVE_MBSNRTOWCS@ +HAVE_MBSRTOWCS = @HAVE_MBSRTOWCS@ +HAVE_MKDTEMP = @HAVE_MKDTEMP@ +HAVE_MKOSTEMP = @HAVE_MKOSTEMP@ +HAVE_MKOSTEMPS = @HAVE_MKOSTEMPS@ +HAVE_MKSTEMP = @HAVE_MKSTEMP@ +HAVE_MKSTEMPS = @HAVE_MKSTEMPS@ +HAVE_MSVC_INVALID_PARAMETER_HANDLER = @HAVE_MSVC_INVALID_PARAMETER_HANDLER@ +HAVE_NL_LANGINFO = @HAVE_NL_LANGINFO@ +HAVE_OS_H = @HAVE_OS_H@ +HAVE_PIPE = @HAVE_PIPE@ +HAVE_PIPE2 = @HAVE_PIPE2@ +HAVE_POSIX_OPENPT = @HAVE_POSIX_OPENPT@ +HAVE_PREAD = @HAVE_PREAD@ +HAVE_PTSNAME = @HAVE_PTSNAME@ +HAVE_PTSNAME_R = @HAVE_PTSNAME_R@ +HAVE_PWRITE = @HAVE_PWRITE@ +HAVE_RANDOM = @HAVE_RANDOM@ +HAVE_RANDOM_H = @HAVE_RANDOM_H@ +HAVE_RANDOM_R = @HAVE_RANDOM_R@ +HAVE_READLINK = @HAVE_READLINK@ +HAVE_READLINKAT = @HAVE_READLINKAT@ +HAVE_REALPATH = @HAVE_REALPATH@ +HAVE_RPMATCH = @HAVE_RPMATCH@ +HAVE_SETENV = @HAVE_SETENV@ +HAVE_SETHOSTNAME = @HAVE_SETHOSTNAME@ +HAVE_SIGNED_SIG_ATOMIC_T = @HAVE_SIGNED_SIG_ATOMIC_T@ +HAVE_SIGNED_WCHAR_T = @HAVE_SIGNED_WCHAR_T@ +HAVE_SIGNED_WINT_T = @HAVE_SIGNED_WINT_T@ +HAVE_SLEEP = @HAVE_SLEEP@ +HAVE_STDINT_H = @HAVE_STDINT_H@ +HAVE_STRCASECMP = @HAVE_STRCASECMP@ +HAVE_STRINGS_H = @HAVE_STRINGS_H@ +HAVE_STRTOD = @HAVE_STRTOD@ +HAVE_STRTOLL = @HAVE_STRTOLL@ +HAVE_STRTOULL = @HAVE_STRTOULL@ +HAVE_STRUCT_RANDOM_DATA = @HAVE_STRUCT_RANDOM_DATA@ +HAVE_SYMLINK = @HAVE_SYMLINK@ +HAVE_SYMLINKAT = @HAVE_SYMLINKAT@ +HAVE_SYS_BITYPES_H = @HAVE_SYS_BITYPES_H@ +HAVE_SYS_FILE_H = @HAVE_SYS_FILE_H@ +HAVE_SYS_INTTYPES_H = @HAVE_SYS_INTTYPES_H@ +HAVE_SYS_LOADAVG_H = @HAVE_SYS_LOADAVG_H@ +HAVE_SYS_PARAM_H = @HAVE_SYS_PARAM_H@ +HAVE_SYS_TYPES_H = @HAVE_SYS_TYPES_H@ +HAVE_UNISTD_H = @HAVE_UNISTD_H@ +HAVE_UNLINKAT = @HAVE_UNLINKAT@ +HAVE_UNLOCKPT = @HAVE_UNLOCKPT@ +HAVE_UNSIGNED_LONG_LONG_INT = @HAVE_UNSIGNED_LONG_LONG_INT@ +HAVE_USLEEP = @HAVE_USLEEP@ +HAVE_WCHAR_H = @HAVE_WCHAR_H@ +HAVE_WCHAR_T = @HAVE_WCHAR_T@ +HAVE_WCPCPY = @HAVE_WCPCPY@ +HAVE_WCPNCPY = @HAVE_WCPNCPY@ +HAVE_WCRTOMB = @HAVE_WCRTOMB@ +HAVE_WCSCASECMP = @HAVE_WCSCASECMP@ +HAVE_WCSCAT = @HAVE_WCSCAT@ +HAVE_WCSCHR = @HAVE_WCSCHR@ +HAVE_WCSCMP = @HAVE_WCSCMP@ +HAVE_WCSCOLL = @HAVE_WCSCOLL@ +HAVE_WCSCPY = @HAVE_WCSCPY@ +HAVE_WCSCSPN = @HAVE_WCSCSPN@ +HAVE_WCSDUP = @HAVE_WCSDUP@ +HAVE_WCSLEN = @HAVE_WCSLEN@ +HAVE_WCSNCASECMP = @HAVE_WCSNCASECMP@ +HAVE_WCSNCAT = @HAVE_WCSNCAT@ +HAVE_WCSNCMP = @HAVE_WCSNCMP@ +HAVE_WCSNCPY = @HAVE_WCSNCPY@ +HAVE_WCSNLEN = @HAVE_WCSNLEN@ +HAVE_WCSNRTOMBS = @HAVE_WCSNRTOMBS@ +HAVE_WCSPBRK = @HAVE_WCSPBRK@ +HAVE_WCSRCHR = @HAVE_WCSRCHR@ +HAVE_WCSRTOMBS = @HAVE_WCSRTOMBS@ +HAVE_WCSSPN = @HAVE_WCSSPN@ +HAVE_WCSSTR = @HAVE_WCSSTR@ +HAVE_WCSTOK = @HAVE_WCSTOK@ +HAVE_WCSWIDTH = @HAVE_WCSWIDTH@ +HAVE_WCSXFRM = @HAVE_WCSXFRM@ +HAVE_WCTRANS_T = @HAVE_WCTRANS_T@ +HAVE_WCTYPE_H = @HAVE_WCTYPE_H@ +HAVE_WCTYPE_T = @HAVE_WCTYPE_T@ +HAVE_WINT_T = @HAVE_WINT_T@ +HAVE_WMEMCHR = @HAVE_WMEMCHR@ +HAVE_WMEMCMP = @HAVE_WMEMCMP@ +HAVE_WMEMCPY = @HAVE_WMEMCPY@ +HAVE_WMEMMOVE = @HAVE_WMEMMOVE@ +HAVE_WMEMSET = @HAVE_WMEMSET@ +HAVE__BOOL = @HAVE__BOOL@ +HAVE__EXIT = @HAVE__EXIT@ +INCLUDE_NEXT = @INCLUDE_NEXT@ +INCLUDE_NEXT_AS_FIRST_DIRECTIVE = @INCLUDE_NEXT_AS_FIRST_DIRECTIVE@ +INSTALL = @INSTALL@ +INSTALL_DATA = @INSTALL_DATA@ +INSTALL_PROGRAM = @INSTALL_PROGRAM@ +INSTALL_SCRIPT = @INSTALL_SCRIPT@ +INSTALL_STRIP_PROGRAM = @INSTALL_STRIP_PROGRAM@ +LDFLAGS = @LDFLAGS@ +LIBGNU_LIBDEPS = @LIBGNU_LIBDEPS@ +LIBGNU_LTLIBDEPS = @LIBGNU_LTLIBDEPS@ +LIBICONV = @LIBICONV@ +LIBINTL = @LIBINTL@ +LIBOBJS = @LIBOBJS@ +LIBS = @LIBS@ +LOCALCHARSET_TESTS_ENVIRONMENT = @LOCALCHARSET_TESTS_ENVIRONMENT@ +LOCALE_FR = @LOCALE_FR@ +LOCALE_FR_UTF8 = @LOCALE_FR_UTF8@ +LOCALE_JA = @LOCALE_JA@ +LOCALE_ZH_CN = @LOCALE_ZH_CN@ +LTLIBICONV = @LTLIBICONV@ +LTLIBINTL = @LTLIBINTL@ +LTLIBOBJS = @LTLIBOBJS@ +MAINT = @MAINT@ +MAKEINFO = @MAKEINFO@ +MKDIR_P = @MKDIR_P@ +MO_PROGRAMS = @MO_PROGRAMS@ +MO_SECTIONS = @MO_SECTIONS@ +MO_SUTOROOT = @MO_SUTOROOT@ +NEXT_AS_FIRST_DIRECTIVE_LANGINFO_H = @NEXT_AS_FIRST_DIRECTIVE_LANGINFO_H@ +NEXT_AS_FIRST_DIRECTIVE_STDDEF_H = @NEXT_AS_FIRST_DIRECTIVE_STDDEF_H@ +NEXT_AS_FIRST_DIRECTIVE_STDINT_H = @NEXT_AS_FIRST_DIRECTIVE_STDINT_H@ +NEXT_AS_FIRST_DIRECTIVE_STDLIB_H = @NEXT_AS_FIRST_DIRECTIVE_STDLIB_H@ +NEXT_AS_FIRST_DIRECTIVE_STRINGS_H = @NEXT_AS_FIRST_DIRECTIVE_STRINGS_H@ +NEXT_AS_FIRST_DIRECTIVE_SYS_FILE_H = @NEXT_AS_FIRST_DIRECTIVE_SYS_FILE_H@ +NEXT_AS_FIRST_DIRECTIVE_SYS_TYPES_H = @NEXT_AS_FIRST_DIRECTIVE_SYS_TYPES_H@ +NEXT_AS_FIRST_DIRECTIVE_UNISTD_H = @NEXT_AS_FIRST_DIRECTIVE_UNISTD_H@ +NEXT_AS_FIRST_DIRECTIVE_WCHAR_H = @NEXT_AS_FIRST_DIRECTIVE_WCHAR_H@ +NEXT_AS_FIRST_DIRECTIVE_WCTYPE_H = @NEXT_AS_FIRST_DIRECTIVE_WCTYPE_H@ +NEXT_LANGINFO_H = @NEXT_LANGINFO_H@ +NEXT_STDDEF_H = @NEXT_STDDEF_H@ +NEXT_STDINT_H = @NEXT_STDINT_H@ +NEXT_STDLIB_H = @NEXT_STDLIB_H@ +NEXT_STRINGS_H = @NEXT_STRINGS_H@ +NEXT_SYS_FILE_H = @NEXT_SYS_FILE_H@ +NEXT_SYS_TYPES_H = @NEXT_SYS_TYPES_H@ +NEXT_UNISTD_H = @NEXT_UNISTD_H@ +NEXT_WCHAR_H = @NEXT_WCHAR_H@ +NEXT_WCTYPE_H = @NEXT_WCTYPE_H@ +OBJEXT = @OBJEXT@ +PACKAGE = @PACKAGE@ +PACKAGE_BUGREPORT = @PACKAGE_BUGREPORT@ +PACKAGE_NAME = @PACKAGE_NAME@ +PACKAGE_STRING = @PACKAGE_STRING@ +PACKAGE_TARNAME = @PACKAGE_TARNAME@ +PACKAGE_URL = @PACKAGE_URL@ +PACKAGE_VERSION = @PACKAGE_VERSION@ +PATH_SEPARATOR = @PATH_SEPARATOR@ +PO_PROGRAMS = @PO_PROGRAMS@ +PO_SECTIONS = @PO_SECTIONS@ +PO_SUTOROOT = @PO_SUTOROOT@ +PRAGMA_COLUMNS = @PRAGMA_COLUMNS@ +PRAGMA_SYSTEM_HEADER = @PRAGMA_SYSTEM_HEADER@ +PTRDIFF_T_SUFFIX = @PTRDIFF_T_SUFFIX@ +RANLIB = @RANLIB@ +REPLACE_BTOWC = @REPLACE_BTOWC@ +REPLACE_CALLOC = @REPLACE_CALLOC@ +REPLACE_CANONICALIZE_FILE_NAME = @REPLACE_CANONICALIZE_FILE_NAME@ +REPLACE_CHOWN = @REPLACE_CHOWN@ +REPLACE_CLOSE = @REPLACE_CLOSE@ +REPLACE_DUP = @REPLACE_DUP@ +REPLACE_DUP2 = @REPLACE_DUP2@ +REPLACE_FCHOWNAT = @REPLACE_FCHOWNAT@ +REPLACE_GETCWD = @REPLACE_GETCWD@ +REPLACE_GETDOMAINNAME = @REPLACE_GETDOMAINNAME@ +REPLACE_GETGROUPS = @REPLACE_GETGROUPS@ +REPLACE_GETLOGIN_R = @REPLACE_GETLOGIN_R@ +REPLACE_GETPAGESIZE = @REPLACE_GETPAGESIZE@ +REPLACE_ISATTY = @REPLACE_ISATTY@ +REPLACE_ISWBLANK = @REPLACE_ISWBLANK@ +REPLACE_ISWCNTRL = @REPLACE_ISWCNTRL@ +REPLACE_LCHOWN = @REPLACE_LCHOWN@ +REPLACE_LINK = @REPLACE_LINK@ +REPLACE_LINKAT = @REPLACE_LINKAT@ +REPLACE_LSEEK = @REPLACE_LSEEK@ +REPLACE_MALLOC = @REPLACE_MALLOC@ +REPLACE_MBRLEN = @REPLACE_MBRLEN@ +REPLACE_MBRTOWC = @REPLACE_MBRTOWC@ +REPLACE_MBSINIT = @REPLACE_MBSINIT@ +REPLACE_MBSNRTOWCS = @REPLACE_MBSNRTOWCS@ +REPLACE_MBSRTOWCS = @REPLACE_MBSRTOWCS@ +REPLACE_MBSTATE_T = @REPLACE_MBSTATE_T@ +REPLACE_MBTOWC = @REPLACE_MBTOWC@ +REPLACE_MKSTEMP = @REPLACE_MKSTEMP@ +REPLACE_NL_LANGINFO = @REPLACE_NL_LANGINFO@ +REPLACE_NULL = @REPLACE_NULL@ +REPLACE_PREAD = @REPLACE_PREAD@ +REPLACE_PTSNAME_R = @REPLACE_PTSNAME_R@ +REPLACE_PUTENV = @REPLACE_PUTENV@ +REPLACE_PWRITE = @REPLACE_PWRITE@ +REPLACE_RANDOM_R = @REPLACE_RANDOM_R@ +REPLACE_READ = @REPLACE_READ@ +REPLACE_READLINK = @REPLACE_READLINK@ +REPLACE_REALLOC = @REPLACE_REALLOC@ +REPLACE_REALPATH = @REPLACE_REALPATH@ +REPLACE_RMDIR = @REPLACE_RMDIR@ +REPLACE_SETENV = @REPLACE_SETENV@ +REPLACE_SLEEP = @REPLACE_SLEEP@ +REPLACE_STRTOD = @REPLACE_STRTOD@ +REPLACE_SYMLINK = @REPLACE_SYMLINK@ +REPLACE_TOWLOWER = @REPLACE_TOWLOWER@ +REPLACE_TTYNAME_R = @REPLACE_TTYNAME_R@ +REPLACE_UNLINK = @REPLACE_UNLINK@ +REPLACE_UNLINKAT = @REPLACE_UNLINKAT@ +REPLACE_UNSETENV = @REPLACE_UNSETENV@ +REPLACE_USLEEP = @REPLACE_USLEEP@ +REPLACE_WCRTOMB = @REPLACE_WCRTOMB@ +REPLACE_WCSNRTOMBS = @REPLACE_WCSNRTOMBS@ +REPLACE_WCSRTOMBS = @REPLACE_WCSRTOMBS@ +REPLACE_WCSWIDTH = @REPLACE_WCSWIDTH@ +REPLACE_WCTOB = @REPLACE_WCTOB@ +REPLACE_WCTOMB = @REPLACE_WCTOMB@ +REPLACE_WCWIDTH = @REPLACE_WCWIDTH@ +REPLACE_WRITE = @REPLACE_WRITE@ +SET_MAKE = @SET_MAKE@ +SHELL = @SHELL@ +SIG_ATOMIC_T_SUFFIX = @SIG_ATOMIC_T_SUFFIX@ +SIZE_T_SUFFIX = @SIZE_T_SUFFIX@ +STDBOOL_H = @STDBOOL_H@ +STDDEF_H = @STDDEF_H@ +STDINT_H = @STDINT_H@ +STRIP = @STRIP@ +UNISTD_H_HAVE_WINSOCK2_H = @UNISTD_H_HAVE_WINSOCK2_H@ +UNISTD_H_HAVE_WINSOCK2_H_AND_USE_SOCKETS = @UNISTD_H_HAVE_WINSOCK2_H_AND_USE_SOCKETS@ +VERSION = @VERSION@ +WCHAR_T_SUFFIX = @WCHAR_T_SUFFIX@ +WINT_T_SUFFIX = @WINT_T_SUFFIX@ +abs_builddir = @abs_builddir@ +abs_srcdir = @abs_srcdir@ +abs_top_builddir = @abs_top_builddir@ +abs_top_srcdir = @abs_top_srcdir@ +ac_ct_CC = @ac_ct_CC@ +ac_ct_CXX = @ac_ct_CXX@ +am__include = @am__include@ +am__leading_dot = @am__leading_dot@ +am__quote = @am__quote@ +am__tar = @am__tar@ +am__untar = @am__untar@ +bindir = @bindir@ +build = @build@ +build_alias = @build_alias@ +build_cpu = @build_cpu@ +build_os = @build_os@ +build_vendor = @build_vendor@ +builddir = @builddir@ +datadir = @datadir@ +datarootdir = @datarootdir@ +docdir = @docdir@ +dvidir = @dvidir@ +exec_prefix = @exec_prefix@ +gl_LIBOBJS = @gl_LIBOBJS@ +gl_LTLIBOBJS = @gl_LTLIBOBJS@ +gltests_LIBOBJS = @gltests_LIBOBJS@ +gltests_LTLIBOBJS = @gltests_LTLIBOBJS@ +gltests_WITNESS = @gltests_WITNESS@ +host = @host@ +host_alias = @host_alias@ +host_cpu = @host_cpu@ +host_os = @host_os@ +host_vendor = @host_vendor@ +htmldir = @htmldir@ +includedir = @includedir@ +infodir = @infodir@ +install_sh = @install_sh@ +libdir = @libdir@ +libexecdir = @libexecdir@ +lispdir = @lispdir@ +localedir = @localedir@ +localstatedir = @localstatedir@ +mandir = @mandir@ +mkdir_p = @mkdir_p@ +oldincludedir = @oldincludedir@ +pdfdir = @pdfdir@ +prefix = @prefix@ +program_transform_name = @program_transform_name@ +psdir = @psdir@ +sbindir = @sbindir@ +sharedstatedir = @sharedstatedir@ +srcdir = @srcdir@ +sysconfdir = @sysconfdir@ +target_alias = @target_alias@ +top_build_prefix = @top_build_prefix@ +top_builddir = @top_builddir@ +top_srcdir = @top_srcdir@ +AUTOMAKE_OPTIONS = 1.5 gnits +SUBDIRS = +noinst_HEADERS = +noinst_LIBRARIES = libgnu.a +noinst_LTLIBRARIES = +EXTRA_DIST = alloca.in.h btowc.c flock.c langinfo.in.h config.charset \ + ref-add.sin ref-del.sin malloc.c malloc.c mbrtowc.c mbsinit.c \ + mbtowc-impl.h mbtowc.c msvc-inval.c msvc-inval.h \ + msvc-nothrow.c msvc-nothrow.h nl_langinfo.c regcomp.c regex.c \ + regex.h regex_internal.c regex_internal.h regexec.c \ + $(top_srcdir)/./snippet/_Noreturn.h \ + $(top_srcdir)/./snippet/arg-nonnull.h \ + $(top_srcdir)/./snippet/c++defs.h \ + $(top_srcdir)/./snippet/warn-on-use.h stdbool.in.h stddef.in.h \ + stdint.in.h stdlib.in.h strcasecmp.c strncasecmp.c streq.h \ + strings.in.h sys_file.in.h sys_types.in.h unistd.in.h verify.h \ + wchar.in.h wcrtomb.c wctype.in.h + +# The BUILT_SOURCES created by this Makefile snippet are not used via #include +# statements but through direct file reference. Therefore this snippet must be +# present in all Makefile.am that need it. This is ensured by the applicability +# 'all' defined above. + +# The BUILT_SOURCES created by this Makefile snippet are not used via #include +# statements but through direct file reference. Therefore this snippet must be +# present in all Makefile.am that need it. This is ensured by the applicability +# 'all' defined above. +BUILT_SOURCES = $(ALLOCA_H) configmake.h langinfo.h arg-nonnull.h \ + c++defs.h warn-on-use.h $(STDBOOL_H) $(STDDEF_H) $(STDINT_H) \ + stdlib.h strings.h sys/file.h sys/types.h unistd.h wchar.h \ + wctype.h +SUFFIXES = .sed .sin +MOSTLYCLEANFILES = core *.stackdump alloca.h alloca.h-t langinfo.h \ + langinfo.h-t arg-nonnull.h arg-nonnull.h-t c++defs.h \ + c++defs.h-t warn-on-use.h warn-on-use.h-t stdbool.h \ + stdbool.h-t stddef.h stddef.h-t stdint.h stdint.h-t stdlib.h \ + stdlib.h-t strings.h strings.h-t sys/file.h sys/file.h-t \ + sys/types.h sys/types.h-t unistd.h unistd.h-t wchar.h \ + wchar.h-t wctype.h wctype.h-t +MOSTLYCLEANDIRS = sys +CLEANFILES = configmake.h configmake.h-t charset.alias ref-add.sed \ + ref-del.sed +DISTCLEANFILES = +MAINTAINERCLEANFILES = +AM_CPPFLAGS = +AM_CFLAGS = +libgnu_a_SOURCES = gettext.h localcharset.h localcharset.c +libgnu_a_LIBADD = $(gl_LIBOBJS) +libgnu_a_DEPENDENCIES = $(gl_LIBOBJS) +EXTRA_libgnu_a_SOURCES = btowc.c flock.c malloc.c malloc.c mbrtowc.c \ + mbsinit.c mbtowc.c msvc-inval.c msvc-nothrow.c nl_langinfo.c \ + regcomp.c regex.c regex_internal.c regexec.c strcasecmp.c \ + strncasecmp.c wcrtomb.c +charset_alias = $(DESTDIR)$(libdir)/charset.alias +charset_tmp = $(DESTDIR)$(libdir)/charset.tmp + +# Because this Makefile snippet defines a variable used by other +# gnulib Makefile snippets, it must be present in all Makefile.am that +# need it. This is ensured by the applicability 'all' defined above. +_NORETURN_H = $(top_srcdir)/./snippet/_Noreturn.h +ARG_NONNULL_H = arg-nonnull.h +CXXDEFS_H = c++defs.h +WARN_ON_USE_H = warn-on-use.h +all: $(BUILT_SOURCES) + $(MAKE) $(AM_MAKEFLAGS) all-recursive + +.SUFFIXES: +.SUFFIXES: .sed .sin .c .o .obj +$(srcdir)/Makefile.in: @MAINTAINER_MODE_TRUE@ $(srcdir)/Makefile.am $(am__configure_deps) + @for dep in $?; do \ + case '$(am__configure_deps)' in \ + *$$dep*) \ + ( cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh ) \ + && { if test -f $@; then exit 0; else break; fi; }; \ + exit 1;; \ + esac; \ + done; \ + echo ' cd $(top_srcdir) && $(AUTOMAKE) --gnits lib/Makefile'; \ + $(am__cd) $(top_srcdir) && \ + $(AUTOMAKE) --gnits lib/Makefile +.PRECIOUS: Makefile +Makefile: $(srcdir)/Makefile.in $(top_builddir)/config.status + @case '$?' in \ + *config.status*) \ + cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh;; \ + *) \ + echo ' cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@ $(am__depfiles_maybe)'; \ + cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@ $(am__depfiles_maybe);; \ + esac; + +$(top_builddir)/config.status: $(top_srcdir)/configure $(CONFIG_STATUS_DEPENDENCIES) + cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh + +$(top_srcdir)/configure: @MAINTAINER_MODE_TRUE@ $(am__configure_deps) + cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh +$(ACLOCAL_M4): @MAINTAINER_MODE_TRUE@ $(am__aclocal_m4_deps) + cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh +$(am__aclocal_m4_deps): + +clean-noinstLIBRARIES: + -test -z "$(noinst_LIBRARIES)" || rm -f $(noinst_LIBRARIES) +libgnu.a: $(libgnu_a_OBJECTS) $(libgnu_a_DEPENDENCIES) $(EXTRA_libgnu_a_DEPENDENCIES) + -rm -f libgnu.a + $(libgnu_a_AR) libgnu.a $(libgnu_a_OBJECTS) $(libgnu_a_LIBADD) + $(RANLIB) libgnu.a + +clean-noinstLTLIBRARIES: + -test -z "$(noinst_LTLIBRARIES)" || rm -f $(noinst_LTLIBRARIES) + @list='$(noinst_LTLIBRARIES)'; for p in $$list; do \ + dir="`echo $$p | sed -e 's|/[^/]*$$||'`"; \ + test "$$dir" != "$$p" || dir=.; \ + echo "rm -f \"$${dir}/so_locations\""; \ + rm -f "$${dir}/so_locations"; \ + done + +mostlyclean-compile: + -rm -f *.$(OBJEXT) + +distclean-compile: + -rm -f *.tab.c + +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/btowc.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/flock.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/localcharset.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/malloc.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/mbrtowc.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/mbsinit.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/mbtowc.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/msvc-inval.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/msvc-nothrow.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/nl_langinfo.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/regcomp.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/regex.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/regex_internal.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/regexec.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/strcasecmp.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/strncasecmp.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/wcrtomb.Po@am__quote@ + +.c.o: +@am__fastdepCC_TRUE@ $(COMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo -c -o $@ $< +@am__fastdepCC_TRUE@ $(am__mv) $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Po +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='$<' object='$@' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(COMPILE) -c $< + +.c.obj: +@am__fastdepCC_TRUE@ $(COMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo -c -o $@ `$(CYGPATH_W) '$<'` +@am__fastdepCC_TRUE@ $(am__mv) $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Po +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='$<' object='$@' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(COMPILE) -c `$(CYGPATH_W) '$<'` + +# This directory's subdirectories are mostly independent; you can cd +# into them and run `make' without going through this Makefile. +# To change the values of `make' variables: instead of editing Makefiles, +# (1) if the variable is set in `config.status', edit `config.status' +# (which will cause the Makefiles to be regenerated when you run `make'); +# (2) otherwise, pass the desired values on the `make' command line. +$(RECURSIVE_TARGETS): + @fail= failcom='exit 1'; \ + for f in x $$MAKEFLAGS; do \ + case $$f in \ + *=* | --[!k]*);; \ + *k*) failcom='fail=yes';; \ + esac; \ + done; \ + dot_seen=no; \ + target=`echo $@ | sed s/-recursive//`; \ + list='$(SUBDIRS)'; for subdir in $$list; do \ + echo "Making $$target in $$subdir"; \ + if test "$$subdir" = "."; then \ + dot_seen=yes; \ + local_target="$$target-am"; \ + else \ + local_target="$$target"; \ + fi; \ + ($(am__cd) $$subdir && $(MAKE) $(AM_MAKEFLAGS) $$local_target) \ + || eval $$failcom; \ + done; \ + if test "$$dot_seen" = "no"; then \ + $(MAKE) $(AM_MAKEFLAGS) "$$target-am" || exit 1; \ + fi; test -z "$$fail" + +$(RECURSIVE_CLEAN_TARGETS): + @fail= failcom='exit 1'; \ + for f in x $$MAKEFLAGS; do \ + case $$f in \ + *=* | --[!k]*);; \ + *k*) failcom='fail=yes';; \ + esac; \ + done; \ + dot_seen=no; \ + case "$@" in \ + distclean-* | maintainer-clean-*) list='$(DIST_SUBDIRS)' ;; \ + *) list='$(SUBDIRS)' ;; \ + esac; \ + rev=''; for subdir in $$list; do \ + if test "$$subdir" = "."; then :; else \ + rev="$$subdir $$rev"; \ + fi; \ + done; \ + rev="$$rev ."; \ + target=`echo $@ | sed s/-recursive//`; \ + for subdir in $$rev; do \ + echo "Making $$target in $$subdir"; \ + if test "$$subdir" = "."; then \ + local_target="$$target-am"; \ + else \ + local_target="$$target"; \ + fi; \ + ($(am__cd) $$subdir && $(MAKE) $(AM_MAKEFLAGS) $$local_target) \ + || eval $$failcom; \ + done && test -z "$$fail" +tags-recursive: + list='$(SUBDIRS)'; for subdir in $$list; do \ + test "$$subdir" = . || ($(am__cd) $$subdir && $(MAKE) $(AM_MAKEFLAGS) tags); \ + done +ctags-recursive: + list='$(SUBDIRS)'; for subdir in $$list; do \ + test "$$subdir" = . || ($(am__cd) $$subdir && $(MAKE) $(AM_MAKEFLAGS) ctags); \ + done + +ID: $(HEADERS) $(SOURCES) $(LISP) $(TAGS_FILES) + list='$(SOURCES) $(HEADERS) $(LISP) $(TAGS_FILES)'; \ + unique=`for i in $$list; do \ + if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \ + done | \ + $(AWK) '{ files[$$0] = 1; nonempty = 1; } \ + END { if (nonempty) { for (i in files) print i; }; }'`; \ + mkid -fID $$unique +tags: TAGS + +TAGS: tags-recursive $(HEADERS) $(SOURCES) $(TAGS_DEPENDENCIES) \ + $(TAGS_FILES) $(LISP) + set x; \ + here=`pwd`; \ + if ($(ETAGS) --etags-include --version) >/dev/null 2>&1; then \ + include_option=--etags-include; \ + empty_fix=.; \ + else \ + include_option=--include; \ + empty_fix=; \ + fi; \ + list='$(SUBDIRS)'; for subdir in $$list; do \ + if test "$$subdir" = .; then :; else \ + test ! -f $$subdir/TAGS || \ + set "$$@" "$$include_option=$$here/$$subdir/TAGS"; \ + fi; \ + done; \ + list='$(SOURCES) $(HEADERS) $(LISP) $(TAGS_FILES)'; \ + unique=`for i in $$list; do \ + if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \ + done | \ + $(AWK) '{ files[$$0] = 1; nonempty = 1; } \ + END { if (nonempty) { for (i in files) print i; }; }'`; \ + shift; \ + if test -z "$(ETAGS_ARGS)$$*$$unique"; then :; else \ + test -n "$$unique" || unique=$$empty_fix; \ + if test $$# -gt 0; then \ + $(ETAGS) $(ETAGSFLAGS) $(AM_ETAGSFLAGS) $(ETAGS_ARGS) \ + "$$@" $$unique; \ + else \ + $(ETAGS) $(ETAGSFLAGS) $(AM_ETAGSFLAGS) $(ETAGS_ARGS) \ + $$unique; \ + fi; \ + fi +ctags: CTAGS +CTAGS: ctags-recursive $(HEADERS) $(SOURCES) $(TAGS_DEPENDENCIES) \ + $(TAGS_FILES) $(LISP) + list='$(SOURCES) $(HEADERS) $(LISP) $(TAGS_FILES)'; \ + unique=`for i in $$list; do \ + if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \ + done | \ + $(AWK) '{ files[$$0] = 1; nonempty = 1; } \ + END { if (nonempty) { for (i in files) print i; }; }'`; \ + test -z "$(CTAGS_ARGS)$$unique" \ + || $(CTAGS) $(CTAGSFLAGS) $(AM_CTAGSFLAGS) $(CTAGS_ARGS) \ + $$unique + +GTAGS: + here=`$(am__cd) $(top_builddir) && pwd` \ + && $(am__cd) $(top_srcdir) \ + && gtags -i $(GTAGS_ARGS) "$$here" + +distclean-tags: + -rm -f TAGS ID GTAGS GRTAGS GSYMS GPATH tags + +distdir: $(DISTFILES) + @srcdirstrip=`echo "$(srcdir)" | sed 's/[].[^$$\\*]/\\\\&/g'`; \ + topsrcdirstrip=`echo "$(top_srcdir)" | sed 's/[].[^$$\\*]/\\\\&/g'`; \ + list='$(DISTFILES)'; \ + dist_files=`for file in $$list; do echo $$file; done | \ + sed -e "s|^$$srcdirstrip/||;t" \ + -e "s|^$$topsrcdirstrip/|$(top_builddir)/|;t"`; \ + case $$dist_files in \ + */*) $(MKDIR_P) `echo "$$dist_files" | \ + sed '/\//!d;s|^|$(distdir)/|;s,/[^/]*$$,,' | \ + sort -u` ;; \ + esac; \ + for file in $$dist_files; do \ + if test -f $$file || test -d $$file; then d=.; else d=$(srcdir); fi; \ + if test -d $$d/$$file; then \ + dir=`echo "/$$file" | sed -e 's,/[^/]*$$,,'`; \ + if test -d "$(distdir)/$$file"; then \ + find "$(distdir)/$$file" -type d ! -perm -700 -exec chmod u+rwx {} \;; \ + fi; \ + if test -d $(srcdir)/$$file && test $$d != $(srcdir); then \ + cp -fpR $(srcdir)/$$file "$(distdir)$$dir" || exit 1; \ + find "$(distdir)/$$file" -type d ! -perm -700 -exec chmod u+rwx {} \;; \ + fi; \ + cp -fpR $$d/$$file "$(distdir)$$dir" || exit 1; \ + else \ + test -f "$(distdir)/$$file" \ + || cp -p $$d/$$file "$(distdir)/$$file" \ + || exit 1; \ + fi; \ + done + @list='$(DIST_SUBDIRS)'; for subdir in $$list; do \ + if test "$$subdir" = .; then :; else \ + $(am__make_dryrun) \ + || test -d "$(distdir)/$$subdir" \ + || $(MKDIR_P) "$(distdir)/$$subdir" \ + || exit 1; \ + dir1=$$subdir; dir2="$(distdir)/$$subdir"; \ + $(am__relativize); \ + new_distdir=$$reldir; \ + dir1=$$subdir; dir2="$(top_distdir)"; \ + $(am__relativize); \ + new_top_distdir=$$reldir; \ + echo " (cd $$subdir && $(MAKE) $(AM_MAKEFLAGS) top_distdir="$$new_top_distdir" distdir="$$new_distdir" \\"; \ + echo " am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir)"; \ + ($(am__cd) $$subdir && \ + $(MAKE) $(AM_MAKEFLAGS) \ + top_distdir="$$new_top_distdir" \ + distdir="$$new_distdir" \ + am__remove_distdir=: \ + am__skip_length_check=: \ + am__skip_mode_fix=: \ + distdir) \ + || exit 1; \ + fi; \ + done +check-am: all-am +check: $(BUILT_SOURCES) + $(MAKE) $(AM_MAKEFLAGS) check-recursive +all-am: Makefile $(LIBRARIES) $(LTLIBRARIES) $(HEADERS) all-local +installdirs: installdirs-recursive +installdirs-am: +install: $(BUILT_SOURCES) + $(MAKE) $(AM_MAKEFLAGS) install-recursive +install-exec: install-exec-recursive +install-data: install-data-recursive +uninstall: uninstall-recursive + +install-am: all-am + @$(MAKE) $(AM_MAKEFLAGS) install-exec-am install-data-am + +installcheck: installcheck-recursive +install-strip: + if test -z '$(STRIP)'; then \ + $(MAKE) $(AM_MAKEFLAGS) INSTALL_PROGRAM="$(INSTALL_STRIP_PROGRAM)" \ + install_sh_PROGRAM="$(INSTALL_STRIP_PROGRAM)" INSTALL_STRIP_FLAG=-s \ + install; \ + else \ + $(MAKE) $(AM_MAKEFLAGS) INSTALL_PROGRAM="$(INSTALL_STRIP_PROGRAM)" \ + install_sh_PROGRAM="$(INSTALL_STRIP_PROGRAM)" INSTALL_STRIP_FLAG=-s \ + "INSTALL_PROGRAM_ENV=STRIPPROG='$(STRIP)'" install; \ + fi +mostlyclean-generic: + -test -z "$(MOSTLYCLEANFILES)" || rm -f $(MOSTLYCLEANFILES) + +clean-generic: + -test -z "$(CLEANFILES)" || rm -f $(CLEANFILES) + +distclean-generic: + -test -z "$(CONFIG_CLEAN_FILES)" || rm -f $(CONFIG_CLEAN_FILES) + -test . = "$(srcdir)" || test -z "$(CONFIG_CLEAN_VPATH_FILES)" || rm -f $(CONFIG_CLEAN_VPATH_FILES) + -test -z "$(DISTCLEANFILES)" || rm -f $(DISTCLEANFILES) + +maintainer-clean-generic: + @echo "This command is intended for maintainers to use" + @echo "it deletes files that may require special tools to rebuild." + -test -z "$(BUILT_SOURCES)" || rm -f $(BUILT_SOURCES) + -test -z "$(MAINTAINERCLEANFILES)" || rm -f $(MAINTAINERCLEANFILES) +clean: clean-recursive + +clean-am: clean-generic clean-noinstLIBRARIES clean-noinstLTLIBRARIES \ + mostlyclean-am + +distclean: distclean-recursive + -rm -rf ./$(DEPDIR) + -rm -f Makefile +distclean-am: clean-am distclean-compile distclean-generic \ + distclean-tags + +dvi: dvi-recursive + +dvi-am: + +html: html-recursive + +html-am: + +info: info-recursive + +info-am: + +install-data-am: + +install-dvi: install-dvi-recursive + +install-dvi-am: + +install-exec-am: install-exec-local + +install-html: install-html-recursive + +install-html-am: + +install-info: install-info-recursive + +install-info-am: + +install-man: + +install-pdf: install-pdf-recursive + +install-pdf-am: + +install-ps: install-ps-recursive + +install-ps-am: + +installcheck-am: + +maintainer-clean: maintainer-clean-recursive + -rm -rf ./$(DEPDIR) + -rm -f Makefile +maintainer-clean-am: distclean-am maintainer-clean-generic + +mostlyclean: mostlyclean-recursive + +mostlyclean-am: mostlyclean-compile mostlyclean-generic \ + mostlyclean-local + +pdf: pdf-recursive + +pdf-am: + +ps: ps-recursive + +ps-am: + +uninstall-am: uninstall-local + +.MAKE: $(RECURSIVE_CLEAN_TARGETS) $(RECURSIVE_TARGETS) all check \ + ctags-recursive install install-am install-strip \ + tags-recursive + +.PHONY: $(RECURSIVE_CLEAN_TARGETS) $(RECURSIVE_TARGETS) CTAGS GTAGS \ + all all-am all-local check check-am clean clean-generic \ + clean-noinstLIBRARIES clean-noinstLTLIBRARIES ctags \ + ctags-recursive distclean distclean-compile distclean-generic \ + distclean-tags distdir dvi dvi-am html html-am info info-am \ + install install-am install-data install-data-am install-dvi \ + install-dvi-am install-exec install-exec-am install-exec-local \ + install-html install-html-am install-info install-info-am \ + install-man install-pdf install-pdf-am install-ps \ + install-ps-am install-strip installcheck installcheck-am \ + installdirs installdirs-am maintainer-clean \ + maintainer-clean-generic mostlyclean mostlyclean-compile \ + mostlyclean-generic mostlyclean-local pdf pdf-am ps ps-am tags \ + tags-recursive uninstall uninstall-am uninstall-local + + +# We need the following in order to create <alloca.h> when the system +# doesn't have one that works with the given compiler. +@GL_GENERATE_ALLOCA_H_TRUE@alloca.h: alloca.in.h $(top_builddir)/config.status +@GL_GENERATE_ALLOCA_H_TRUE@ $(AM_V_GEN)rm -f $@-t $@ && \ +@GL_GENERATE_ALLOCA_H_TRUE@ { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ +@GL_GENERATE_ALLOCA_H_TRUE@ cat $(srcdir)/alloca.in.h; \ +@GL_GENERATE_ALLOCA_H_TRUE@ } > $@-t && \ +@GL_GENERATE_ALLOCA_H_TRUE@ mv -f $@-t $@ +@GL_GENERATE_ALLOCA_H_FALSE@alloca.h: $(top_builddir)/config.status +@GL_GENERATE_ALLOCA_H_FALSE@ rm -f $@ + +# Listed in the same order as the GNU makefile conventions, and +# provided by autoconf 2.59c+. +# The Automake-defined pkg* macros are appended, in the order +# listed in the Automake 1.10a+ documentation. +configmake.h: Makefile + $(AM_V_GEN)rm -f $@-t && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + echo '#define PREFIX "$(prefix)"'; \ + echo '#define EXEC_PREFIX "$(exec_prefix)"'; \ + echo '#define BINDIR "$(bindir)"'; \ + echo '#define SBINDIR "$(sbindir)"'; \ + echo '#define LIBEXECDIR "$(libexecdir)"'; \ + echo '#define DATAROOTDIR "$(datarootdir)"'; \ + echo '#define DATADIR "$(datadir)"'; \ + echo '#define SYSCONFDIR "$(sysconfdir)"'; \ + echo '#define SHAREDSTATEDIR "$(sharedstatedir)"'; \ + echo '#define LOCALSTATEDIR "$(localstatedir)"'; \ + echo '#define INCLUDEDIR "$(includedir)"'; \ + echo '#define OLDINCLUDEDIR "$(oldincludedir)"'; \ + echo '#define DOCDIR "$(docdir)"'; \ + echo '#define INFODIR "$(infodir)"'; \ + echo '#define HTMLDIR "$(htmldir)"'; \ + echo '#define DVIDIR "$(dvidir)"'; \ + echo '#define PDFDIR "$(pdfdir)"'; \ + echo '#define PSDIR "$(psdir)"'; \ + echo '#define LIBDIR "$(libdir)"'; \ + echo '#define LISPDIR "$(lispdir)"'; \ + echo '#define LOCALEDIR "$(localedir)"'; \ + echo '#define MANDIR "$(mandir)"'; \ + echo '#define MANEXT "$(manext)"'; \ + echo '#define PKGDATADIR "$(pkgdatadir)"'; \ + echo '#define PKGINCLUDEDIR "$(pkgincludedir)"'; \ + echo '#define PKGLIBDIR "$(pkglibdir)"'; \ + echo '#define PKGLIBEXECDIR "$(pkglibexecdir)"'; \ + } | sed '/""/d' > $@-t + mv -f $@-t $@ + +# We need the following in order to create an empty placeholder for +# <langinfo.h> when the system doesn't have one. +langinfo.h: langinfo.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''HAVE_LANGINFO_H''@|$(HAVE_LANGINFO_H)|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_LANGINFO_H''@|$(NEXT_LANGINFO_H)|g' \ + -e 's/@''GNULIB_NL_LANGINFO''@/$(GNULIB_NL_LANGINFO)/g' \ + -e 's|@''HAVE_LANGINFO_CODESET''@|$(HAVE_LANGINFO_CODESET)|g' \ + -e 's|@''HAVE_LANGINFO_T_FMT_AMPM''@|$(HAVE_LANGINFO_T_FMT_AMPM)|g' \ + -e 's|@''HAVE_LANGINFO_ERA''@|$(HAVE_LANGINFO_ERA)|g' \ + -e 's|@''HAVE_LANGINFO_YESEXPR''@|$(HAVE_LANGINFO_YESEXPR)|g' \ + -e 's|@''HAVE_NL_LANGINFO''@|$(HAVE_NL_LANGINFO)|g' \ + -e 's|@''REPLACE_NL_LANGINFO''@|$(REPLACE_NL_LANGINFO)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)' \ + < $(srcdir)/langinfo.in.h; \ + } > $@-t && \ + mv $@-t $@ + +# We need the following in order to install a simple file in $(libdir) +# which is shared with other installed packages. We use a list of referencing +# packages so that "make uninstall" will remove the file if and only if it +# is not used by another installed package. +# On systems with glibc-2.1 or newer, the file is redundant, therefore we +# avoid installing it. + +all-local: charset.alias ref-add.sed ref-del.sed +install-exec-local: install-exec-localcharset +install-exec-localcharset: all-local + if test $(GLIBC21) = no; then \ + case '$(host_os)' in \ + darwin[56]*) \ + need_charset_alias=true ;; \ + darwin* | cygwin* | mingw* | pw32* | cegcc*) \ + need_charset_alias=false ;; \ + *) \ + need_charset_alias=true ;; \ + esac ; \ + else \ + need_charset_alias=false ; \ + fi ; \ + if $$need_charset_alias; then \ + $(mkinstalldirs) $(DESTDIR)$(libdir) ; \ + fi ; \ + if test -f $(charset_alias); then \ + sed -f ref-add.sed $(charset_alias) > $(charset_tmp) ; \ + $(INSTALL_DATA) $(charset_tmp) $(charset_alias) ; \ + rm -f $(charset_tmp) ; \ + else \ + if $$need_charset_alias; then \ + sed -f ref-add.sed charset.alias > $(charset_tmp) ; \ + $(INSTALL_DATA) $(charset_tmp) $(charset_alias) ; \ + rm -f $(charset_tmp) ; \ + fi ; \ + fi + +uninstall-local: uninstall-localcharset +uninstall-localcharset: all-local + if test -f $(charset_alias); then \ + sed -f ref-del.sed $(charset_alias) > $(charset_tmp); \ + if grep '^# Packages using this file: $$' $(charset_tmp) \ + > /dev/null; then \ + rm -f $(charset_alias); \ + else \ + $(INSTALL_DATA) $(charset_tmp) $(charset_alias); \ + fi; \ + rm -f $(charset_tmp); \ + fi + +charset.alias: config.charset + $(AM_V_GEN)rm -f t-$@ $@ && \ + $(SHELL) $(srcdir)/config.charset '$(host)' > t-$@ && \ + mv t-$@ $@ +.sin.sed: + $(AM_V_GEN)rm -f t-$@ $@ && \ + sed -e '/^#/d' -e 's/@''PACKAGE''@/$(PACKAGE)/g' $< > t-$@ && \ + mv t-$@ $@ +# The arg-nonnull.h that gets inserted into generated .h files is the same as +# build-aux/snippet/arg-nonnull.h, except that it has the copyright header cut +# off. +arg-nonnull.h: $(top_srcdir)/./snippet/arg-nonnull.h + $(AM_V_GEN)rm -f $@-t $@ && \ + sed -n -e '/GL_ARG_NONNULL/,$$p' \ + < $(top_srcdir)/./snippet/arg-nonnull.h \ + > $@-t && \ + mv $@-t $@ +# The c++defs.h that gets inserted into generated .h files is the same as +# build-aux/snippet/c++defs.h, except that it has the copyright header cut off. +c++defs.h: $(top_srcdir)/./snippet/c++defs.h + $(AM_V_GEN)rm -f $@-t $@ && \ + sed -n -e '/_GL_CXXDEFS/,$$p' \ + < $(top_srcdir)/./snippet/c++defs.h \ + > $@-t && \ + mv $@-t $@ +# The warn-on-use.h that gets inserted into generated .h files is the same as +# build-aux/snippet/warn-on-use.h, except that it has the copyright header cut +# off. +warn-on-use.h: $(top_srcdir)/./snippet/warn-on-use.h + $(AM_V_GEN)rm -f $@-t $@ && \ + sed -n -e '/^.ifndef/,$$p' \ + < $(top_srcdir)/./snippet/warn-on-use.h \ + > $@-t && \ + mv $@-t $@ + +# We need the following in order to create <stdbool.h> when the system +# doesn't have one that works. +@GL_GENERATE_STDBOOL_H_TRUE@stdbool.h: stdbool.in.h $(top_builddir)/config.status +@GL_GENERATE_STDBOOL_H_TRUE@ $(AM_V_GEN)rm -f $@-t $@ && \ +@GL_GENERATE_STDBOOL_H_TRUE@ { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ +@GL_GENERATE_STDBOOL_H_TRUE@ sed -e 's/@''HAVE__BOOL''@/$(HAVE__BOOL)/g' < $(srcdir)/stdbool.in.h; \ +@GL_GENERATE_STDBOOL_H_TRUE@ } > $@-t && \ +@GL_GENERATE_STDBOOL_H_TRUE@ mv $@-t $@ +@GL_GENERATE_STDBOOL_H_FALSE@stdbool.h: $(top_builddir)/config.status +@GL_GENERATE_STDBOOL_H_FALSE@ rm -f $@ + +# We need the following in order to create <stddef.h> when the system +# doesn't have one that works with the given compiler. +@GL_GENERATE_STDDEF_H_TRUE@stddef.h: stddef.in.h $(top_builddir)/config.status +@GL_GENERATE_STDDEF_H_TRUE@ $(AM_V_GEN)rm -f $@-t $@ && \ +@GL_GENERATE_STDDEF_H_TRUE@ { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */' && \ +@GL_GENERATE_STDDEF_H_TRUE@ sed -e 's|@''GUARD_PREFIX''@|GL|g' \ +@GL_GENERATE_STDDEF_H_TRUE@ -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ +@GL_GENERATE_STDDEF_H_TRUE@ -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ +@GL_GENERATE_STDDEF_H_TRUE@ -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ +@GL_GENERATE_STDDEF_H_TRUE@ -e 's|@''NEXT_STDDEF_H''@|$(NEXT_STDDEF_H)|g' \ +@GL_GENERATE_STDDEF_H_TRUE@ -e 's|@''HAVE_WCHAR_T''@|$(HAVE_WCHAR_T)|g' \ +@GL_GENERATE_STDDEF_H_TRUE@ -e 's|@''REPLACE_NULL''@|$(REPLACE_NULL)|g' \ +@GL_GENERATE_STDDEF_H_TRUE@ < $(srcdir)/stddef.in.h; \ +@GL_GENERATE_STDDEF_H_TRUE@ } > $@-t && \ +@GL_GENERATE_STDDEF_H_TRUE@ mv $@-t $@ +@GL_GENERATE_STDDEF_H_FALSE@stddef.h: $(top_builddir)/config.status +@GL_GENERATE_STDDEF_H_FALSE@ rm -f $@ + +# We need the following in order to create <stdint.h> when the system +# doesn't have one that works with the given compiler. +@GL_GENERATE_STDINT_H_TRUE@stdint.h: stdint.in.h $(top_builddir)/config.status +@GL_GENERATE_STDINT_H_TRUE@ $(AM_V_GEN)rm -f $@-t $@ && \ +@GL_GENERATE_STDINT_H_TRUE@ { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ +@GL_GENERATE_STDINT_H_TRUE@ sed -e 's|@''GUARD_PREFIX''@|GL|g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_STDINT_H''@/$(HAVE_STDINT_H)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's|@''NEXT_STDINT_H''@|$(NEXT_STDINT_H)|g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_SYS_TYPES_H''@/$(HAVE_SYS_TYPES_H)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_INTTYPES_H''@/$(HAVE_INTTYPES_H)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_SYS_INTTYPES_H''@/$(HAVE_SYS_INTTYPES_H)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_SYS_BITYPES_H''@/$(HAVE_SYS_BITYPES_H)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_WCHAR_H''@/$(HAVE_WCHAR_H)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_LONG_LONG_INT''@/$(HAVE_LONG_LONG_INT)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_UNSIGNED_LONG_LONG_INT''@/$(HAVE_UNSIGNED_LONG_LONG_INT)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''APPLE_UNIVERSAL_BUILD''@/$(APPLE_UNIVERSAL_BUILD)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''BITSIZEOF_PTRDIFF_T''@/$(BITSIZEOF_PTRDIFF_T)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''PTRDIFF_T_SUFFIX''@/$(PTRDIFF_T_SUFFIX)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''BITSIZEOF_SIG_ATOMIC_T''@/$(BITSIZEOF_SIG_ATOMIC_T)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_SIGNED_SIG_ATOMIC_T''@/$(HAVE_SIGNED_SIG_ATOMIC_T)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''SIG_ATOMIC_T_SUFFIX''@/$(SIG_ATOMIC_T_SUFFIX)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''BITSIZEOF_SIZE_T''@/$(BITSIZEOF_SIZE_T)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''SIZE_T_SUFFIX''@/$(SIZE_T_SUFFIX)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''BITSIZEOF_WCHAR_T''@/$(BITSIZEOF_WCHAR_T)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_SIGNED_WCHAR_T''@/$(HAVE_SIGNED_WCHAR_T)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''WCHAR_T_SUFFIX''@/$(WCHAR_T_SUFFIX)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''BITSIZEOF_WINT_T''@/$(BITSIZEOF_WINT_T)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''HAVE_SIGNED_WINT_T''@/$(HAVE_SIGNED_WINT_T)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ -e 's/@''WINT_T_SUFFIX''@/$(WINT_T_SUFFIX)/g' \ +@GL_GENERATE_STDINT_H_TRUE@ < $(srcdir)/stdint.in.h; \ +@GL_GENERATE_STDINT_H_TRUE@ } > $@-t && \ +@GL_GENERATE_STDINT_H_TRUE@ mv $@-t $@ +@GL_GENERATE_STDINT_H_FALSE@stdint.h: $(top_builddir)/config.status +@GL_GENERATE_STDINT_H_FALSE@ rm -f $@ + +# We need the following in order to create <stdlib.h> when the system +# doesn't have one that works with the given compiler. +stdlib.h: stdlib.in.h $(top_builddir)/config.status $(CXXDEFS_H) \ + $(_NORETURN_H) $(ARG_NONNULL_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */' && \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_STDLIB_H''@|$(NEXT_STDLIB_H)|g' \ + -e 's/@''GNULIB__EXIT''@/$(GNULIB__EXIT)/g' \ + -e 's/@''GNULIB_ATOLL''@/$(GNULIB_ATOLL)/g' \ + -e 's/@''GNULIB_CALLOC_POSIX''@/$(GNULIB_CALLOC_POSIX)/g' \ + -e 's/@''GNULIB_CANONICALIZE_FILE_NAME''@/$(GNULIB_CANONICALIZE_FILE_NAME)/g' \ + -e 's/@''GNULIB_GETLOADAVG''@/$(GNULIB_GETLOADAVG)/g' \ + -e 's/@''GNULIB_GETSUBOPT''@/$(GNULIB_GETSUBOPT)/g' \ + -e 's/@''GNULIB_GRANTPT''@/$(GNULIB_GRANTPT)/g' \ + -e 's/@''GNULIB_MALLOC_POSIX''@/$(GNULIB_MALLOC_POSIX)/g' \ + -e 's/@''GNULIB_MBTOWC''@/$(GNULIB_MBTOWC)/g' \ + -e 's/@''GNULIB_MKDTEMP''@/$(GNULIB_MKDTEMP)/g' \ + -e 's/@''GNULIB_MKOSTEMP''@/$(GNULIB_MKOSTEMP)/g' \ + -e 's/@''GNULIB_MKOSTEMPS''@/$(GNULIB_MKOSTEMPS)/g' \ + -e 's/@''GNULIB_MKSTEMP''@/$(GNULIB_MKSTEMP)/g' \ + -e 's/@''GNULIB_MKSTEMPS''@/$(GNULIB_MKSTEMPS)/g' \ + -e 's/@''GNULIB_POSIX_OPENPT''@/$(GNULIB_POSIX_OPENPT)/g' \ + -e 's/@''GNULIB_PTSNAME''@/$(GNULIB_PTSNAME)/g' \ + -e 's/@''GNULIB_PTSNAME_R''@/$(GNULIB_PTSNAME_R)/g' \ + -e 's/@''GNULIB_PUTENV''@/$(GNULIB_PUTENV)/g' \ + -e 's/@''GNULIB_RANDOM''@/$(GNULIB_RANDOM)/g' \ + -e 's/@''GNULIB_RANDOM_R''@/$(GNULIB_RANDOM_R)/g' \ + -e 's/@''GNULIB_REALLOC_POSIX''@/$(GNULIB_REALLOC_POSIX)/g' \ + -e 's/@''GNULIB_REALPATH''@/$(GNULIB_REALPATH)/g' \ + -e 's/@''GNULIB_RPMATCH''@/$(GNULIB_RPMATCH)/g' \ + -e 's/@''GNULIB_SETENV''@/$(GNULIB_SETENV)/g' \ + -e 's/@''GNULIB_STRTOD''@/$(GNULIB_STRTOD)/g' \ + -e 's/@''GNULIB_STRTOLL''@/$(GNULIB_STRTOLL)/g' \ + -e 's/@''GNULIB_STRTOULL''@/$(GNULIB_STRTOULL)/g' \ + -e 's/@''GNULIB_SYSTEM_POSIX''@/$(GNULIB_SYSTEM_POSIX)/g' \ + -e 's/@''GNULIB_UNLOCKPT''@/$(GNULIB_UNLOCKPT)/g' \ + -e 's/@''GNULIB_UNSETENV''@/$(GNULIB_UNSETENV)/g' \ + -e 's/@''GNULIB_WCTOMB''@/$(GNULIB_WCTOMB)/g' \ + < $(srcdir)/stdlib.in.h | \ + sed -e 's|@''HAVE__EXIT''@|$(HAVE__EXIT)|g' \ + -e 's|@''HAVE_ATOLL''@|$(HAVE_ATOLL)|g' \ + -e 's|@''HAVE_CANONICALIZE_FILE_NAME''@|$(HAVE_CANONICALIZE_FILE_NAME)|g' \ + -e 's|@''HAVE_DECL_GETLOADAVG''@|$(HAVE_DECL_GETLOADAVG)|g' \ + -e 's|@''HAVE_GETSUBOPT''@|$(HAVE_GETSUBOPT)|g' \ + -e 's|@''HAVE_GRANTPT''@|$(HAVE_GRANTPT)|g' \ + -e 's|@''HAVE_MKDTEMP''@|$(HAVE_MKDTEMP)|g' \ + -e 's|@''HAVE_MKOSTEMP''@|$(HAVE_MKOSTEMP)|g' \ + -e 's|@''HAVE_MKOSTEMPS''@|$(HAVE_MKOSTEMPS)|g' \ + -e 's|@''HAVE_MKSTEMP''@|$(HAVE_MKSTEMP)|g' \ + -e 's|@''HAVE_MKSTEMPS''@|$(HAVE_MKSTEMPS)|g' \ + -e 's|@''HAVE_POSIX_OPENPT''@|$(HAVE_POSIX_OPENPT)|g' \ + -e 's|@''HAVE_PTSNAME''@|$(HAVE_PTSNAME)|g' \ + -e 's|@''HAVE_PTSNAME_R''@|$(HAVE_PTSNAME_R)|g' \ + -e 's|@''HAVE_RANDOM''@|$(HAVE_RANDOM)|g' \ + -e 's|@''HAVE_RANDOM_H''@|$(HAVE_RANDOM_H)|g' \ + -e 's|@''HAVE_RANDOM_R''@|$(HAVE_RANDOM_R)|g' \ + -e 's|@''HAVE_REALPATH''@|$(HAVE_REALPATH)|g' \ + -e 's|@''HAVE_RPMATCH''@|$(HAVE_RPMATCH)|g' \ + -e 's|@''HAVE_DECL_SETENV''@|$(HAVE_DECL_SETENV)|g' \ + -e 's|@''HAVE_STRTOD''@|$(HAVE_STRTOD)|g' \ + -e 's|@''HAVE_STRTOLL''@|$(HAVE_STRTOLL)|g' \ + -e 's|@''HAVE_STRTOULL''@|$(HAVE_STRTOULL)|g' \ + -e 's|@''HAVE_STRUCT_RANDOM_DATA''@|$(HAVE_STRUCT_RANDOM_DATA)|g' \ + -e 's|@''HAVE_SYS_LOADAVG_H''@|$(HAVE_SYS_LOADAVG_H)|g' \ + -e 's|@''HAVE_UNLOCKPT''@|$(HAVE_UNLOCKPT)|g' \ + -e 's|@''HAVE_DECL_UNSETENV''@|$(HAVE_DECL_UNSETENV)|g' \ + -e 's|@''REPLACE_CALLOC''@|$(REPLACE_CALLOC)|g' \ + -e 's|@''REPLACE_CANONICALIZE_FILE_NAME''@|$(REPLACE_CANONICALIZE_FILE_NAME)|g' \ + -e 's|@''REPLACE_MALLOC''@|$(REPLACE_MALLOC)|g' \ + -e 's|@''REPLACE_MBTOWC''@|$(REPLACE_MBTOWC)|g' \ + -e 's|@''REPLACE_MKSTEMP''@|$(REPLACE_MKSTEMP)|g' \ + -e 's|@''REPLACE_PTSNAME_R''@|$(REPLACE_PTSNAME_R)|g' \ + -e 's|@''REPLACE_PUTENV''@|$(REPLACE_PUTENV)|g' \ + -e 's|@''REPLACE_RANDOM_R''@|$(REPLACE_RANDOM_R)|g' \ + -e 's|@''REPLACE_REALLOC''@|$(REPLACE_REALLOC)|g' \ + -e 's|@''REPLACE_REALPATH''@|$(REPLACE_REALPATH)|g' \ + -e 's|@''REPLACE_SETENV''@|$(REPLACE_SETENV)|g' \ + -e 's|@''REPLACE_STRTOD''@|$(REPLACE_STRTOD)|g' \ + -e 's|@''REPLACE_UNSETENV''@|$(REPLACE_UNSETENV)|g' \ + -e 's|@''REPLACE_WCTOMB''@|$(REPLACE_WCTOMB)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _Noreturn/r $(_NORETURN_H)' \ + -e '/definition of _GL_ARG_NONNULL/r $(ARG_NONNULL_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)'; \ + } > $@-t && \ + mv $@-t $@ + +# We need the following in order to create <strings.h> when the system +# doesn't have one that works with the given compiler. +strings.h: strings.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(WARN_ON_USE_H) $(ARG_NONNULL_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */' && \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''HAVE_STRINGS_H''@|$(HAVE_STRINGS_H)|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_STRINGS_H''@|$(NEXT_STRINGS_H)|g' \ + -e 's|@''GNULIB_FFS''@|$(GNULIB_FFS)|g' \ + -e 's|@''HAVE_FFS''@|$(HAVE_FFS)|g' \ + -e 's|@''HAVE_STRCASECMP''@|$(HAVE_STRCASECMP)|g' \ + -e 's|@''HAVE_DECL_STRNCASECMP''@|$(HAVE_DECL_STRNCASECMP)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_ARG_NONNULL/r $(ARG_NONNULL_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)' \ + < $(srcdir)/strings.in.h; \ + } > $@-t && \ + mv $@-t $@ + +# We need the following in order to create <sys/file.h> when the system +# has one that is incomplete. +sys/file.h: sys_file.in.h $(top_builddir)/config.status $(WARN_ON_USE_H) + $(AM_V_at)$(MKDIR_P) sys + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's/@''HAVE_SYS_FILE_H''@/$(HAVE_SYS_FILE_H)/g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_SYS_FILE_H''@|$(NEXT_SYS_FILE_H)|g' \ + -e 's/@''HAVE_FLOCK''@/$(HAVE_FLOCK)/g' \ + -e 's/@''GNULIB_FLOCK''@/$(GNULIB_FLOCK)/g' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)' \ + < $(srcdir)/sys_file.in.h; \ + } > $@-t && \ + mv $@-t $@ + +# We need the following in order to create <sys/types.h> when the system +# doesn't have one that works with the given compiler. +sys/types.h: sys_types.in.h $(top_builddir)/config.status + $(AM_V_at)$(MKDIR_P) sys + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_SYS_TYPES_H''@|$(NEXT_SYS_TYPES_H)|g' \ + < $(srcdir)/sys_types.in.h; \ + } > $@-t && \ + mv $@-t $@ + +# We need the following in order to create an empty placeholder for +# <unistd.h> when the system doesn't have one. +unistd.h: unistd.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(ARG_NONNULL_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''HAVE_UNISTD_H''@|$(HAVE_UNISTD_H)|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_UNISTD_H''@|$(NEXT_UNISTD_H)|g' \ + -e 's/@''GNULIB_CHDIR''@/$(GNULIB_CHDIR)/g' \ + -e 's/@''GNULIB_CHOWN''@/$(GNULIB_CHOWN)/g' \ + -e 's/@''GNULIB_CLOSE''@/$(GNULIB_CLOSE)/g' \ + -e 's/@''GNULIB_DUP''@/$(GNULIB_DUP)/g' \ + -e 's/@''GNULIB_DUP2''@/$(GNULIB_DUP2)/g' \ + -e 's/@''GNULIB_DUP3''@/$(GNULIB_DUP3)/g' \ + -e 's/@''GNULIB_ENVIRON''@/$(GNULIB_ENVIRON)/g' \ + -e 's/@''GNULIB_EUIDACCESS''@/$(GNULIB_EUIDACCESS)/g' \ + -e 's/@''GNULIB_FACCESSAT''@/$(GNULIB_FACCESSAT)/g' \ + -e 's/@''GNULIB_FCHDIR''@/$(GNULIB_FCHDIR)/g' \ + -e 's/@''GNULIB_FCHOWNAT''@/$(GNULIB_FCHOWNAT)/g' \ + -e 's/@''GNULIB_FDATASYNC''@/$(GNULIB_FDATASYNC)/g' \ + -e 's/@''GNULIB_FSYNC''@/$(GNULIB_FSYNC)/g' \ + -e 's/@''GNULIB_FTRUNCATE''@/$(GNULIB_FTRUNCATE)/g' \ + -e 's/@''GNULIB_GETCWD''@/$(GNULIB_GETCWD)/g' \ + -e 's/@''GNULIB_GETDOMAINNAME''@/$(GNULIB_GETDOMAINNAME)/g' \ + -e 's/@''GNULIB_GETDTABLESIZE''@/$(GNULIB_GETDTABLESIZE)/g' \ + -e 's/@''GNULIB_GETGROUPS''@/$(GNULIB_GETGROUPS)/g' \ + -e 's/@''GNULIB_GETHOSTNAME''@/$(GNULIB_GETHOSTNAME)/g' \ + -e 's/@''GNULIB_GETLOGIN''@/$(GNULIB_GETLOGIN)/g' \ + -e 's/@''GNULIB_GETLOGIN_R''@/$(GNULIB_GETLOGIN_R)/g' \ + -e 's/@''GNULIB_GETPAGESIZE''@/$(GNULIB_GETPAGESIZE)/g' \ + -e 's/@''GNULIB_GETUSERSHELL''@/$(GNULIB_GETUSERSHELL)/g' \ + -e 's/@''GNULIB_GROUP_MEMBER''@/$(GNULIB_GROUP_MEMBER)/g' \ + -e 's/@''GNULIB_ISATTY''@/$(GNULIB_ISATTY)/g' \ + -e 's/@''GNULIB_LCHOWN''@/$(GNULIB_LCHOWN)/g' \ + -e 's/@''GNULIB_LINK''@/$(GNULIB_LINK)/g' \ + -e 's/@''GNULIB_LINKAT''@/$(GNULIB_LINKAT)/g' \ + -e 's/@''GNULIB_LSEEK''@/$(GNULIB_LSEEK)/g' \ + -e 's/@''GNULIB_PIPE''@/$(GNULIB_PIPE)/g' \ + -e 's/@''GNULIB_PIPE2''@/$(GNULIB_PIPE2)/g' \ + -e 's/@''GNULIB_PREAD''@/$(GNULIB_PREAD)/g' \ + -e 's/@''GNULIB_PWRITE''@/$(GNULIB_PWRITE)/g' \ + -e 's/@''GNULIB_READ''@/$(GNULIB_READ)/g' \ + -e 's/@''GNULIB_READLINK''@/$(GNULIB_READLINK)/g' \ + -e 's/@''GNULIB_READLINKAT''@/$(GNULIB_READLINKAT)/g' \ + -e 's/@''GNULIB_RMDIR''@/$(GNULIB_RMDIR)/g' \ + -e 's/@''GNULIB_SETHOSTNAME''@/$(GNULIB_SETHOSTNAME)/g' \ + -e 's/@''GNULIB_SLEEP''@/$(GNULIB_SLEEP)/g' \ + -e 's/@''GNULIB_SYMLINK''@/$(GNULIB_SYMLINK)/g' \ + -e 's/@''GNULIB_SYMLINKAT''@/$(GNULIB_SYMLINKAT)/g' \ + -e 's/@''GNULIB_TTYNAME_R''@/$(GNULIB_TTYNAME_R)/g' \ + -e 's/@''GNULIB_UNISTD_H_GETOPT''@/0$(GNULIB_GL_UNISTD_H_GETOPT)/g' \ + -e 's/@''GNULIB_UNISTD_H_NONBLOCKING''@/$(GNULIB_UNISTD_H_NONBLOCKING)/g' \ + -e 's/@''GNULIB_UNISTD_H_SIGPIPE''@/$(GNULIB_UNISTD_H_SIGPIPE)/g' \ + -e 's/@''GNULIB_UNLINK''@/$(GNULIB_UNLINK)/g' \ + -e 's/@''GNULIB_UNLINKAT''@/$(GNULIB_UNLINKAT)/g' \ + -e 's/@''GNULIB_USLEEP''@/$(GNULIB_USLEEP)/g' \ + -e 's/@''GNULIB_WRITE''@/$(GNULIB_WRITE)/g' \ + < $(srcdir)/unistd.in.h | \ + sed -e 's|@''HAVE_CHOWN''@|$(HAVE_CHOWN)|g' \ + -e 's|@''HAVE_DUP2''@|$(HAVE_DUP2)|g' \ + -e 's|@''HAVE_DUP3''@|$(HAVE_DUP3)|g' \ + -e 's|@''HAVE_EUIDACCESS''@|$(HAVE_EUIDACCESS)|g' \ + -e 's|@''HAVE_FACCESSAT''@|$(HAVE_FACCESSAT)|g' \ + -e 's|@''HAVE_FCHDIR''@|$(HAVE_FCHDIR)|g' \ + -e 's|@''HAVE_FCHOWNAT''@|$(HAVE_FCHOWNAT)|g' \ + -e 's|@''HAVE_FDATASYNC''@|$(HAVE_FDATASYNC)|g' \ + -e 's|@''HAVE_FSYNC''@|$(HAVE_FSYNC)|g' \ + -e 's|@''HAVE_FTRUNCATE''@|$(HAVE_FTRUNCATE)|g' \ + -e 's|@''HAVE_GETDTABLESIZE''@|$(HAVE_GETDTABLESIZE)|g' \ + -e 's|@''HAVE_GETGROUPS''@|$(HAVE_GETGROUPS)|g' \ + -e 's|@''HAVE_GETHOSTNAME''@|$(HAVE_GETHOSTNAME)|g' \ + -e 's|@''HAVE_GETLOGIN''@|$(HAVE_GETLOGIN)|g' \ + -e 's|@''HAVE_GETPAGESIZE''@|$(HAVE_GETPAGESIZE)|g' \ + -e 's|@''HAVE_GROUP_MEMBER''@|$(HAVE_GROUP_MEMBER)|g' \ + -e 's|@''HAVE_LCHOWN''@|$(HAVE_LCHOWN)|g' \ + -e 's|@''HAVE_LINK''@|$(HAVE_LINK)|g' \ + -e 's|@''HAVE_LINKAT''@|$(HAVE_LINKAT)|g' \ + -e 's|@''HAVE_PIPE''@|$(HAVE_PIPE)|g' \ + -e 's|@''HAVE_PIPE2''@|$(HAVE_PIPE2)|g' \ + -e 's|@''HAVE_PREAD''@|$(HAVE_PREAD)|g' \ + -e 's|@''HAVE_PWRITE''@|$(HAVE_PWRITE)|g' \ + -e 's|@''HAVE_READLINK''@|$(HAVE_READLINK)|g' \ + -e 's|@''HAVE_READLINKAT''@|$(HAVE_READLINKAT)|g' \ + -e 's|@''HAVE_SETHOSTNAME''@|$(HAVE_SETHOSTNAME)|g' \ + -e 's|@''HAVE_SLEEP''@|$(HAVE_SLEEP)|g' \ + -e 's|@''HAVE_SYMLINK''@|$(HAVE_SYMLINK)|g' \ + -e 's|@''HAVE_SYMLINKAT''@|$(HAVE_SYMLINKAT)|g' \ + -e 's|@''HAVE_UNLINKAT''@|$(HAVE_UNLINKAT)|g' \ + -e 's|@''HAVE_USLEEP''@|$(HAVE_USLEEP)|g' \ + -e 's|@''HAVE_DECL_ENVIRON''@|$(HAVE_DECL_ENVIRON)|g' \ + -e 's|@''HAVE_DECL_FCHDIR''@|$(HAVE_DECL_FCHDIR)|g' \ + -e 's|@''HAVE_DECL_FDATASYNC''@|$(HAVE_DECL_FDATASYNC)|g' \ + -e 's|@''HAVE_DECL_GETDOMAINNAME''@|$(HAVE_DECL_GETDOMAINNAME)|g' \ + -e 's|@''HAVE_DECL_GETLOGIN_R''@|$(HAVE_DECL_GETLOGIN_R)|g' \ + -e 's|@''HAVE_DECL_GETPAGESIZE''@|$(HAVE_DECL_GETPAGESIZE)|g' \ + -e 's|@''HAVE_DECL_GETUSERSHELL''@|$(HAVE_DECL_GETUSERSHELL)|g' \ + -e 's|@''HAVE_DECL_SETHOSTNAME''@|$(HAVE_DECL_SETHOSTNAME)|g' \ + -e 's|@''HAVE_DECL_TTYNAME_R''@|$(HAVE_DECL_TTYNAME_R)|g' \ + -e 's|@''HAVE_OS_H''@|$(HAVE_OS_H)|g' \ + -e 's|@''HAVE_SYS_PARAM_H''@|$(HAVE_SYS_PARAM_H)|g' \ + | \ + sed -e 's|@''REPLACE_CHOWN''@|$(REPLACE_CHOWN)|g' \ + -e 's|@''REPLACE_CLOSE''@|$(REPLACE_CLOSE)|g' \ + -e 's|@''REPLACE_DUP''@|$(REPLACE_DUP)|g' \ + -e 's|@''REPLACE_DUP2''@|$(REPLACE_DUP2)|g' \ + -e 's|@''REPLACE_FCHOWNAT''@|$(REPLACE_FCHOWNAT)|g' \ + -e 's|@''REPLACE_GETCWD''@|$(REPLACE_GETCWD)|g' \ + -e 's|@''REPLACE_GETDOMAINNAME''@|$(REPLACE_GETDOMAINNAME)|g' \ + -e 's|@''REPLACE_GETLOGIN_R''@|$(REPLACE_GETLOGIN_R)|g' \ + -e 's|@''REPLACE_GETGROUPS''@|$(REPLACE_GETGROUPS)|g' \ + -e 's|@''REPLACE_GETPAGESIZE''@|$(REPLACE_GETPAGESIZE)|g' \ + -e 's|@''REPLACE_ISATTY''@|$(REPLACE_ISATTY)|g' \ + -e 's|@''REPLACE_LCHOWN''@|$(REPLACE_LCHOWN)|g' \ + -e 's|@''REPLACE_LINK''@|$(REPLACE_LINK)|g' \ + -e 's|@''REPLACE_LINKAT''@|$(REPLACE_LINKAT)|g' \ + -e 's|@''REPLACE_LSEEK''@|$(REPLACE_LSEEK)|g' \ + -e 's|@''REPLACE_PREAD''@|$(REPLACE_PREAD)|g' \ + -e 's|@''REPLACE_PWRITE''@|$(REPLACE_PWRITE)|g' \ + -e 's|@''REPLACE_READ''@|$(REPLACE_READ)|g' \ + -e 's|@''REPLACE_READLINK''@|$(REPLACE_READLINK)|g' \ + -e 's|@''REPLACE_RMDIR''@|$(REPLACE_RMDIR)|g' \ + -e 's|@''REPLACE_SLEEP''@|$(REPLACE_SLEEP)|g' \ + -e 's|@''REPLACE_SYMLINK''@|$(REPLACE_SYMLINK)|g' \ + -e 's|@''REPLACE_TTYNAME_R''@|$(REPLACE_TTYNAME_R)|g' \ + -e 's|@''REPLACE_UNLINK''@|$(REPLACE_UNLINK)|g' \ + -e 's|@''REPLACE_UNLINKAT''@|$(REPLACE_UNLINKAT)|g' \ + -e 's|@''REPLACE_USLEEP''@|$(REPLACE_USLEEP)|g' \ + -e 's|@''REPLACE_WRITE''@|$(REPLACE_WRITE)|g' \ + -e 's|@''UNISTD_H_HAVE_WINSOCK2_H''@|$(UNISTD_H_HAVE_WINSOCK2_H)|g' \ + -e 's|@''UNISTD_H_HAVE_WINSOCK2_H_AND_USE_SOCKETS''@|$(UNISTD_H_HAVE_WINSOCK2_H_AND_USE_SOCKETS)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_ARG_NONNULL/r $(ARG_NONNULL_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)'; \ + } > $@-t && \ + mv $@-t $@ + +# We need the following in order to create <wchar.h> when the system +# version does not work standalone. +wchar.h: wchar.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(ARG_NONNULL_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''HAVE_FEATURES_H''@|$(HAVE_FEATURES_H)|g' \ + -e 's|@''NEXT_WCHAR_H''@|$(NEXT_WCHAR_H)|g' \ + -e 's|@''HAVE_WCHAR_H''@|$(HAVE_WCHAR_H)|g' \ + -e 's/@''GNULIB_BTOWC''@/$(GNULIB_BTOWC)/g' \ + -e 's/@''GNULIB_WCTOB''@/$(GNULIB_WCTOB)/g' \ + -e 's/@''GNULIB_MBSINIT''@/$(GNULIB_MBSINIT)/g' \ + -e 's/@''GNULIB_MBRTOWC''@/$(GNULIB_MBRTOWC)/g' \ + -e 's/@''GNULIB_MBRLEN''@/$(GNULIB_MBRLEN)/g' \ + -e 's/@''GNULIB_MBSRTOWCS''@/$(GNULIB_MBSRTOWCS)/g' \ + -e 's/@''GNULIB_MBSNRTOWCS''@/$(GNULIB_MBSNRTOWCS)/g' \ + -e 's/@''GNULIB_WCRTOMB''@/$(GNULIB_WCRTOMB)/g' \ + -e 's/@''GNULIB_WCSRTOMBS''@/$(GNULIB_WCSRTOMBS)/g' \ + -e 's/@''GNULIB_WCSNRTOMBS''@/$(GNULIB_WCSNRTOMBS)/g' \ + -e 's/@''GNULIB_WCWIDTH''@/$(GNULIB_WCWIDTH)/g' \ + -e 's/@''GNULIB_WMEMCHR''@/$(GNULIB_WMEMCHR)/g' \ + -e 's/@''GNULIB_WMEMCMP''@/$(GNULIB_WMEMCMP)/g' \ + -e 's/@''GNULIB_WMEMCPY''@/$(GNULIB_WMEMCPY)/g' \ + -e 's/@''GNULIB_WMEMMOVE''@/$(GNULIB_WMEMMOVE)/g' \ + -e 's/@''GNULIB_WMEMSET''@/$(GNULIB_WMEMSET)/g' \ + -e 's/@''GNULIB_WCSLEN''@/$(GNULIB_WCSLEN)/g' \ + -e 's/@''GNULIB_WCSNLEN''@/$(GNULIB_WCSNLEN)/g' \ + -e 's/@''GNULIB_WCSCPY''@/$(GNULIB_WCSCPY)/g' \ + -e 's/@''GNULIB_WCPCPY''@/$(GNULIB_WCPCPY)/g' \ + -e 's/@''GNULIB_WCSNCPY''@/$(GNULIB_WCSNCPY)/g' \ + -e 's/@''GNULIB_WCPNCPY''@/$(GNULIB_WCPNCPY)/g' \ + -e 's/@''GNULIB_WCSCAT''@/$(GNULIB_WCSCAT)/g' \ + -e 's/@''GNULIB_WCSNCAT''@/$(GNULIB_WCSNCAT)/g' \ + -e 's/@''GNULIB_WCSCMP''@/$(GNULIB_WCSCMP)/g' \ + -e 's/@''GNULIB_WCSNCMP''@/$(GNULIB_WCSNCMP)/g' \ + -e 's/@''GNULIB_WCSCASECMP''@/$(GNULIB_WCSCASECMP)/g' \ + -e 's/@''GNULIB_WCSNCASECMP''@/$(GNULIB_WCSNCASECMP)/g' \ + -e 's/@''GNULIB_WCSCOLL''@/$(GNULIB_WCSCOLL)/g' \ + -e 's/@''GNULIB_WCSXFRM''@/$(GNULIB_WCSXFRM)/g' \ + -e 's/@''GNULIB_WCSDUP''@/$(GNULIB_WCSDUP)/g' \ + -e 's/@''GNULIB_WCSCHR''@/$(GNULIB_WCSCHR)/g' \ + -e 's/@''GNULIB_WCSRCHR''@/$(GNULIB_WCSRCHR)/g' \ + -e 's/@''GNULIB_WCSCSPN''@/$(GNULIB_WCSCSPN)/g' \ + -e 's/@''GNULIB_WCSSPN''@/$(GNULIB_WCSSPN)/g' \ + -e 's/@''GNULIB_WCSPBRK''@/$(GNULIB_WCSPBRK)/g' \ + -e 's/@''GNULIB_WCSSTR''@/$(GNULIB_WCSSTR)/g' \ + -e 's/@''GNULIB_WCSTOK''@/$(GNULIB_WCSTOK)/g' \ + -e 's/@''GNULIB_WCSWIDTH''@/$(GNULIB_WCSWIDTH)/g' \ + < $(srcdir)/wchar.in.h | \ + sed -e 's|@''HAVE_WINT_T''@|$(HAVE_WINT_T)|g' \ + -e 's|@''HAVE_BTOWC''@|$(HAVE_BTOWC)|g' \ + -e 's|@''HAVE_MBSINIT''@|$(HAVE_MBSINIT)|g' \ + -e 's|@''HAVE_MBRTOWC''@|$(HAVE_MBRTOWC)|g' \ + -e 's|@''HAVE_MBRLEN''@|$(HAVE_MBRLEN)|g' \ + -e 's|@''HAVE_MBSRTOWCS''@|$(HAVE_MBSRTOWCS)|g' \ + -e 's|@''HAVE_MBSNRTOWCS''@|$(HAVE_MBSNRTOWCS)|g' \ + -e 's|@''HAVE_WCRTOMB''@|$(HAVE_WCRTOMB)|g' \ + -e 's|@''HAVE_WCSRTOMBS''@|$(HAVE_WCSRTOMBS)|g' \ + -e 's|@''HAVE_WCSNRTOMBS''@|$(HAVE_WCSNRTOMBS)|g' \ + -e 's|@''HAVE_WMEMCHR''@|$(HAVE_WMEMCHR)|g' \ + -e 's|@''HAVE_WMEMCMP''@|$(HAVE_WMEMCMP)|g' \ + -e 's|@''HAVE_WMEMCPY''@|$(HAVE_WMEMCPY)|g' \ + -e 's|@''HAVE_WMEMMOVE''@|$(HAVE_WMEMMOVE)|g' \ + -e 's|@''HAVE_WMEMSET''@|$(HAVE_WMEMSET)|g' \ + -e 's|@''HAVE_WCSLEN''@|$(HAVE_WCSLEN)|g' \ + -e 's|@''HAVE_WCSNLEN''@|$(HAVE_WCSNLEN)|g' \ + -e 's|@''HAVE_WCSCPY''@|$(HAVE_WCSCPY)|g' \ + -e 's|@''HAVE_WCPCPY''@|$(HAVE_WCPCPY)|g' \ + -e 's|@''HAVE_WCSNCPY''@|$(HAVE_WCSNCPY)|g' \ + -e 's|@''HAVE_WCPNCPY''@|$(HAVE_WCPNCPY)|g' \ + -e 's|@''HAVE_WCSCAT''@|$(HAVE_WCSCAT)|g' \ + -e 's|@''HAVE_WCSNCAT''@|$(HAVE_WCSNCAT)|g' \ + -e 's|@''HAVE_WCSCMP''@|$(HAVE_WCSCMP)|g' \ + -e 's|@''HAVE_WCSNCMP''@|$(HAVE_WCSNCMP)|g' \ + -e 's|@''HAVE_WCSCASECMP''@|$(HAVE_WCSCASECMP)|g' \ + -e 's|@''HAVE_WCSNCASECMP''@|$(HAVE_WCSNCASECMP)|g' \ + -e 's|@''HAVE_WCSCOLL''@|$(HAVE_WCSCOLL)|g' \ + -e 's|@''HAVE_WCSXFRM''@|$(HAVE_WCSXFRM)|g' \ + -e 's|@''HAVE_WCSDUP''@|$(HAVE_WCSDUP)|g' \ + -e 's|@''HAVE_WCSCHR''@|$(HAVE_WCSCHR)|g' \ + -e 's|@''HAVE_WCSRCHR''@|$(HAVE_WCSRCHR)|g' \ + -e 's|@''HAVE_WCSCSPN''@|$(HAVE_WCSCSPN)|g' \ + -e 's|@''HAVE_WCSSPN''@|$(HAVE_WCSSPN)|g' \ + -e 's|@''HAVE_WCSPBRK''@|$(HAVE_WCSPBRK)|g' \ + -e 's|@''HAVE_WCSSTR''@|$(HAVE_WCSSTR)|g' \ + -e 's|@''HAVE_WCSTOK''@|$(HAVE_WCSTOK)|g' \ + -e 's|@''HAVE_WCSWIDTH''@|$(HAVE_WCSWIDTH)|g' \ + -e 's|@''HAVE_DECL_WCTOB''@|$(HAVE_DECL_WCTOB)|g' \ + -e 's|@''HAVE_DECL_WCWIDTH''@|$(HAVE_DECL_WCWIDTH)|g' \ + | \ + sed -e 's|@''REPLACE_MBSTATE_T''@|$(REPLACE_MBSTATE_T)|g' \ + -e 's|@''REPLACE_BTOWC''@|$(REPLACE_BTOWC)|g' \ + -e 's|@''REPLACE_WCTOB''@|$(REPLACE_WCTOB)|g' \ + -e 's|@''REPLACE_MBSINIT''@|$(REPLACE_MBSINIT)|g' \ + -e 's|@''REPLACE_MBRTOWC''@|$(REPLACE_MBRTOWC)|g' \ + -e 's|@''REPLACE_MBRLEN''@|$(REPLACE_MBRLEN)|g' \ + -e 's|@''REPLACE_MBSRTOWCS''@|$(REPLACE_MBSRTOWCS)|g' \ + -e 's|@''REPLACE_MBSNRTOWCS''@|$(REPLACE_MBSNRTOWCS)|g' \ + -e 's|@''REPLACE_WCRTOMB''@|$(REPLACE_WCRTOMB)|g' \ + -e 's|@''REPLACE_WCSRTOMBS''@|$(REPLACE_WCSRTOMBS)|g' \ + -e 's|@''REPLACE_WCSNRTOMBS''@|$(REPLACE_WCSNRTOMBS)|g' \ + -e 's|@''REPLACE_WCWIDTH''@|$(REPLACE_WCWIDTH)|g' \ + -e 's|@''REPLACE_WCSWIDTH''@|$(REPLACE_WCSWIDTH)|g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_ARG_NONNULL/r $(ARG_NONNULL_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)'; \ + } > $@-t && \ + mv $@-t $@ + +# We need the following in order to create <wctype.h> when the system +# doesn't have one that works with the given compiler. +wctype.h: wctype.in.h $(top_builddir)/config.status $(CXXDEFS_H) $(WARN_ON_USE_H) + $(AM_V_GEN)rm -f $@-t $@ && \ + { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ + sed -e 's|@''GUARD_PREFIX''@|GL|g' \ + -e 's/@''HAVE_WCTYPE_H''@/$(HAVE_WCTYPE_H)/g' \ + -e 's|@''INCLUDE_NEXT''@|$(INCLUDE_NEXT)|g' \ + -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \ + -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \ + -e 's|@''NEXT_WCTYPE_H''@|$(NEXT_WCTYPE_H)|g' \ + -e 's/@''GNULIB_ISWBLANK''@/$(GNULIB_ISWBLANK)/g' \ + -e 's/@''GNULIB_WCTYPE''@/$(GNULIB_WCTYPE)/g' \ + -e 's/@''GNULIB_ISWCTYPE''@/$(GNULIB_ISWCTYPE)/g' \ + -e 's/@''GNULIB_WCTRANS''@/$(GNULIB_WCTRANS)/g' \ + -e 's/@''GNULIB_TOWCTRANS''@/$(GNULIB_TOWCTRANS)/g' \ + -e 's/@''HAVE_ISWBLANK''@/$(HAVE_ISWBLANK)/g' \ + -e 's/@''HAVE_ISWCNTRL''@/$(HAVE_ISWCNTRL)/g' \ + -e 's/@''HAVE_WCTYPE_T''@/$(HAVE_WCTYPE_T)/g' \ + -e 's/@''HAVE_WCTRANS_T''@/$(HAVE_WCTRANS_T)/g' \ + -e 's/@''HAVE_WINT_T''@/$(HAVE_WINT_T)/g' \ + -e 's/@''REPLACE_ISWBLANK''@/$(REPLACE_ISWBLANK)/g' \ + -e 's/@''REPLACE_ISWCNTRL''@/$(REPLACE_ISWCNTRL)/g' \ + -e 's/@''REPLACE_TOWLOWER''@/$(REPLACE_TOWLOWER)/g' \ + -e '/definitions of _GL_FUNCDECL_RPL/r $(CXXDEFS_H)' \ + -e '/definition of _GL_WARN_ON_USE/r $(WARN_ON_USE_H)' \ + < $(srcdir)/wctype.in.h; \ + } > $@-t && \ + mv $@-t $@ + +mostlyclean-local: mostlyclean-generic + @for dir in '' $(MOSTLYCLEANDIRS); do \ + if test -n "$$dir" && test -d $$dir; then \ + echo "rmdir $$dir"; rmdir $$dir; \ + fi; \ + done; \ + : + +# Tell versions [3.59,3.63) of GNU make to not export all variables. +# Otherwise a system limit (for SysV at least) may be exceeded. +.NOEXPORT: diff --git a/lib/alloca.in.h b/lib/alloca.in.h new file mode 100644 index 0000000..e94eb68 --- /dev/null +++ b/lib/alloca.in.h @@ -0,0 +1,56 @@ +/* Memory allocation on the stack. + + Copyright (C) 1995, 1999, 2001-2004, 2006-2012 Free Software Foundation, + Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see + <http://www.gnu.org/licenses/>. + */ + +/* Avoid using the symbol _ALLOCA_H here, as Bison assumes _ALLOCA_H + means there is a real alloca function. */ +#ifndef _GL_ALLOCA_H +#define _GL_ALLOCA_H + +/* alloca (N) returns a pointer to N bytes of memory + allocated on the stack, which will last until the function returns. + Use of alloca should be avoided: + - inside arguments of function calls - undefined behaviour, + - in inline functions - the allocation may actually last until the + calling function returns, + - for huge N (say, N >= 65536) - you never know how large (or small) + the stack is, and when the stack cannot fulfill the memory allocation + request, the program just crashes. + */ + +#ifndef alloca +# ifdef __GNUC__ +# define alloca __builtin_alloca +# elif defined _AIX +# define alloca __alloca +# elif defined _MSC_VER +# include <malloc.h> +# define alloca _alloca +# elif defined __DECC && defined __VMS +# define alloca __ALLOCA +# else +# include <stddef.h> +# ifdef __cplusplus +extern "C" +# endif +void *alloca (size_t); +# endif +#endif + +#endif /* _GL_ALLOCA_H */ diff --git a/lib/btowc.c b/lib/btowc.c new file mode 100644 index 0000000..cec9eca --- /dev/null +++ b/lib/btowc.c @@ -0,0 +1,39 @@ +/* Convert unibyte character to wide character. + Copyright (C) 2008, 2010-2012 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2008. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +/* Specification. */ +#include <wchar.h> + +#include <stdio.h> +#include <stdlib.h> + +wint_t +btowc (int c) +{ + if (c != EOF) + { + char buf[1]; + wchar_t wc; + + buf[0] = c; + if (mbtowc (&wc, buf, 1) >= 0) + return wc; + } + return WEOF; +} diff --git a/lib/config.charset b/lib/config.charset new file mode 100755 index 0000000..89b3017 --- /dev/null +++ b/lib/config.charset @@ -0,0 +1,684 @@ +#! /bin/sh +# Output a system dependent table of character encoding aliases. +# +# Copyright (C) 2000-2004, 2006-2012 Free Software Foundation, Inc. +# +# This program 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, or (at your option) +# any later version. +# +# This program 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 this program; if not, see <http://www.gnu.org/licenses/>. +# +# The table consists of lines of the form +# ALIAS CANONICAL +# +# ALIAS is the (system dependent) result of "nl_langinfo (CODESET)". +# ALIAS is compared in a case sensitive way. +# +# CANONICAL is the GNU canonical name for this character encoding. +# It must be an encoding supported by libiconv. Support by GNU libc is +# also desirable. CANONICAL is case insensitive. Usually an upper case +# MIME charset name is preferred. +# The current list of GNU canonical charset names is as follows. +# +# name MIME? used by which systems +# (darwin = MacOS X, woe32 = native Windows) +# +# ASCII, ANSI_X3.4-1968 glibc solaris freebsd netbsd darwin cygwin +# ISO-8859-1 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin +# ISO-8859-2 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin +# ISO-8859-3 Y glibc solaris cygwin +# ISO-8859-4 Y osf solaris freebsd netbsd openbsd darwin +# ISO-8859-5 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin +# ISO-8859-6 Y glibc aix hpux solaris cygwin +# ISO-8859-7 Y glibc aix hpux irix osf solaris netbsd openbsd darwin cygwin +# ISO-8859-8 Y glibc aix hpux osf solaris cygwin +# ISO-8859-9 Y glibc aix hpux irix osf solaris darwin cygwin +# ISO-8859-13 glibc netbsd openbsd darwin cygwin +# ISO-8859-14 glibc cygwin +# ISO-8859-15 glibc aix osf solaris freebsd netbsd openbsd darwin cygwin +# KOI8-R Y glibc solaris freebsd netbsd openbsd darwin +# KOI8-U Y glibc freebsd netbsd openbsd darwin cygwin +# KOI8-T glibc +# CP437 dos +# CP775 dos +# CP850 aix osf dos +# CP852 dos +# CP855 dos +# CP856 aix +# CP857 dos +# CP861 dos +# CP862 dos +# CP864 dos +# CP865 dos +# CP866 freebsd netbsd openbsd darwin dos +# CP869 dos +# CP874 woe32 dos +# CP922 aix +# CP932 aix cygwin woe32 dos +# CP943 aix +# CP949 osf darwin woe32 dos +# CP950 woe32 dos +# CP1046 aix +# CP1124 aix +# CP1125 dos +# CP1129 aix +# CP1131 darwin +# CP1250 woe32 +# CP1251 glibc solaris netbsd openbsd darwin cygwin woe32 +# CP1252 aix woe32 +# CP1253 woe32 +# CP1254 woe32 +# CP1255 glibc woe32 +# CP1256 woe32 +# CP1257 woe32 +# GB2312 Y glibc aix hpux irix solaris freebsd netbsd darwin +# EUC-JP Y glibc aix hpux irix osf solaris freebsd netbsd darwin +# EUC-KR Y glibc aix hpux irix osf solaris freebsd netbsd darwin cygwin +# EUC-TW glibc aix hpux irix osf solaris netbsd +# BIG5 Y glibc aix hpux osf solaris freebsd netbsd darwin cygwin +# BIG5-HKSCS glibc solaris darwin +# GBK glibc aix osf solaris darwin cygwin woe32 dos +# GB18030 glibc solaris netbsd darwin +# SHIFT_JIS Y hpux osf solaris freebsd netbsd darwin +# JOHAB glibc solaris woe32 +# TIS-620 glibc aix hpux osf solaris cygwin +# VISCII Y glibc +# TCVN5712-1 glibc +# ARMSCII-8 glibc darwin +# GEORGIAN-PS glibc cygwin +# PT154 glibc +# HP-ROMAN8 hpux +# HP-ARABIC8 hpux +# HP-GREEK8 hpux +# HP-HEBREW8 hpux +# HP-TURKISH8 hpux +# HP-KANA8 hpux +# DEC-KANJI osf +# DEC-HANYU osf +# UTF-8 Y glibc aix hpux osf solaris netbsd darwin cygwin +# +# Note: Names which are not marked as being a MIME name should not be used in +# Internet protocols for information interchange (mail, news, etc.). +# +# Note: ASCII and ANSI_X3.4-1968 are synonymous canonical names. Applications +# must understand both names and treat them as equivalent. +# +# The first argument passed to this file is the canonical host specification, +# CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM +# or +# CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM + +host="$1" +os=`echo "$host" | sed -e 's/^[^-]*-[^-]*-\(.*\)$/\1/'` +echo "# This file contains a table of character encoding aliases," +echo "# suitable for operating system '${os}'." +echo "# It was automatically generated from config.charset." +# List of references, updated during installation: +echo "# Packages using this file: " +case "$os" in + linux-gnulibc1*) + # Linux libc5 doesn't have nl_langinfo(CODESET); therefore + # localcharset.c falls back to using the full locale name + # from the environment variables. + echo "C ASCII" + echo "POSIX ASCII" + for l in af af_ZA ca ca_ES da da_DK de de_AT de_BE de_CH de_DE de_LU \ + en en_AU en_BW en_CA en_DK en_GB en_IE en_NZ en_US en_ZA \ + en_ZW es es_AR es_BO es_CL es_CO es_DO es_EC es_ES es_GT \ + es_HN es_MX es_PA es_PE es_PY es_SV es_US es_UY es_VE et \ + et_EE eu eu_ES fi fi_FI fo fo_FO fr fr_BE fr_CA fr_CH fr_FR \ + fr_LU ga ga_IE gl gl_ES id id_ID in in_ID is is_IS it it_CH \ + it_IT kl kl_GL nl nl_BE nl_NL no no_NO pt pt_BR pt_PT sv \ + sv_FI sv_SE; do + echo "$l ISO-8859-1" + echo "$l.iso-8859-1 ISO-8859-1" + echo "$l.iso-8859-15 ISO-8859-15" + echo "$l.iso-8859-15@euro ISO-8859-15" + echo "$l@euro ISO-8859-15" + echo "$l.cp-437 CP437" + echo "$l.cp-850 CP850" + echo "$l.cp-1252 CP1252" + echo "$l.cp-1252@euro CP1252" + #echo "$l.atari-st ATARI-ST" # not a commonly used encoding + echo "$l.utf-8 UTF-8" + echo "$l.utf-8@euro UTF-8" + done + for l in cs cs_CZ hr hr_HR hu hu_HU pl pl_PL ro ro_RO sk sk_SK sl \ + sl_SI sr sr_CS sr_YU; do + echo "$l ISO-8859-2" + echo "$l.iso-8859-2 ISO-8859-2" + echo "$l.cp-852 CP852" + echo "$l.cp-1250 CP1250" + echo "$l.utf-8 UTF-8" + done + for l in mk mk_MK ru ru_RU; do + echo "$l ISO-8859-5" + echo "$l.iso-8859-5 ISO-8859-5" + echo "$l.koi8-r KOI8-R" + echo "$l.cp-866 CP866" + echo "$l.cp-1251 CP1251" + echo "$l.utf-8 UTF-8" + done + for l in ar ar_SA; do + echo "$l ISO-8859-6" + echo "$l.iso-8859-6 ISO-8859-6" + echo "$l.cp-864 CP864" + #echo "$l.cp-868 CP868" # not a commonly used encoding + echo "$l.cp-1256 CP1256" + echo "$l.utf-8 UTF-8" + done + for l in el el_GR gr gr_GR; do + echo "$l ISO-8859-7" + echo "$l.iso-8859-7 ISO-8859-7" + echo "$l.cp-869 CP869" + echo "$l.cp-1253 CP1253" + echo "$l.cp-1253@euro CP1253" + echo "$l.utf-8 UTF-8" + echo "$l.utf-8@euro UTF-8" + done + for l in he he_IL iw iw_IL; do + echo "$l ISO-8859-8" + echo "$l.iso-8859-8 ISO-8859-8" + echo "$l.cp-862 CP862" + echo "$l.cp-1255 CP1255" + echo "$l.utf-8 UTF-8" + done + for l in tr tr_TR; do + echo "$l ISO-8859-9" + echo "$l.iso-8859-9 ISO-8859-9" + echo "$l.cp-857 CP857" + echo "$l.cp-1254 CP1254" + echo "$l.utf-8 UTF-8" + done + for l in lt lt_LT lv lv_LV; do + #echo "$l BALTIC" # not a commonly used encoding, wrong encoding name + echo "$l ISO-8859-13" + done + for l in ru_UA uk uk_UA; do + echo "$l KOI8-U" + done + for l in zh zh_CN; do + #echo "$l GB_2312-80" # not a commonly used encoding, wrong encoding name + echo "$l GB2312" + done + for l in ja ja_JP ja_JP.EUC; do + echo "$l EUC-JP" + done + for l in ko ko_KR; do + echo "$l EUC-KR" + done + for l in th th_TH; do + echo "$l TIS-620" + done + for l in fa fa_IR; do + #echo "$l ISIRI-3342" # a broken encoding + echo "$l.utf-8 UTF-8" + done + ;; + linux* | *-gnu*) + # With glibc-2.1 or newer, we don't need any canonicalization, + # because glibc has iconv and both glibc and libiconv support all + # GNU canonical names directly. Therefore, the Makefile does not + # need to install the alias file at all. + # The following applies only to glibc-2.0.x and older libcs. + echo "ISO_646.IRV:1983 ASCII" + ;; + aix*) + echo "ISO8859-1 ISO-8859-1" + echo "ISO8859-2 ISO-8859-2" + echo "ISO8859-5 ISO-8859-5" + echo "ISO8859-6 ISO-8859-6" + echo "ISO8859-7 ISO-8859-7" + echo "ISO8859-8 ISO-8859-8" + echo "ISO8859-9 ISO-8859-9" + echo "ISO8859-15 ISO-8859-15" + echo "IBM-850 CP850" + echo "IBM-856 CP856" + echo "IBM-921 ISO-8859-13" + echo "IBM-922 CP922" + echo "IBM-932 CP932" + echo "IBM-943 CP943" + echo "IBM-1046 CP1046" + echo "IBM-1124 CP1124" + echo "IBM-1129 CP1129" + echo "IBM-1252 CP1252" + echo "IBM-eucCN GB2312" + echo "IBM-eucJP EUC-JP" + echo "IBM-eucKR EUC-KR" + echo "IBM-eucTW EUC-TW" + echo "big5 BIG5" + echo "GBK GBK" + echo "TIS-620 TIS-620" + echo "UTF-8 UTF-8" + ;; + hpux*) + echo "iso88591 ISO-8859-1" + echo "iso88592 ISO-8859-2" + echo "iso88595 ISO-8859-5" + echo "iso88596 ISO-8859-6" + echo "iso88597 ISO-8859-7" + echo "iso88598 ISO-8859-8" + echo "iso88599 ISO-8859-9" + echo "iso885915 ISO-8859-15" + echo "roman8 HP-ROMAN8" + echo "arabic8 HP-ARABIC8" + echo "greek8 HP-GREEK8" + echo "hebrew8 HP-HEBREW8" + echo "turkish8 HP-TURKISH8" + echo "kana8 HP-KANA8" + echo "tis620 TIS-620" + echo "big5 BIG5" + echo "eucJP EUC-JP" + echo "eucKR EUC-KR" + echo "eucTW EUC-TW" + echo "hp15CN GB2312" + #echo "ccdc ?" # what is this? + echo "SJIS SHIFT_JIS" + echo "utf8 UTF-8" + ;; + irix*) + echo "ISO8859-1 ISO-8859-1" + echo "ISO8859-2 ISO-8859-2" + echo "ISO8859-5 ISO-8859-5" + echo "ISO8859-7 ISO-8859-7" + echo "ISO8859-9 ISO-8859-9" + echo "eucCN GB2312" + echo "eucJP EUC-JP" + echo "eucKR EUC-KR" + echo "eucTW EUC-TW" + ;; + osf*) + echo "ISO8859-1 ISO-8859-1" + echo "ISO8859-2 ISO-8859-2" + echo "ISO8859-4 ISO-8859-4" + echo "ISO8859-5 ISO-8859-5" + echo "ISO8859-7 ISO-8859-7" + echo "ISO8859-8 ISO-8859-8" + echo "ISO8859-9 ISO-8859-9" + echo "ISO8859-15 ISO-8859-15" + echo "cp850 CP850" + echo "big5 BIG5" + echo "dechanyu DEC-HANYU" + echo "dechanzi GB2312" + echo "deckanji DEC-KANJI" + echo "deckorean EUC-KR" + echo "eucJP EUC-JP" + echo "eucKR EUC-KR" + echo "eucTW EUC-TW" + echo "GBK GBK" + echo "KSC5601 CP949" + echo "sdeckanji EUC-JP" + echo "SJIS SHIFT_JIS" + echo "TACTIS TIS-620" + echo "UTF-8 UTF-8" + ;; + solaris*) + echo "646 ASCII" + echo "ISO8859-1 ISO-8859-1" + echo "ISO8859-2 ISO-8859-2" + echo "ISO8859-3 ISO-8859-3" + echo "ISO8859-4 ISO-8859-4" + echo "ISO8859-5 ISO-8859-5" + echo "ISO8859-6 ISO-8859-6" + echo "ISO8859-7 ISO-8859-7" + echo "ISO8859-8 ISO-8859-8" + echo "ISO8859-9 ISO-8859-9" + echo "ISO8859-15 ISO-8859-15" + echo "koi8-r KOI8-R" + echo "ansi-1251 CP1251" + echo "BIG5 BIG5" + echo "Big5-HKSCS BIG5-HKSCS" + echo "gb2312 GB2312" + echo "GBK GBK" + echo "GB18030 GB18030" + echo "cns11643 EUC-TW" + echo "5601 EUC-KR" + echo "ko_KR.johap92 JOHAB" + echo "eucJP EUC-JP" + echo "PCK SHIFT_JIS" + echo "TIS620.2533 TIS-620" + #echo "sun_eu_greek ?" # what is this? + echo "UTF-8 UTF-8" + ;; + freebsd* | os2*) + # FreeBSD 4.2 doesn't have nl_langinfo(CODESET); therefore + # localcharset.c falls back to using the full locale name + # from the environment variables. + # Likewise for OS/2. OS/2 has XFree86 just like FreeBSD. Just + # reuse FreeBSD's locale data for OS/2. + echo "C ASCII" + echo "US-ASCII ASCII" + for l in la_LN lt_LN; do + echo "$l.ASCII ASCII" + done + for l in da_DK de_AT de_CH de_DE en_AU en_CA en_GB en_US es_ES \ + fi_FI fr_BE fr_CA fr_CH fr_FR is_IS it_CH it_IT la_LN \ + lt_LN nl_BE nl_NL no_NO pt_PT sv_SE; do + echo "$l.ISO_8859-1 ISO-8859-1" + echo "$l.DIS_8859-15 ISO-8859-15" + done + for l in cs_CZ hr_HR hu_HU la_LN lt_LN pl_PL sl_SI; do + echo "$l.ISO_8859-2 ISO-8859-2" + done + for l in la_LN lt_LT; do + echo "$l.ISO_8859-4 ISO-8859-4" + done + for l in ru_RU ru_SU; do + echo "$l.KOI8-R KOI8-R" + echo "$l.ISO_8859-5 ISO-8859-5" + echo "$l.CP866 CP866" + done + echo "uk_UA.KOI8-U KOI8-U" + echo "zh_TW.BIG5 BIG5" + echo "zh_TW.Big5 BIG5" + echo "zh_CN.EUC GB2312" + echo "ja_JP.EUC EUC-JP" + echo "ja_JP.SJIS SHIFT_JIS" + echo "ja_JP.Shift_JIS SHIFT_JIS" + echo "ko_KR.EUC EUC-KR" + ;; + netbsd*) + echo "646 ASCII" + echo "ISO8859-1 ISO-8859-1" + echo "ISO8859-2 ISO-8859-2" + echo "ISO8859-4 ISO-8859-4" + echo "ISO8859-5 ISO-8859-5" + echo "ISO8859-7 ISO-8859-7" + echo "ISO8859-13 ISO-8859-13" + echo "ISO8859-15 ISO-8859-15" + echo "eucCN GB2312" + echo "eucJP EUC-JP" + echo "eucKR EUC-KR" + echo "eucTW EUC-TW" + echo "BIG5 BIG5" + echo "SJIS SHIFT_JIS" + ;; + openbsd*) + echo "646 ASCII" + echo "ISO8859-1 ISO-8859-1" + echo "ISO8859-2 ISO-8859-2" + echo "ISO8859-4 ISO-8859-4" + echo "ISO8859-5 ISO-8859-5" + echo "ISO8859-7 ISO-8859-7" + echo "ISO8859-13 ISO-8859-13" + echo "ISO8859-15 ISO-8859-15" + ;; + darwin[56]*) + # Darwin 6.8 doesn't have nl_langinfo(CODESET); therefore + # localcharset.c falls back to using the full locale name + # from the environment variables. + echo "C ASCII" + for l in en_AU en_CA en_GB en_US la_LN; do + echo "$l.US-ASCII ASCII" + done + for l in da_DK de_AT de_CH de_DE en_AU en_CA en_GB en_US es_ES \ + fi_FI fr_BE fr_CA fr_CH fr_FR is_IS it_CH it_IT nl_BE \ + nl_NL no_NO pt_PT sv_SE; do + echo "$l ISO-8859-1" + echo "$l.ISO8859-1 ISO-8859-1" + echo "$l.ISO8859-15 ISO-8859-15" + done + for l in la_LN; do + echo "$l.ISO8859-1 ISO-8859-1" + echo "$l.ISO8859-15 ISO-8859-15" + done + for l in cs_CZ hr_HR hu_HU la_LN pl_PL sl_SI; do + echo "$l.ISO8859-2 ISO-8859-2" + done + for l in la_LN lt_LT; do + echo "$l.ISO8859-4 ISO-8859-4" + done + for l in ru_RU; do + echo "$l.KOI8-R KOI8-R" + echo "$l.ISO8859-5 ISO-8859-5" + echo "$l.CP866 CP866" + done + for l in bg_BG; do + echo "$l.CP1251 CP1251" + done + echo "uk_UA.KOI8-U KOI8-U" + echo "zh_TW.BIG5 BIG5" + echo "zh_TW.Big5 BIG5" + echo "zh_CN.EUC GB2312" + echo "ja_JP.EUC EUC-JP" + echo "ja_JP.SJIS SHIFT_JIS" + echo "ko_KR.EUC EUC-KR" + ;; + darwin*) + # Darwin 7.5 has nl_langinfo(CODESET), but sometimes its value is + # useless: + # - It returns the empty string when LANG is set to a locale of the + # form ll_CC, although ll_CC/LC_CTYPE is a symlink to an UTF-8 + # LC_CTYPE file. + # - The environment variables LANG, LC_CTYPE, LC_ALL are not set by + # the system; nl_langinfo(CODESET) returns "US-ASCII" in this case. + # - The documentation says: + # "... all code that calls BSD system routines should ensure + # that the const *char parameters of these routines are in UTF-8 + # encoding. All BSD system functions expect their string + # parameters to be in UTF-8 encoding and nothing else." + # It also says + # "An additional caveat is that string parameters for files, + # paths, and other file-system entities must be in canonical + # UTF-8. In a canonical UTF-8 Unicode string, all decomposable + # characters are decomposed ..." + # but this is not true: You can pass non-decomposed UTF-8 strings + # to file system functions, and it is the OS which will convert + # them to decomposed UTF-8 before accessing the file system. + # - The Apple Terminal application displays UTF-8 by default. + # - However, other applications are free to use different encodings: + # - xterm uses ISO-8859-1 by default. + # - TextEdit uses MacRoman by default. + # We prefer UTF-8 over decomposed UTF-8-MAC because one should + # minimize the use of decomposed Unicode. Unfortunately, through the + # Darwin file system, decomposed UTF-8 strings are leaked into user + # space nevertheless. + # Then there are also the locales with encodings other than US-ASCII + # and UTF-8. These locales can be occasionally useful to users (e.g. + # when grepping through ISO-8859-1 encoded text files), when all their + # file names are in US-ASCII. + echo "ISO8859-1 ISO-8859-1" + echo "ISO8859-2 ISO-8859-2" + echo "ISO8859-4 ISO-8859-4" + echo "ISO8859-5 ISO-8859-5" + echo "ISO8859-7 ISO-8859-7" + echo "ISO8859-9 ISO-8859-9" + echo "ISO8859-13 ISO-8859-13" + echo "ISO8859-15 ISO-8859-15" + echo "KOI8-R KOI8-R" + echo "KOI8-U KOI8-U" + echo "CP866 CP866" + echo "CP949 CP949" + echo "CP1131 CP1131" + echo "CP1251 CP1251" + echo "eucCN GB2312" + echo "GB2312 GB2312" + echo "eucJP EUC-JP" + echo "eucKR EUC-KR" + echo "Big5 BIG5" + echo "Big5HKSCS BIG5-HKSCS" + echo "GBK GBK" + echo "GB18030 GB18030" + echo "SJIS SHIFT_JIS" + echo "ARMSCII-8 ARMSCII-8" + echo "PT154 PT154" + #echo "ISCII-DEV ?" + echo "* UTF-8" + ;; + beos* | haiku*) + # BeOS and Haiku have a single locale, and it has UTF-8 encoding. + echo "* UTF-8" + ;; + msdosdjgpp*) + # DJGPP 2.03 doesn't have nl_langinfo(CODESET); therefore + # localcharset.c falls back to using the full locale name + # from the environment variables. + echo "#" + echo "# The encodings given here may not all be correct." + echo "# If you find that the encoding given for your language and" + echo "# country is not the one your DOS machine actually uses, just" + echo "# correct it in this file, and send a mail to" + echo "# Juan Manuel Guerrero <juan.guerrero@gmx.de>" + echo "# and Bruno Haible <bruno@clisp.org>." + echo "#" + echo "C ASCII" + # ISO-8859-1 languages + echo "ca CP850" + echo "ca_ES CP850" + echo "da CP865" # not CP850 ?? + echo "da_DK CP865" # not CP850 ?? + echo "de CP850" + echo "de_AT CP850" + echo "de_CH CP850" + echo "de_DE CP850" + echo "en CP850" + echo "en_AU CP850" # not CP437 ?? + echo "en_CA CP850" + echo "en_GB CP850" + echo "en_NZ CP437" + echo "en_US CP437" + echo "en_ZA CP850" # not CP437 ?? + echo "es CP850" + echo "es_AR CP850" + echo "es_BO CP850" + echo "es_CL CP850" + echo "es_CO CP850" + echo "es_CR CP850" + echo "es_CU CP850" + echo "es_DO CP850" + echo "es_EC CP850" + echo "es_ES CP850" + echo "es_GT CP850" + echo "es_HN CP850" + echo "es_MX CP850" + echo "es_NI CP850" + echo "es_PA CP850" + echo "es_PY CP850" + echo "es_PE CP850" + echo "es_SV CP850" + echo "es_UY CP850" + echo "es_VE CP850" + echo "et CP850" + echo "et_EE CP850" + echo "eu CP850" + echo "eu_ES CP850" + echo "fi CP850" + echo "fi_FI CP850" + echo "fr CP850" + echo "fr_BE CP850" + echo "fr_CA CP850" + echo "fr_CH CP850" + echo "fr_FR CP850" + echo "ga CP850" + echo "ga_IE CP850" + echo "gd CP850" + echo "gd_GB CP850" + echo "gl CP850" + echo "gl_ES CP850" + echo "id CP850" # not CP437 ?? + echo "id_ID CP850" # not CP437 ?? + echo "is CP861" # not CP850 ?? + echo "is_IS CP861" # not CP850 ?? + echo "it CP850" + echo "it_CH CP850" + echo "it_IT CP850" + echo "lt CP775" + echo "lt_LT CP775" + echo "lv CP775" + echo "lv_LV CP775" + echo "nb CP865" # not CP850 ?? + echo "nb_NO CP865" # not CP850 ?? + echo "nl CP850" + echo "nl_BE CP850" + echo "nl_NL CP850" + echo "nn CP865" # not CP850 ?? + echo "nn_NO CP865" # not CP850 ?? + echo "no CP865" # not CP850 ?? + echo "no_NO CP865" # not CP850 ?? + echo "pt CP850" + echo "pt_BR CP850" + echo "pt_PT CP850" + echo "sv CP850" + echo "sv_SE CP850" + # ISO-8859-2 languages + echo "cs CP852" + echo "cs_CZ CP852" + echo "hr CP852" + echo "hr_HR CP852" + echo "hu CP852" + echo "hu_HU CP852" + echo "pl CP852" + echo "pl_PL CP852" + echo "ro CP852" + echo "ro_RO CP852" + echo "sk CP852" + echo "sk_SK CP852" + echo "sl CP852" + echo "sl_SI CP852" + echo "sq CP852" + echo "sq_AL CP852" + echo "sr CP852" # CP852 or CP866 or CP855 ?? + echo "sr_CS CP852" # CP852 or CP866 or CP855 ?? + echo "sr_YU CP852" # CP852 or CP866 or CP855 ?? + # ISO-8859-3 languages + echo "mt CP850" + echo "mt_MT CP850" + # ISO-8859-5 languages + echo "be CP866" + echo "be_BE CP866" + echo "bg CP866" # not CP855 ?? + echo "bg_BG CP866" # not CP855 ?? + echo "mk CP866" # not CP855 ?? + echo "mk_MK CP866" # not CP855 ?? + echo "ru CP866" + echo "ru_RU CP866" + echo "uk CP1125" + echo "uk_UA CP1125" + # ISO-8859-6 languages + echo "ar CP864" + echo "ar_AE CP864" + echo "ar_DZ CP864" + echo "ar_EG CP864" + echo "ar_IQ CP864" + echo "ar_IR CP864" + echo "ar_JO CP864" + echo "ar_KW CP864" + echo "ar_MA CP864" + echo "ar_OM CP864" + echo "ar_QA CP864" + echo "ar_SA CP864" + echo "ar_SY CP864" + # ISO-8859-7 languages + echo "el CP869" + echo "el_GR CP869" + # ISO-8859-8 languages + echo "he CP862" + echo "he_IL CP862" + # ISO-8859-9 languages + echo "tr CP857" + echo "tr_TR CP857" + # Japanese + echo "ja CP932" + echo "ja_JP CP932" + # Chinese + echo "zh_CN GBK" + echo "zh_TW CP950" # not CP938 ?? + # Korean + echo "kr CP949" # not CP934 ?? + echo "kr_KR CP949" # not CP934 ?? + # Thai + echo "th CP874" + echo "th_TH CP874" + # Other + echo "eo CP850" + echo "eo_EO CP850" + ;; +esac diff --git a/lib/flock.c b/lib/flock.c new file mode 100644 index 0000000..6d26905 --- /dev/null +++ b/lib/flock.c @@ -0,0 +1,220 @@ +/* Emulate flock on platforms that lack it, primarily Windows and MinGW. + + This is derived from sqlite3 sources. + http://www.sqlite.org/cvstrac/rlog?f=sqlite/src/os_win.c + http://www.sqlite.org/copyright.html + + Written by Richard W.M. Jones <rjones.at.redhat.com> + + Copyright (C) 2008-2012 Free Software Foundation, Inc. + + This 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. + + 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 + General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> +#include <sys/file.h> + +#if (defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__ + +/* LockFileEx */ +# define WIN32_LEAN_AND_MEAN +# include <windows.h> + +# include <errno.h> + +/* _get_osfhandle */ +# include "msvc-nothrow.h" + +/* Determine the current size of a file. Because the other braindead + * APIs we'll call need lower/upper 32 bit pairs, keep the file size + * like that too. + */ +static BOOL +file_size (HANDLE h, DWORD * lower, DWORD * upper) +{ + *lower = GetFileSize (h, upper); + return 1; +} + +/* LOCKFILE_FAIL_IMMEDIATELY is undefined on some Windows systems. */ +# ifndef LOCKFILE_FAIL_IMMEDIATELY +# define LOCKFILE_FAIL_IMMEDIATELY 1 +# endif + +/* Acquire a lock. */ +static BOOL +do_lock (HANDLE h, int non_blocking, int exclusive) +{ + BOOL res; + DWORD size_lower, size_upper; + OVERLAPPED ovlp; + int flags = 0; + + /* We're going to lock the whole file, so get the file size. */ + res = file_size (h, &size_lower, &size_upper); + if (!res) + return 0; + + /* Start offset is 0, and also zero the remaining members of this struct. */ + memset (&ovlp, 0, sizeof ovlp); + + if (non_blocking) + flags |= LOCKFILE_FAIL_IMMEDIATELY; + if (exclusive) + flags |= LOCKFILE_EXCLUSIVE_LOCK; + + return LockFileEx (h, flags, 0, size_lower, size_upper, &ovlp); +} + +/* Unlock reader or exclusive lock. */ +static BOOL +do_unlock (HANDLE h) +{ + int res; + DWORD size_lower, size_upper; + + res = file_size (h, &size_lower, &size_upper); + if (!res) + return 0; + + return UnlockFile (h, 0, 0, size_lower, size_upper); +} + +/* Now our BSD-like flock operation. */ +int +flock (int fd, int operation) +{ + HANDLE h = (HANDLE) _get_osfhandle (fd); + DWORD res; + int non_blocking; + + if (h == INVALID_HANDLE_VALUE) + { + errno = EBADF; + return -1; + } + + non_blocking = operation & LOCK_NB; + operation &= ~LOCK_NB; + + switch (operation) + { + case LOCK_SH: + res = do_lock (h, non_blocking, 0); + break; + case LOCK_EX: + res = do_lock (h, non_blocking, 1); + break; + case LOCK_UN: + res = do_unlock (h); + break; + default: + errno = EINVAL; + return -1; + } + + /* Map Windows errors into Unix errnos. As usual MSDN fails to + * document the permissible error codes. + */ + if (!res) + { + DWORD err = GetLastError (); + switch (err) + { + /* This means someone else is holding a lock. */ + case ERROR_LOCK_VIOLATION: + errno = EAGAIN; + break; + + /* Out of memory. */ + case ERROR_NOT_ENOUGH_MEMORY: + errno = ENOMEM; + break; + + case ERROR_BAD_COMMAND: + errno = EINVAL; + break; + + /* Unlikely to be other errors, but at least don't lose the + * error code. + */ + default: + errno = err; + } + + return -1; + } + + return 0; +} + +#else /* !Windows */ + +# ifdef HAVE_STRUCT_FLOCK_L_TYPE +/* We know how to implement flock in terms of fcntl. */ + +# include <fcntl.h> + +# ifdef HAVE_UNISTD_H +# include <unistd.h> +# endif + +# include <errno.h> +# include <string.h> + +int +flock (int fd, int operation) +{ + int cmd, r; + struct flock fl; + + if (operation & LOCK_NB) + cmd = F_SETLK; + else + cmd = F_SETLKW; + operation &= ~LOCK_NB; + + memset (&fl, 0, sizeof fl); + fl.l_whence = SEEK_SET; + /* l_start & l_len are 0, which as a special case means "whole file". */ + + switch (operation) + { + case LOCK_SH: + fl.l_type = F_RDLCK; + break; + case LOCK_EX: + fl.l_type = F_WRLCK; + break; + case LOCK_UN: + fl.l_type = F_UNLCK; + break; + default: + errno = EINVAL; + return -1; + } + + r = fcntl (fd, cmd, &fl); + if (r == -1 && errno == EACCES) + errno = EAGAIN; + + return r; +} + +# else /* !HAVE_STRUCT_FLOCK_L_TYPE */ + +# error "This platform lacks flock function, and Gnulib doesn't provide a replacement. This is a bug in Gnulib." + +# endif /* !HAVE_STRUCT_FLOCK_L_TYPE */ + +#endif /* !Windows */ diff --git a/lib/gettext.h b/lib/gettext.h new file mode 100644 index 0000000..75875cd --- /dev/null +++ b/lib/gettext.h @@ -0,0 +1,285 @@ +/* Convenience header for conditional use of GNU <libintl.h>. + Copyright (C) 1995-1998, 2000-2002, 2004-2006, 2009-2012 Free Software + Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#ifndef _LIBGETTEXT_H +#define _LIBGETTEXT_H 1 + +/* NLS can be disabled through the configure --disable-nls option. */ +#if ENABLE_NLS + +/* Get declarations of GNU message catalog functions. */ +# include <libintl.h> + +/* You can set the DEFAULT_TEXT_DOMAIN macro to specify the domain used by + the gettext() and ngettext() macros. This is an alternative to calling + textdomain(), and is useful for libraries. */ +# ifdef DEFAULT_TEXT_DOMAIN +# undef gettext +# define gettext(Msgid) \ + dgettext (DEFAULT_TEXT_DOMAIN, Msgid) +# undef ngettext +# define ngettext(Msgid1, Msgid2, N) \ + dngettext (DEFAULT_TEXT_DOMAIN, Msgid1, Msgid2, N) +# endif + +#else + +/* Solaris /usr/include/locale.h includes /usr/include/libintl.h, which + chokes if dcgettext is defined as a macro. So include it now, to make + later inclusions of <locale.h> a NOP. We don't include <libintl.h> + as well because people using "gettext.h" will not include <libintl.h>, + and also including <libintl.h> would fail on SunOS 4, whereas <locale.h> + is OK. */ +#if defined(__sun) +# include <locale.h> +#endif + +/* Many header files from the libstdc++ coming with g++ 3.3 or newer include + <libintl.h>, which chokes if dcgettext is defined as a macro. So include + it now, to make later inclusions of <libintl.h> a NOP. */ +#if defined(__cplusplus) && defined(__GNUG__) && (__GNUC__ >= 3) +# include <cstdlib> +# if (__GLIBC__ >= 2 && !defined __UCLIBC__) || _GLIBCXX_HAVE_LIBINTL_H +# include <libintl.h> +# endif +#endif + +/* Disabled NLS. + The casts to 'const char *' serve the purpose of producing warnings + for invalid uses of the value returned from these functions. + On pre-ANSI systems without 'const', the config.h file is supposed to + contain "#define const". */ +# undef gettext +# define gettext(Msgid) ((const char *) (Msgid)) +# undef dgettext +# define dgettext(Domainname, Msgid) ((void) (Domainname), gettext (Msgid)) +# undef dcgettext +# define dcgettext(Domainname, Msgid, Category) \ + ((void) (Category), dgettext (Domainname, Msgid)) +# undef ngettext +# define ngettext(Msgid1, Msgid2, N) \ + ((N) == 1 \ + ? ((void) (Msgid2), (const char *) (Msgid1)) \ + : ((void) (Msgid1), (const char *) (Msgid2))) +# undef dngettext +# define dngettext(Domainname, Msgid1, Msgid2, N) \ + ((void) (Domainname), ngettext (Msgid1, Msgid2, N)) +# undef dcngettext +# define dcngettext(Domainname, Msgid1, Msgid2, N, Category) \ + ((void) (Category), dngettext (Domainname, Msgid1, Msgid2, N)) +# undef textdomain +# define textdomain(Domainname) ((const char *) (Domainname)) +# undef bindtextdomain +# define bindtextdomain(Domainname, Dirname) \ + ((void) (Domainname), (const char *) (Dirname)) +# undef bind_textdomain_codeset +# define bind_textdomain_codeset(Domainname, Codeset) \ + ((void) (Domainname), (const char *) (Codeset)) + +#endif + +/* Prefer gnulib's setlocale override over libintl's setlocale override. */ +#ifdef GNULIB_defined_setlocale +# undef setlocale +# define setlocale rpl_setlocale +#endif + +/* A pseudo function call that serves as a marker for the automated + extraction of messages, but does not call gettext(). The run-time + translation is done at a different place in the code. + The argument, String, should be a literal string. Concatenated strings + and other string expressions won't work. + The macro's expansion is not parenthesized, so that it is suitable as + initializer for static 'char[]' or 'const char[]' variables. */ +#define gettext_noop(String) String + +/* The separator between msgctxt and msgid in a .mo file. */ +#define GETTEXT_CONTEXT_GLUE "\004" + +/* Pseudo function calls, taking a MSGCTXT and a MSGID instead of just a + MSGID. MSGCTXT and MSGID must be string literals. MSGCTXT should be + short and rarely need to change. + The letter 'p' stands for 'particular' or 'special'. */ +#ifdef DEFAULT_TEXT_DOMAIN +# define pgettext(Msgctxt, Msgid) \ + pgettext_aux (DEFAULT_TEXT_DOMAIN, Msgctxt GETTEXT_CONTEXT_GLUE Msgid, Msgid, LC_MESSAGES) +#else +# define pgettext(Msgctxt, Msgid) \ + pgettext_aux (NULL, Msgctxt GETTEXT_CONTEXT_GLUE Msgid, Msgid, LC_MESSAGES) +#endif +#define dpgettext(Domainname, Msgctxt, Msgid) \ + pgettext_aux (Domainname, Msgctxt GETTEXT_CONTEXT_GLUE Msgid, Msgid, LC_MESSAGES) +#define dcpgettext(Domainname, Msgctxt, Msgid, Category) \ + pgettext_aux (Domainname, Msgctxt GETTEXT_CONTEXT_GLUE Msgid, Msgid, Category) +#ifdef DEFAULT_TEXT_DOMAIN +# define npgettext(Msgctxt, Msgid, MsgidPlural, N) \ + npgettext_aux (DEFAULT_TEXT_DOMAIN, Msgctxt GETTEXT_CONTEXT_GLUE Msgid, Msgid, MsgidPlural, N, LC_MESSAGES) +#else +# define npgettext(Msgctxt, Msgid, MsgidPlural, N) \ + npgettext_aux (NULL, Msgctxt GETTEXT_CONTEXT_GLUE Msgid, Msgid, MsgidPlural, N, LC_MESSAGES) +#endif +#define dnpgettext(Domainname, Msgctxt, Msgid, MsgidPlural, N) \ + npgettext_aux (Domainname, Msgctxt GETTEXT_CONTEXT_GLUE Msgid, Msgid, MsgidPlural, N, LC_MESSAGES) +#define dcnpgettext(Domainname, Msgctxt, Msgid, MsgidPlural, N, Category) \ + npgettext_aux (Domainname, Msgctxt GETTEXT_CONTEXT_GLUE Msgid, Msgid, MsgidPlural, N, Category) + +#ifdef __GNUC__ +__inline +#else +#ifdef __cplusplus +inline +#endif +#endif +static const char * +pgettext_aux (const char *domain, + const char *msg_ctxt_id, const char *msgid, + int category) +{ + const char *translation = dcgettext (domain, msg_ctxt_id, category); + if (translation == msg_ctxt_id) + return msgid; + else + return translation; +} + +#ifdef __GNUC__ +__inline +#else +#ifdef __cplusplus +inline +#endif +#endif +static const char * +npgettext_aux (const char *domain, + const char *msg_ctxt_id, const char *msgid, + const char *msgid_plural, unsigned long int n, + int category) +{ + const char *translation = + dcngettext (domain, msg_ctxt_id, msgid_plural, n, category); + if (translation == msg_ctxt_id || translation == msgid_plural) + return (n == 1 ? msgid : msgid_plural); + else + return translation; +} + +/* The same thing extended for non-constant arguments. Here MSGCTXT and MSGID + can be arbitrary expressions. But for string literals these macros are + less efficient than those above. */ + +#include <string.h> + +#define _LIBGETTEXT_HAVE_VARIABLE_SIZE_ARRAYS \ + (((__GNUC__ >= 3 || __GNUG__ >= 2) && !defined __STRICT_ANSI__) \ + /* || __STDC_VERSION__ >= 199901L */ ) + +#if !_LIBGETTEXT_HAVE_VARIABLE_SIZE_ARRAYS +#include <stdlib.h> +#endif + +#define pgettext_expr(Msgctxt, Msgid) \ + dcpgettext_expr (NULL, Msgctxt, Msgid, LC_MESSAGES) +#define dpgettext_expr(Domainname, Msgctxt, Msgid) \ + dcpgettext_expr (Domainname, Msgctxt, Msgid, LC_MESSAGES) + +#ifdef __GNUC__ +__inline +#else +#ifdef __cplusplus +inline +#endif +#endif +static const char * +dcpgettext_expr (const char *domain, + const char *msgctxt, const char *msgid, + int category) +{ + size_t msgctxt_len = strlen (msgctxt) + 1; + size_t msgid_len = strlen (msgid) + 1; + const char *translation; +#if _LIBGETTEXT_HAVE_VARIABLE_SIZE_ARRAYS + char msg_ctxt_id[msgctxt_len + msgid_len]; +#else + char buf[1024]; + char *msg_ctxt_id = + (msgctxt_len + msgid_len <= sizeof (buf) + ? buf + : (char *) malloc (msgctxt_len + msgid_len)); + if (msg_ctxt_id != NULL) +#endif + { + memcpy (msg_ctxt_id, msgctxt, msgctxt_len - 1); + msg_ctxt_id[msgctxt_len - 1] = '\004'; + memcpy (msg_ctxt_id + msgctxt_len, msgid, msgid_len); + translation = dcgettext (domain, msg_ctxt_id, category); +#if !_LIBGETTEXT_HAVE_VARIABLE_SIZE_ARRAYS + if (msg_ctxt_id != buf) + free (msg_ctxt_id); +#endif + if (translation != msg_ctxt_id) + return translation; + } + return msgid; +} + +#define npgettext_expr(Msgctxt, Msgid, MsgidPlural, N) \ + dcnpgettext_expr (NULL, Msgctxt, Msgid, MsgidPlural, N, LC_MESSAGES) +#define dnpgettext_expr(Domainname, Msgctxt, Msgid, MsgidPlural, N) \ + dcnpgettext_expr (Domainname, Msgctxt, Msgid, MsgidPlural, N, LC_MESSAGES) + +#ifdef __GNUC__ +__inline +#else +#ifdef __cplusplus +inline +#endif +#endif +static const char * +dcnpgettext_expr (const char *domain, + const char *msgctxt, const char *msgid, + const char *msgid_plural, unsigned long int n, + int category) +{ + size_t msgctxt_len = strlen (msgctxt) + 1; + size_t msgid_len = strlen (msgid) + 1; + const char *translation; +#if _LIBGETTEXT_HAVE_VARIABLE_SIZE_ARRAYS + char msg_ctxt_id[msgctxt_len + msgid_len]; +#else + char buf[1024]; + char *msg_ctxt_id = + (msgctxt_len + msgid_len <= sizeof (buf) + ? buf + : (char *) malloc (msgctxt_len + msgid_len)); + if (msg_ctxt_id != NULL) +#endif + { + memcpy (msg_ctxt_id, msgctxt, msgctxt_len - 1); + msg_ctxt_id[msgctxt_len - 1] = '\004'; + memcpy (msg_ctxt_id + msgctxt_len, msgid, msgid_len); + translation = dcngettext (domain, msg_ctxt_id, msgid_plural, n, category); +#if !_LIBGETTEXT_HAVE_VARIABLE_SIZE_ARRAYS + if (msg_ctxt_id != buf) + free (msg_ctxt_id); +#endif + if (!(translation == msg_ctxt_id || translation == msgid_plural)) + return translation; + } + return (n == 1 ? msgid : msgid_plural); +} + +#endif /* _LIBGETTEXT_H */ diff --git a/lib/langinfo.in.h b/lib/langinfo.in.h new file mode 100644 index 0000000..605c9f7 --- /dev/null +++ b/lib/langinfo.in.h @@ -0,0 +1,176 @@ +/* Substitute for and wrapper around <langinfo.h>. + Copyright (C) 2009-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +/* + * POSIX <langinfo.h> for platforms that lack it or have an incomplete one. + * <http://www.opengroup.org/onlinepubs/9699919799/basedefs/langinfo.h.html> + */ + +#ifndef _@GUARD_PREFIX@_LANGINFO_H + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +/* The include_next requires a split double-inclusion guard. */ +#if @HAVE_LANGINFO_H@ +# @INCLUDE_NEXT@ @NEXT_LANGINFO_H@ +#endif + +#ifndef _@GUARD_PREFIX@_LANGINFO_H +#define _@GUARD_PREFIX@_LANGINFO_H + + +#if !@HAVE_LANGINFO_H@ + +/* A platform that lacks <langinfo.h>. */ + +/* Assume that it also lacks <nl_types.h> and the nl_item type. */ +# if !GNULIB_defined_nl_item +typedef int nl_item; +# define GNULIB_defined_nl_item 1 +# endif + +/* nl_langinfo items of the LC_CTYPE category */ +# define CODESET 10000 +/* nl_langinfo items of the LC_NUMERIC category */ +# define RADIXCHAR 10001 +# define THOUSEP 10002 +/* nl_langinfo items of the LC_TIME category */ +# define D_T_FMT 10003 +# define D_FMT 10004 +# define T_FMT 10005 +# define T_FMT_AMPM 10006 +# define AM_STR 10007 +# define PM_STR 10008 +# define DAY_1 10009 +# define DAY_2 (DAY_1 + 1) +# define DAY_3 (DAY_1 + 2) +# define DAY_4 (DAY_1 + 3) +# define DAY_5 (DAY_1 + 4) +# define DAY_6 (DAY_1 + 5) +# define DAY_7 (DAY_1 + 6) +# define ABDAY_1 10016 +# define ABDAY_2 (ABDAY_1 + 1) +# define ABDAY_3 (ABDAY_1 + 2) +# define ABDAY_4 (ABDAY_1 + 3) +# define ABDAY_5 (ABDAY_1 + 4) +# define ABDAY_6 (ABDAY_1 + 5) +# define ABDAY_7 (ABDAY_1 + 6) +# define MON_1 10023 +# define MON_2 (MON_1 + 1) +# define MON_3 (MON_1 + 2) +# define MON_4 (MON_1 + 3) +# define MON_5 (MON_1 + 4) +# define MON_6 (MON_1 + 5) +# define MON_7 (MON_1 + 6) +# define MON_8 (MON_1 + 7) +# define MON_9 (MON_1 + 8) +# define MON_10 (MON_1 + 9) +# define MON_11 (MON_1 + 10) +# define MON_12 (MON_1 + 11) +# define ABMON_1 10035 +# define ABMON_2 (ABMON_1 + 1) +# define ABMON_3 (ABMON_1 + 2) +# define ABMON_4 (ABMON_1 + 3) +# define ABMON_5 (ABMON_1 + 4) +# define ABMON_6 (ABMON_1 + 5) +# define ABMON_7 (ABMON_1 + 6) +# define ABMON_8 (ABMON_1 + 7) +# define ABMON_9 (ABMON_1 + 8) +# define ABMON_10 (ABMON_1 + 9) +# define ABMON_11 (ABMON_1 + 10) +# define ABMON_12 (ABMON_1 + 11) +# define ERA 10047 +# define ERA_D_FMT 10048 +# define ERA_D_T_FMT 10049 +# define ERA_T_FMT 10050 +# define ALT_DIGITS 10051 +/* nl_langinfo items of the LC_MONETARY category */ +# define CRNCYSTR 10052 +/* nl_langinfo items of the LC_MESSAGES category */ +# define YESEXPR 10053 +# define NOEXPR 10054 + +#else + +/* A platform that has <langinfo.h>. */ + +# if !@HAVE_LANGINFO_CODESET@ +# define CODESET 10000 +# define GNULIB_defined_CODESET 1 +# endif + +# if !@HAVE_LANGINFO_T_FMT_AMPM@ +# define T_FMT_AMPM 10006 +# define GNULIB_defined_T_FMT_AMPM 1 +# endif + +# if !@HAVE_LANGINFO_ERA@ +# define ERA 10047 +# define ERA_D_FMT 10048 +# define ERA_D_T_FMT 10049 +# define ERA_T_FMT 10050 +# define ALT_DIGITS 10051 +# define GNULIB_defined_ERA 1 +# endif + +# if !@HAVE_LANGINFO_YESEXPR@ +# define YESEXPR 10053 +# define NOEXPR 10054 +# define GNULIB_defined_YESEXPR 1 +# endif + +#endif + +/* The definitions of _GL_FUNCDECL_RPL etc. are copied here. */ + +/* The definition of _GL_WARN_ON_USE is copied here. */ + +/* Declare overridden functions. */ + + +/* Return a piece of locale dependent information. + Note: The difference between nl_langinfo (CODESET) and locale_charset () + is that the latter normalizes the encoding names to GNU conventions. */ + +#if @GNULIB_NL_LANGINFO@ +# if @REPLACE_NL_LANGINFO@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef nl_langinfo +# define nl_langinfo rpl_nl_langinfo +# endif +_GL_FUNCDECL_RPL (nl_langinfo, char *, (nl_item item)); +_GL_CXXALIAS_RPL (nl_langinfo, char *, (nl_item item)); +# else +# if !@HAVE_NL_LANGINFO@ +_GL_FUNCDECL_SYS (nl_langinfo, char *, (nl_item item)); +# endif +_GL_CXXALIAS_SYS (nl_langinfo, char *, (nl_item item)); +# endif +_GL_CXXALIASWARN (nl_langinfo); +#elif defined GNULIB_POSIXCHECK +# undef nl_langinfo +# if HAVE_RAW_DECL_NL_LANGINFO +_GL_WARN_ON_USE (nl_langinfo, "nl_langinfo is not portable - " + "use gnulib module nl_langinfo for portability"); +# endif +#endif + + +#endif /* _@GUARD_PREFIX@_LANGINFO_H */ +#endif /* _@GUARD_PREFIX@_LANGINFO_H */ diff --git a/lib/localcharset.c b/lib/localcharset.c new file mode 100644 index 0000000..113205d --- /dev/null +++ b/lib/localcharset.c @@ -0,0 +1,546 @@ +/* Determine a canonical name for the current locale's character encoding. + + Copyright (C) 2000-2006, 2008-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +/* Written by Bruno Haible <bruno@clisp.org>. */ + +#include <config.h> + +/* Specification. */ +#include "localcharset.h" + +#include <fcntl.h> +#include <stddef.h> +#include <stdio.h> +#include <string.h> +#include <stdlib.h> + +#if defined __APPLE__ && defined __MACH__ && HAVE_LANGINFO_CODESET +# define DARWIN7 /* Darwin 7 or newer, i.e. MacOS X 10.3 or newer */ +#endif + +#if defined _WIN32 || defined __WIN32__ +# define WINDOWS_NATIVE +#endif + +#if defined __EMX__ +/* Assume EMX program runs on OS/2, even if compiled under DOS. */ +# ifndef OS2 +# define OS2 +# endif +#endif + +#if !defined WINDOWS_NATIVE +# include <unistd.h> +# if HAVE_LANGINFO_CODESET +# include <langinfo.h> +# else +# if 0 /* see comment below */ +# include <locale.h> +# endif +# endif +# ifdef __CYGWIN__ +# define WIN32_LEAN_AND_MEAN +# include <windows.h> +# endif +#elif defined WINDOWS_NATIVE +# define WIN32_LEAN_AND_MEAN +# include <windows.h> +#endif +#if defined OS2 +# define INCL_DOS +# include <os2.h> +#endif + +#if ENABLE_RELOCATABLE +# include "relocatable.h" +#else +# define relocate(pathname) (pathname) +#endif + +/* Get LIBDIR. */ +#ifndef LIBDIR +# include "configmake.h" +#endif + +/* Define O_NOFOLLOW to 0 on platforms where it does not exist. */ +#ifndef O_NOFOLLOW +# define O_NOFOLLOW 0 +#endif + +#if defined _WIN32 || defined __WIN32__ || defined __CYGWIN__ || defined __EMX__ || defined __DJGPP__ + /* Native Windows, Cygwin, OS/2, DOS */ +# define ISSLASH(C) ((C) == '/' || (C) == '\\') +#endif + +#ifndef DIRECTORY_SEPARATOR +# define DIRECTORY_SEPARATOR '/' +#endif + +#ifndef ISSLASH +# define ISSLASH(C) ((C) == DIRECTORY_SEPARATOR) +#endif + +#if HAVE_DECL_GETC_UNLOCKED +# undef getc +# define getc getc_unlocked +#endif + +/* The following static variable is declared 'volatile' to avoid a + possible multithread problem in the function get_charset_aliases. If we + are running in a threaded environment, and if two threads initialize + 'charset_aliases' simultaneously, both will produce the same value, + and everything will be ok if the two assignments to 'charset_aliases' + are atomic. But I don't know what will happen if the two assignments mix. */ +#if __STDC__ != 1 +# define volatile /* empty */ +#endif +/* Pointer to the contents of the charset.alias file, if it has already been + read, else NULL. Its format is: + ALIAS_1 '\0' CANONICAL_1 '\0' ... ALIAS_n '\0' CANONICAL_n '\0' '\0' */ +static const char * volatile charset_aliases; + +/* Return a pointer to the contents of the charset.alias file. */ +static const char * +get_charset_aliases (void) +{ + const char *cp; + + cp = charset_aliases; + if (cp == NULL) + { +#if !(defined DARWIN7 || defined VMS || defined WINDOWS_NATIVE || defined __CYGWIN__) + const char *dir; + const char *base = "charset.alias"; + char *file_name; + + /* Make it possible to override the charset.alias location. This is + necessary for running the testsuite before "make install". */ + dir = getenv ("CHARSETALIASDIR"); + if (dir == NULL || dir[0] == '\0') + dir = relocate (LIBDIR); + + /* Concatenate dir and base into freshly allocated file_name. */ + { + size_t dir_len = strlen (dir); + size_t base_len = strlen (base); + int add_slash = (dir_len > 0 && !ISSLASH (dir[dir_len - 1])); + file_name = (char *) malloc (dir_len + add_slash + base_len + 1); + if (file_name != NULL) + { + memcpy (file_name, dir, dir_len); + if (add_slash) + file_name[dir_len] = DIRECTORY_SEPARATOR; + memcpy (file_name + dir_len + add_slash, base, base_len + 1); + } + } + + if (file_name == NULL) + /* Out of memory. Treat the file as empty. */ + cp = ""; + else + { + int fd; + + /* Open the file. Reject symbolic links on platforms that support + O_NOFOLLOW. This is a security feature. Without it, an attacker + could retrieve parts of the contents (namely, the tail of the + first line that starts with "* ") of an arbitrary file by placing + a symbolic link to that file under the name "charset.alias" in + some writable directory and defining the environment variable + CHARSETALIASDIR to point to that directory. */ + fd = open (file_name, + O_RDONLY | (HAVE_WORKING_O_NOFOLLOW ? O_NOFOLLOW : 0)); + if (fd < 0) + /* File not found. Treat it as empty. */ + cp = ""; + else + { + FILE *fp; + + fp = fdopen (fd, "r"); + if (fp == NULL) + { + /* Out of memory. Treat the file as empty. */ + close (fd); + cp = ""; + } + else + { + /* Parse the file's contents. */ + char *res_ptr = NULL; + size_t res_size = 0; + + for (;;) + { + int c; + char buf1[50+1]; + char buf2[50+1]; + size_t l1, l2; + char *old_res_ptr; + + c = getc (fp); + if (c == EOF) + break; + if (c == '\n' || c == ' ' || c == '\t') + continue; + if (c == '#') + { + /* Skip comment, to end of line. */ + do + c = getc (fp); + while (!(c == EOF || c == '\n')); + if (c == EOF) + break; + continue; + } + ungetc (c, fp); + if (fscanf (fp, "%50s %50s", buf1, buf2) < 2) + break; + l1 = strlen (buf1); + l2 = strlen (buf2); + old_res_ptr = res_ptr; + if (res_size == 0) + { + res_size = l1 + 1 + l2 + 1; + res_ptr = (char *) malloc (res_size + 1); + } + else + { + res_size += l1 + 1 + l2 + 1; + res_ptr = (char *) realloc (res_ptr, res_size + 1); + } + if (res_ptr == NULL) + { + /* Out of memory. */ + res_size = 0; + free (old_res_ptr); + break; + } + strcpy (res_ptr + res_size - (l2 + 1) - (l1 + 1), buf1); + strcpy (res_ptr + res_size - (l2 + 1), buf2); + } + fclose (fp); + if (res_size == 0) + cp = ""; + else + { + *(res_ptr + res_size) = '\0'; + cp = res_ptr; + } + } + } + + free (file_name); + } + +#else + +# if defined DARWIN7 + /* To avoid the trouble of installing a file that is shared by many + GNU packages -- many packaging systems have problems with this --, + simply inline the aliases here. */ + cp = "ISO8859-1" "\0" "ISO-8859-1" "\0" + "ISO8859-2" "\0" "ISO-8859-2" "\0" + "ISO8859-4" "\0" "ISO-8859-4" "\0" + "ISO8859-5" "\0" "ISO-8859-5" "\0" + "ISO8859-7" "\0" "ISO-8859-7" "\0" + "ISO8859-9" "\0" "ISO-8859-9" "\0" + "ISO8859-13" "\0" "ISO-8859-13" "\0" + "ISO8859-15" "\0" "ISO-8859-15" "\0" + "KOI8-R" "\0" "KOI8-R" "\0" + "KOI8-U" "\0" "KOI8-U" "\0" + "CP866" "\0" "CP866" "\0" + "CP949" "\0" "CP949" "\0" + "CP1131" "\0" "CP1131" "\0" + "CP1251" "\0" "CP1251" "\0" + "eucCN" "\0" "GB2312" "\0" + "GB2312" "\0" "GB2312" "\0" + "eucJP" "\0" "EUC-JP" "\0" + "eucKR" "\0" "EUC-KR" "\0" + "Big5" "\0" "BIG5" "\0" + "Big5HKSCS" "\0" "BIG5-HKSCS" "\0" + "GBK" "\0" "GBK" "\0" + "GB18030" "\0" "GB18030" "\0" + "SJIS" "\0" "SHIFT_JIS" "\0" + "ARMSCII-8" "\0" "ARMSCII-8" "\0" + "PT154" "\0" "PT154" "\0" + /*"ISCII-DEV" "\0" "?" "\0"*/ + "*" "\0" "UTF-8" "\0"; +# endif + +# if defined VMS + /* To avoid the troubles of an extra file charset.alias_vms in the + sources of many GNU packages, simply inline the aliases here. */ + /* The list of encodings is taken from the OpenVMS 7.3-1 documentation + "Compaq C Run-Time Library Reference Manual for OpenVMS systems" + section 10.7 "Handling Different Character Sets". */ + cp = "ISO8859-1" "\0" "ISO-8859-1" "\0" + "ISO8859-2" "\0" "ISO-8859-2" "\0" + "ISO8859-5" "\0" "ISO-8859-5" "\0" + "ISO8859-7" "\0" "ISO-8859-7" "\0" + "ISO8859-8" "\0" "ISO-8859-8" "\0" + "ISO8859-9" "\0" "ISO-8859-9" "\0" + /* Japanese */ + "eucJP" "\0" "EUC-JP" "\0" + "SJIS" "\0" "SHIFT_JIS" "\0" + "DECKANJI" "\0" "DEC-KANJI" "\0" + "SDECKANJI" "\0" "EUC-JP" "\0" + /* Chinese */ + "eucTW" "\0" "EUC-TW" "\0" + "DECHANYU" "\0" "DEC-HANYU" "\0" + "DECHANZI" "\0" "GB2312" "\0" + /* Korean */ + "DECKOREAN" "\0" "EUC-KR" "\0"; +# endif + +# if defined WINDOWS_NATIVE || defined __CYGWIN__ + /* To avoid the troubles of installing a separate file in the same + directory as the DLL and of retrieving the DLL's directory at + runtime, simply inline the aliases here. */ + + cp = "CP936" "\0" "GBK" "\0" + "CP1361" "\0" "JOHAB" "\0" + "CP20127" "\0" "ASCII" "\0" + "CP20866" "\0" "KOI8-R" "\0" + "CP20936" "\0" "GB2312" "\0" + "CP21866" "\0" "KOI8-RU" "\0" + "CP28591" "\0" "ISO-8859-1" "\0" + "CP28592" "\0" "ISO-8859-2" "\0" + "CP28593" "\0" "ISO-8859-3" "\0" + "CP28594" "\0" "ISO-8859-4" "\0" + "CP28595" "\0" "ISO-8859-5" "\0" + "CP28596" "\0" "ISO-8859-6" "\0" + "CP28597" "\0" "ISO-8859-7" "\0" + "CP28598" "\0" "ISO-8859-8" "\0" + "CP28599" "\0" "ISO-8859-9" "\0" + "CP28605" "\0" "ISO-8859-15" "\0" + "CP38598" "\0" "ISO-8859-8" "\0" + "CP51932" "\0" "EUC-JP" "\0" + "CP51936" "\0" "GB2312" "\0" + "CP51949" "\0" "EUC-KR" "\0" + "CP51950" "\0" "EUC-TW" "\0" + "CP54936" "\0" "GB18030" "\0" + "CP65001" "\0" "UTF-8" "\0"; +# endif +#endif + + charset_aliases = cp; + } + + return cp; +} + +/* Determine the current locale's character encoding, and canonicalize it + into one of the canonical names listed in config.charset. + The result must not be freed; it is statically allocated. + If the canonical name cannot be determined, the result is a non-canonical + name. */ + +#ifdef STATIC +STATIC +#endif +const char * +locale_charset (void) +{ + const char *codeset; + const char *aliases; + +#if !(defined WINDOWS_NATIVE || defined OS2) + +# if HAVE_LANGINFO_CODESET + + /* Most systems support nl_langinfo (CODESET) nowadays. */ + codeset = nl_langinfo (CODESET); + +# ifdef __CYGWIN__ + /* Cygwin < 1.7 does not have locales. nl_langinfo (CODESET) always + returns "US-ASCII". Return the suffix of the locale name from the + environment variables (if present) or the codepage as a number. */ + if (codeset != NULL && strcmp (codeset, "US-ASCII") == 0) + { + const char *locale; + static char buf[2 + 10 + 1]; + + locale = getenv ("LC_ALL"); + if (locale == NULL || locale[0] == '\0') + { + locale = getenv ("LC_CTYPE"); + if (locale == NULL || locale[0] == '\0') + locale = getenv ("LANG"); + } + if (locale != NULL && locale[0] != '\0') + { + /* If the locale name contains an encoding after the dot, return + it. */ + const char *dot = strchr (locale, '.'); + + if (dot != NULL) + { + const char *modifier; + + dot++; + /* Look for the possible @... trailer and remove it, if any. */ + modifier = strchr (dot, '@'); + if (modifier == NULL) + return dot; + if (modifier - dot < sizeof (buf)) + { + memcpy (buf, dot, modifier - dot); + buf [modifier - dot] = '\0'; + return buf; + } + } + } + + /* The Windows API has a function returning the locale's codepage as a + number: GetACP(). This encoding is used by Cygwin, unless the user + has set the environment variable CYGWIN=codepage:oem (which very few + people do). + Output directed to console windows needs to be converted (to + GetOEMCP() if the console is using a raster font, or to + GetConsoleOutputCP() if it is using a TrueType font). Cygwin does + this conversion transparently (see winsup/cygwin/fhandler_console.cc), + converting to GetConsoleOutputCP(). This leads to correct results, + except when SetConsoleOutputCP has been called and a raster font is + in use. */ + sprintf (buf, "CP%u", GetACP ()); + codeset = buf; + } +# endif + +# else + + /* On old systems which lack it, use setlocale or getenv. */ + const char *locale = NULL; + + /* But most old systems don't have a complete set of locales. Some + (like SunOS 4 or DJGPP) have only the C locale. Therefore we don't + use setlocale here; it would return "C" when it doesn't support the + locale name the user has set. */ +# if 0 + locale = setlocale (LC_CTYPE, NULL); +# endif + if (locale == NULL || locale[0] == '\0') + { + locale = getenv ("LC_ALL"); + if (locale == NULL || locale[0] == '\0') + { + locale = getenv ("LC_CTYPE"); + if (locale == NULL || locale[0] == '\0') + locale = getenv ("LANG"); + } + } + + /* On some old systems, one used to set locale = "iso8859_1". On others, + you set it to "language_COUNTRY.charset". In any case, we resolve it + through the charset.alias file. */ + codeset = locale; + +# endif + +#elif defined WINDOWS_NATIVE + + static char buf[2 + 10 + 1]; + + /* The Windows API has a function returning the locale's codepage as a + number: GetACP(). + When the output goes to a console window, it needs to be provided in + GetOEMCP() encoding if the console is using a raster font, or in + GetConsoleOutputCP() encoding if it is using a TrueType font. + But in GUI programs and for output sent to files and pipes, GetACP() + encoding is the best bet. */ + sprintf (buf, "CP%u", GetACP ()); + codeset = buf; + +#elif defined OS2 + + const char *locale; + static char buf[2 + 10 + 1]; + ULONG cp[3]; + ULONG cplen; + + /* Allow user to override the codeset, as set in the operating system, + with standard language environment variables. */ + locale = getenv ("LC_ALL"); + if (locale == NULL || locale[0] == '\0') + { + locale = getenv ("LC_CTYPE"); + if (locale == NULL || locale[0] == '\0') + locale = getenv ("LANG"); + } + if (locale != NULL && locale[0] != '\0') + { + /* If the locale name contains an encoding after the dot, return it. */ + const char *dot = strchr (locale, '.'); + + if (dot != NULL) + { + const char *modifier; + + dot++; + /* Look for the possible @... trailer and remove it, if any. */ + modifier = strchr (dot, '@'); + if (modifier == NULL) + return dot; + if (modifier - dot < sizeof (buf)) + { + memcpy (buf, dot, modifier - dot); + buf [modifier - dot] = '\0'; + return buf; + } + } + + /* Resolve through the charset.alias file. */ + codeset = locale; + } + else + { + /* OS/2 has a function returning the locale's codepage as a number. */ + if (DosQueryCp (sizeof (cp), cp, &cplen)) + codeset = ""; + else + { + sprintf (buf, "CP%u", cp[0]); + codeset = buf; + } + } + +#endif + + if (codeset == NULL) + /* The canonical name cannot be determined. */ + codeset = ""; + + /* Resolve alias. */ + for (aliases = get_charset_aliases (); + *aliases != '\0'; + aliases += strlen (aliases) + 1, aliases += strlen (aliases) + 1) + if (strcmp (codeset, aliases) == 0 + || (aliases[0] == '*' && aliases[1] == '\0')) + { + codeset = aliases + strlen (aliases) + 1; + break; + } + + /* Don't return an empty string. GNU libc and GNU libiconv interpret + the empty string as denoting "the locale's character encoding", + thus GNU libiconv would call this function a second time. */ + if (codeset[0] == '\0') + codeset = "ASCII"; + + return codeset; +} diff --git a/lib/localcharset.h b/lib/localcharset.h new file mode 100644 index 0000000..39dc593 --- /dev/null +++ b/lib/localcharset.h @@ -0,0 +1,40 @@ +/* Determine a canonical name for the current locale's character encoding. + Copyright (C) 2000-2003, 2009-2012 Free Software Foundation, Inc. + This file is part of the GNU CHARSET Library. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#ifndef _LOCALCHARSET_H +#define _LOCALCHARSET_H + + +#ifdef __cplusplus +extern "C" { +#endif + + +/* Determine the current locale's character encoding, and canonicalize it + into one of the canonical names listed in config.charset. + The result must not be freed; it is statically allocated. + If the canonical name cannot be determined, the result is a non-canonical + name. */ +extern const char * locale_charset (void); + + +#ifdef __cplusplus +} +#endif + + +#endif /* _LOCALCHARSET_H */ diff --git a/lib/malloc.c b/lib/malloc.c new file mode 100644 index 0000000..e0d5c89 --- /dev/null +++ b/lib/malloc.c @@ -0,0 +1,56 @@ +/* malloc() function that is glibc compatible. + + Copyright (C) 1997-1998, 2006-2007, 2009-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +/* written by Jim Meyering and Bruno Haible */ + +#define _GL_USE_STDLIB_ALLOC 1 +#include <config.h> +/* Only the AC_FUNC_MALLOC macro defines 'malloc' already in config.h. */ +#ifdef malloc +# define NEED_MALLOC_GNU 1 +# undef malloc +/* Whereas the gnulib module 'malloc-gnu' defines HAVE_MALLOC_GNU. */ +#elif GNULIB_MALLOC_GNU && !HAVE_MALLOC_GNU +# define NEED_MALLOC_GNU 1 +#endif + +#include <stdlib.h> + +#include <errno.h> + +/* Allocate an N-byte block of memory from the heap. + If N is zero, allocate a 1-byte block. */ + +void * +rpl_malloc (size_t n) +{ + void *result; + +#if NEED_MALLOC_GNU + if (n == 0) + n = 1; +#endif + + result = malloc (n); + +#if !HAVE_MALLOC_POSIX + if (result == NULL) + errno = ENOMEM; +#endif + + return result; +} diff --git a/lib/mbrtowc.c b/lib/mbrtowc.c new file mode 100644 index 0000000..e3f354b --- /dev/null +++ b/lib/mbrtowc.c @@ -0,0 +1,402 @@ +/* Convert multibyte character to wide character. + Copyright (C) 1999-2002, 2005-2012 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2008. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +/* Specification. */ +#include <wchar.h> + +#if GNULIB_defined_mbstate_t +/* Implement mbrtowc() on top of mbtowc(). */ + +# include <errno.h> +# include <stdlib.h> + +# include "localcharset.h" +# include "streq.h" +# include "verify.h" + + +verify (sizeof (mbstate_t) >= 4); + +static char internal_state[4]; + +size_t +mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps) +{ + char *pstate = (char *)ps; + + if (s == NULL) + { + pwc = NULL; + s = ""; + n = 1; + } + + if (n == 0) + return (size_t)(-2); + + /* Here n > 0. */ + + if (pstate == NULL) + pstate = internal_state; + + { + size_t nstate = pstate[0]; + char buf[4]; + const char *p; + size_t m; + + switch (nstate) + { + case 0: + p = s; + m = n; + break; + case 3: + buf[2] = pstate[3]; + /*FALLTHROUGH*/ + case 2: + buf[1] = pstate[2]; + /*FALLTHROUGH*/ + case 1: + buf[0] = pstate[1]; + p = buf; + m = nstate; + buf[m++] = s[0]; + if (n >= 2 && m < 4) + { + buf[m++] = s[1]; + if (n >= 3 && m < 4) + buf[m++] = s[2]; + } + break; + default: + errno = EINVAL; + return (size_t)(-1); + } + + /* Here m > 0. */ + +# if __GLIBC__ || defined __UCLIBC__ + /* Work around bug <http://sourceware.org/bugzilla/show_bug.cgi?id=9674> */ + mbtowc (NULL, NULL, 0); +# endif + { + int res = mbtowc (pwc, p, m); + + if (res >= 0) + { + if (pwc != NULL && ((*pwc == 0) != (res == 0))) + abort (); + if (nstate >= (res > 0 ? res : 1)) + abort (); + res -= nstate; + pstate[0] = 0; + return res; + } + + /* mbtowc does not distinguish between invalid and incomplete multibyte + sequences. But mbrtowc needs to make this distinction. + There are two possible approaches: + - Use iconv() and its return value. + - Use built-in knowledge about the possible encodings. + Given the low quality of implementation of iconv() on the systems that + lack mbrtowc(), we use the second approach. + The possible encodings are: + - 8-bit encodings, + - EUC-JP, EUC-KR, GB2312, EUC-TW, BIG5, GB18030, SJIS, + - UTF-8. + Use specialized code for each. */ + if (m >= 4 || m >= MB_CUR_MAX) + goto invalid; + /* Here MB_CUR_MAX > 1 and 0 < m < 4. */ + { + const char *encoding = locale_charset (); + + if (STREQ_OPT (encoding, "UTF-8", 'U', 'T', 'F', '-', '8', 0, 0, 0, 0)) + { + /* Cf. unistr/u8-mblen.c. */ + unsigned char c = (unsigned char) p[0]; + + if (c >= 0xc2) + { + if (c < 0xe0) + { + if (m == 1) + goto incomplete; + } + else if (c < 0xf0) + { + if (m == 1) + goto incomplete; + if (m == 2) + { + unsigned char c2 = (unsigned char) p[1]; + + if ((c2 ^ 0x80) < 0x40 + && (c >= 0xe1 || c2 >= 0xa0) + && (c != 0xed || c2 < 0xa0)) + goto incomplete; + } + } + else if (c <= 0xf4) + { + if (m == 1) + goto incomplete; + else /* m == 2 || m == 3 */ + { + unsigned char c2 = (unsigned char) p[1]; + + if ((c2 ^ 0x80) < 0x40 + && (c >= 0xf1 || c2 >= 0x90) + && (c < 0xf4 || (c == 0xf4 && c2 < 0x90))) + { + if (m == 2) + goto incomplete; + else /* m == 3 */ + { + unsigned char c3 = (unsigned char) p[2]; + + if ((c3 ^ 0x80) < 0x40) + goto incomplete; + } + } + } + } + } + goto invalid; + } + + /* As a reference for this code, you can use the GNU libiconv + implementation. Look for uses of the RET_TOOFEW macro. */ + + if (STREQ_OPT (encoding, + "EUC-JP", 'E', 'U', 'C', '-', 'J', 'P', 0, 0, 0)) + { + if (m == 1) + { + unsigned char c = (unsigned char) p[0]; + + if ((c >= 0xa1 && c < 0xff) || c == 0x8e || c == 0x8f) + goto incomplete; + } + if (m == 2) + { + unsigned char c = (unsigned char) p[0]; + + if (c == 0x8f) + { + unsigned char c2 = (unsigned char) p[1]; + + if (c2 >= 0xa1 && c2 < 0xff) + goto incomplete; + } + } + goto invalid; + } + if (STREQ_OPT (encoding, + "EUC-KR", 'E', 'U', 'C', '-', 'K', 'R', 0, 0, 0) + || STREQ_OPT (encoding, + "GB2312", 'G', 'B', '2', '3', '1', '2', 0, 0, 0) + || STREQ_OPT (encoding, + "BIG5", 'B', 'I', 'G', '5', 0, 0, 0, 0, 0)) + { + if (m == 1) + { + unsigned char c = (unsigned char) p[0]; + + if (c >= 0xa1 && c < 0xff) + goto incomplete; + } + goto invalid; + } + if (STREQ_OPT (encoding, + "EUC-TW", 'E', 'U', 'C', '-', 'T', 'W', 0, 0, 0)) + { + if (m == 1) + { + unsigned char c = (unsigned char) p[0]; + + if ((c >= 0xa1 && c < 0xff) || c == 0x8e) + goto incomplete; + } + else /* m == 2 || m == 3 */ + { + unsigned char c = (unsigned char) p[0]; + + if (c == 0x8e) + goto incomplete; + } + goto invalid; + } + if (STREQ_OPT (encoding, + "GB18030", 'G', 'B', '1', '8', '0', '3', '0', 0, 0)) + { + if (m == 1) + { + unsigned char c = (unsigned char) p[0]; + + if ((c >= 0x90 && c <= 0xe3) || (c >= 0xf8 && c <= 0xfe)) + goto incomplete; + } + else /* m == 2 || m == 3 */ + { + unsigned char c = (unsigned char) p[0]; + + if (c >= 0x90 && c <= 0xe3) + { + unsigned char c2 = (unsigned char) p[1]; + + if (c2 >= 0x30 && c2 <= 0x39) + { + if (m == 2) + goto incomplete; + else /* m == 3 */ + { + unsigned char c3 = (unsigned char) p[2]; + + if (c3 >= 0x81 && c3 <= 0xfe) + goto incomplete; + } + } + } + } + goto invalid; + } + if (STREQ_OPT (encoding, "SJIS", 'S', 'J', 'I', 'S', 0, 0, 0, 0, 0)) + { + if (m == 1) + { + unsigned char c = (unsigned char) p[0]; + + if ((c >= 0x81 && c <= 0x9f) || (c >= 0xe0 && c <= 0xea) + || (c >= 0xf0 && c <= 0xf9)) + goto incomplete; + } + goto invalid; + } + + /* An unknown multibyte encoding. */ + goto incomplete; + } + + incomplete: + { + size_t k = nstate; + /* Here 0 <= k < m < 4. */ + pstate[++k] = s[0]; + if (k < m) + { + pstate[++k] = s[1]; + if (k < m) + pstate[++k] = s[2]; + } + if (k != m) + abort (); + } + pstate[0] = m; + return (size_t)(-2); + + invalid: + errno = EILSEQ; + /* The conversion state is undefined, says POSIX. */ + return (size_t)(-1); + } + } +} + +#else +/* Override the system's mbrtowc() function. */ + +# undef mbrtowc + +size_t +rpl_mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps) +{ +# if MBRTOWC_NULL_ARG2_BUG || MBRTOWC_RETVAL_BUG + if (s == NULL) + { + pwc = NULL; + s = ""; + n = 1; + } +# endif + +# if MBRTOWC_RETVAL_BUG + { + static mbstate_t internal_state; + + /* Override mbrtowc's internal state. We cannot call mbsinit() on the + hidden internal state, but we can call it on our variable. */ + if (ps == NULL) + ps = &internal_state; + + if (!mbsinit (ps)) + { + /* Parse the rest of the multibyte character byte for byte. */ + size_t count = 0; + for (; n > 0; s++, n--) + { + wchar_t wc; + size_t ret = mbrtowc (&wc, s, 1, ps); + + if (ret == (size_t)(-1)) + return (size_t)(-1); + count++; + if (ret != (size_t)(-2)) + { + /* The multibyte character has been completed. */ + if (pwc != NULL) + *pwc = wc; + return (wc == 0 ? 0 : count); + } + } + return (size_t)(-2); + } + } +# endif + +# if MBRTOWC_NUL_RETVAL_BUG + { + wchar_t wc; + size_t ret = mbrtowc (&wc, s, n, ps); + + if (ret != (size_t)(-1) && ret != (size_t)(-2)) + { + if (pwc != NULL) + *pwc = wc; + if (wc == 0) + ret = 0; + } + return ret; + } +# else + { +# if MBRTOWC_NULL_ARG1_BUG + wchar_t dummy; + + if (pwc == NULL) + pwc = &dummy; +# endif + + return mbrtowc (pwc, s, n, ps); + } +# endif +} + +#endif diff --git a/lib/mbsinit.c b/lib/mbsinit.c new file mode 100644 index 0000000..638142e --- /dev/null +++ b/lib/mbsinit.c @@ -0,0 +1,61 @@ +/* Test for initial conversion state. + Copyright (C) 2008-2012 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2008. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +/* Specification. */ +#include <wchar.h> + +#include "verify.h" + +#if (defined _WIN32 || defined __WIN32__) && !defined __CYGWIN__ + +/* On native Windows, 'mbstate_t' is defined as 'int'. */ + +int +mbsinit (const mbstate_t *ps) +{ + return ps == NULL || *ps == 0; +} + +#else + +/* Platforms that lack mbsinit() also lack mbrlen(), mbrtowc(), mbsrtowcs() + and wcrtomb(), wcsrtombs(). + We assume that + - sizeof (mbstate_t) >= 4, + - only stateless encodings are supported (such as UTF-8 and EUC-JP, but + not ISO-2022 variants), + - for each encoding, the number of bytes for a wide character is <= 4. + (This maximum is attained for UTF-8, GB18030, EUC-TW.) + We define the meaning of mbstate_t as follows: + - In mb -> wc direction, mbstate_t's first byte contains the number of + buffered bytes (in the range 0..3), followed by up to 3 buffered bytes. + - In wc -> mb direction, mbstate_t contains no information. In other + words, it is always in the initial state. */ + +verify (sizeof (mbstate_t) >= 4); + +int +mbsinit (const mbstate_t *ps) +{ + const char *pstate = (const char *)ps; + + return pstate == NULL || pstate[0] == 0; +} + +#endif diff --git a/lib/mbtowc-impl.h b/lib/mbtowc-impl.h new file mode 100644 index 0000000..bb75ed2 --- /dev/null +++ b/lib/mbtowc-impl.h @@ -0,0 +1,44 @@ +/* Convert multibyte character to wide character. + Copyright (C) 2011-2012 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2011. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +/* We don't need a static internal state, because the encoding is not state + dependent, and when mbrtowc returns (size_t)(-2). we throw the result + away. */ + +int +mbtowc (wchar_t *pwc, const char *s, size_t n) +{ + if (s == NULL) + return 0; + else + { + mbstate_t state; + wchar_t wc; + size_t result; + + memset (&state, 0, sizeof (mbstate_t)); + result = mbrtowc (&wc, s, n, &state); + if (result == (size_t)-1 || result == (size_t)-2) + { + errno = EILSEQ; + return -1; + } + if (pwc != NULL) + *pwc = wc; + return (wc == 0 ? 0 : result); + } +} diff --git a/lib/mbtowc.c b/lib/mbtowc.c new file mode 100644 index 0000000..97eab4e --- /dev/null +++ b/lib/mbtowc.c @@ -0,0 +1,26 @@ +/* Convert multibyte character to wide character. + Copyright (C) 2011-2012 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2011. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +#include <stdlib.h> + +#include <errno.h> +#include <string.h> +#include <wchar.h> + +#include "mbtowc-impl.h" diff --git a/lib/msvc-inval.c b/lib/msvc-inval.c new file mode 100644 index 0000000..ba76a7e --- /dev/null +++ b/lib/msvc-inval.c @@ -0,0 +1,129 @@ +/* Invalid parameter handler for MSVC runtime libraries. + Copyright (C) 2011-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +/* Specification. */ +#include "msvc-inval.h" + +#if HAVE_MSVC_INVALID_PARAMETER_HANDLER \ + && !(MSVC_INVALID_PARAMETER_HANDLING == SANE_LIBRARY_HANDLING) + +/* Get _invalid_parameter_handler type and _set_invalid_parameter_handler + declaration. */ +# include <stdlib.h> + +# if MSVC_INVALID_PARAMETER_HANDLING == DEFAULT_HANDLING + +static void cdecl +gl_msvc_invalid_parameter_handler (const wchar_t *expression, + const wchar_t *function, + const wchar_t *file, + unsigned int line, + uintptr_t dummy) +{ +} + +# else + +/* Get declarations of the native Windows API functions. */ +# define WIN32_LEAN_AND_MEAN +# include <windows.h> + +# if defined _MSC_VER + +static void cdecl +gl_msvc_invalid_parameter_handler (const wchar_t *expression, + const wchar_t *function, + const wchar_t *file, + unsigned int line, + uintptr_t dummy) +{ + RaiseException (STATUS_GNULIB_INVALID_PARAMETER, 0, 0, NULL); +} + +# else + +/* An index to thread-local storage. */ +static DWORD tls_index; +static int tls_initialized /* = 0 */; + +/* Used as a fallback only. */ +static struct gl_msvc_inval_per_thread not_per_thread; + +struct gl_msvc_inval_per_thread * +gl_msvc_inval_current (void) +{ + if (!tls_initialized) + { + tls_index = TlsAlloc (); + tls_initialized = 1; + } + if (tls_index == TLS_OUT_OF_INDEXES) + /* TlsAlloc had failed. */ + return ¬_per_thread; + else + { + struct gl_msvc_inval_per_thread *pointer = + (struct gl_msvc_inval_per_thread *) TlsGetValue (tls_index); + if (pointer == NULL) + { + /* First call. Allocate a new 'struct gl_msvc_inval_per_thread'. */ + pointer = + (struct gl_msvc_inval_per_thread *) + malloc (sizeof (struct gl_msvc_inval_per_thread)); + if (pointer == NULL) + /* Could not allocate memory. Use the global storage. */ + pointer = ¬_per_thread; + TlsSetValue (tls_index, pointer); + } + return pointer; + } +} + +static void cdecl +gl_msvc_invalid_parameter_handler (const wchar_t *expression, + const wchar_t *function, + const wchar_t *file, + unsigned int line, + uintptr_t dummy) +{ + struct gl_msvc_inval_per_thread *current = gl_msvc_inval_current (); + if (current->restart_valid) + longjmp (current->restart, 1); + else + /* An invalid parameter notification from outside the gnulib code. + Give the caller a chance to intervene. */ + RaiseException (STATUS_GNULIB_INVALID_PARAMETER, 0, 0, NULL); +} + +# endif + +# endif + +static int gl_msvc_inval_initialized /* = 0 */; + +void +gl_msvc_inval_ensure_handler (void) +{ + if (gl_msvc_inval_initialized == 0) + { + _set_invalid_parameter_handler (gl_msvc_invalid_parameter_handler); + gl_msvc_inval_initialized = 1; + } +} + +#endif diff --git a/lib/msvc-inval.h b/lib/msvc-inval.h new file mode 100644 index 0000000..eb6930b --- /dev/null +++ b/lib/msvc-inval.h @@ -0,0 +1,222 @@ +/* Invalid parameter handler for MSVC runtime libraries. + Copyright (C) 2011-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#ifndef _MSVC_INVAL_H +#define _MSVC_INVAL_H + +/* With MSVC runtime libraries with the "invalid parameter handler" concept, + functions like fprintf(), dup2(), or close() crash when the caller passes + an invalid argument. But POSIX wants error codes (such as EINVAL or EBADF) + instead. + This file defines macros that turn such an invalid parameter notification + into a non-local exit. An error code can then be produced at the target + of this exit. You can thus write code like + + TRY_MSVC_INVAL + { + <Code that can trigger an invalid parameter notification + but does not do 'return', 'break', 'continue', nor 'goto'.> + } + CATCH_MSVC_INVAL + { + <Code that handles an invalid parameter notification + but does not do 'return', 'break', 'continue', nor 'goto'.> + } + DONE_MSVC_INVAL; + + This entire block expands to a single statement. + + The handling of invalid parameters can be done in three ways: + + * The default way, which is reasonable for programs (not libraries): + AC_DEFINE([MSVC_INVALID_PARAMETER_HANDLING], [DEFAULT_HANDLING]) + + * The way for libraries that make "hairy" calls (like close(-1), or + fclose(fp) where fileno(fp) is closed, or simply getdtablesize()): + AC_DEFINE([MSVC_INVALID_PARAMETER_HANDLING], [HAIRY_LIBRARY_HANDLING]) + + * The way for libraries that make no "hairy" calls: + AC_DEFINE([MSVC_INVALID_PARAMETER_HANDLING], [SANE_LIBRARY_HANDLING]) + */ + +#define DEFAULT_HANDLING 0 +#define HAIRY_LIBRARY_HANDLING 1 +#define SANE_LIBRARY_HANDLING 2 + +#if HAVE_MSVC_INVALID_PARAMETER_HANDLER \ + && !(MSVC_INVALID_PARAMETER_HANDLING == SANE_LIBRARY_HANDLING) +/* A native Windows platform with the "invalid parameter handler" concept, + and either DEFAULT_HANDLING or HAIRY_LIBRARY_HANDLING. */ + +# if MSVC_INVALID_PARAMETER_HANDLING == DEFAULT_HANDLING +/* Default handling. */ + +# ifdef __cplusplus +extern "C" { +# endif + +/* Ensure that the invalid parameter handler in installed that just returns. + Because we assume no other part of the program installs a different + invalid parameter handler, this solution is multithread-safe. */ +extern void gl_msvc_inval_ensure_handler (void); + +# ifdef __cplusplus +} +# endif + +# define TRY_MSVC_INVAL \ + do \ + { \ + gl_msvc_inval_ensure_handler (); \ + if (1) +# define CATCH_MSVC_INVAL \ + else +# define DONE_MSVC_INVAL \ + } \ + while (0) + +# else +/* Handling for hairy libraries. */ + +# include <excpt.h> + +/* Gnulib can define its own status codes, as described in the page + "Raising Software Exceptions" on microsoft.com + <http://msdn.microsoft.com/en-us/library/het71c37.aspx>. + Our status codes are composed of + - 0xE0000000, mandatory for all user-defined status codes, + - 0x474E550, a API identifier ("GNU"), + - 0, 1, 2, ..., used to distinguish different status codes from the + same API. */ +# define STATUS_GNULIB_INVALID_PARAMETER (0xE0000000 + 0x474E550 + 0) + +# if defined _MSC_VER +/* A compiler that supports __try/__except, as described in the page + "try-except statement" on microsoft.com + <http://msdn.microsoft.com/en-us/library/s58ftw19.aspx>. + With __try/__except, we can use the multithread-safe exception handling. */ + +# ifdef __cplusplus +extern "C" { +# endif + +/* Ensure that the invalid parameter handler in installed that raises a + software exception with code STATUS_GNULIB_INVALID_PARAMETER. + Because we assume no other part of the program installs a different + invalid parameter handler, this solution is multithread-safe. */ +extern void gl_msvc_inval_ensure_handler (void); + +# ifdef __cplusplus +} +# endif + +# define TRY_MSVC_INVAL \ + do \ + { \ + gl_msvc_inval_ensure_handler (); \ + __try +# define CATCH_MSVC_INVAL \ + __except (GetExceptionCode () == STATUS_GNULIB_INVALID_PARAMETER \ + ? EXCEPTION_EXECUTE_HANDLER \ + : EXCEPTION_CONTINUE_SEARCH) +# define DONE_MSVC_INVAL \ + } \ + while (0) + +# else +/* Any compiler. + We can only use setjmp/longjmp. */ + +# include <setjmp.h> + +# ifdef __cplusplus +extern "C" { +# endif + +struct gl_msvc_inval_per_thread +{ + /* The restart that will resume execution at the code between + CATCH_MSVC_INVAL and DONE_MSVC_INVAL. It is enabled only between + TRY_MSVC_INVAL and CATCH_MSVC_INVAL. */ + jmp_buf restart; + + /* Tells whether the contents of restart is valid. */ + int restart_valid; +}; + +/* Ensure that the invalid parameter handler in installed that passes + control to the gl_msvc_inval_restart if it is valid, or raises a + software exception with code STATUS_GNULIB_INVALID_PARAMETER otherwise. + Because we assume no other part of the program installs a different + invalid parameter handler, this solution is multithread-safe. */ +extern void gl_msvc_inval_ensure_handler (void); + +/* Return a pointer to the per-thread data for the current thread. */ +extern struct gl_msvc_inval_per_thread *gl_msvc_inval_current (void); + +# ifdef __cplusplus +} +# endif + +# define TRY_MSVC_INVAL \ + do \ + { \ + struct gl_msvc_inval_per_thread *msvc_inval_current; \ + gl_msvc_inval_ensure_handler (); \ + msvc_inval_current = gl_msvc_inval_current (); \ + /* First, initialize gl_msvc_inval_restart. */ \ + if (setjmp (msvc_inval_current->restart) == 0) \ + { \ + /* Then, mark it as valid. */ \ + msvc_inval_current->restart_valid = 1; +# define CATCH_MSVC_INVAL \ + /* Execution completed. \ + Mark gl_msvc_inval_restart as invalid. */ \ + msvc_inval_current->restart_valid = 0; \ + } \ + else \ + { \ + /* Execution triggered an invalid parameter notification. \ + Mark gl_msvc_inval_restart as invalid. */ \ + msvc_inval_current->restart_valid = 0; +# define DONE_MSVC_INVAL \ + } \ + } \ + while (0) + +# endif + +# endif + +#else +/* A platform that does not need to the invalid parameter handler, + or when SANE_LIBRARY_HANDLING is desired. */ + +/* The braces here avoid GCC warnings like + "warning: suggest explicit braces to avoid ambiguous 'else'". */ +# define TRY_MSVC_INVAL \ + do \ + { \ + if (1) +# define CATCH_MSVC_INVAL \ + else +# define DONE_MSVC_INVAL \ + } \ + while (0) + +#endif + +#endif /* _MSVC_INVAL_H */ diff --git a/lib/msvc-nothrow.c b/lib/msvc-nothrow.c new file mode 100644 index 0000000..e5cf181 --- /dev/null +++ b/lib/msvc-nothrow.c @@ -0,0 +1,49 @@ +/* Wrappers that don't throw invalid parameter notifications + with MSVC runtime libraries. + Copyright (C) 2011-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +/* Specification. */ +#include "msvc-nothrow.h" + +/* Get declarations of the native Windows API functions. */ +#define WIN32_LEAN_AND_MEAN +#include <windows.h> + +#include "msvc-inval.h" + +#undef _get_osfhandle + +#if HAVE_MSVC_INVALID_PARAMETER_HANDLER +intptr_t +_gl_nothrow_get_osfhandle (int fd) +{ + intptr_t result; + + TRY_MSVC_INVAL + { + result = _get_osfhandle (fd); + } + CATCH_MSVC_INVAL + { + result = (intptr_t) INVALID_HANDLE_VALUE; + } + DONE_MSVC_INVAL; + + return result; +} +#endif diff --git a/lib/msvc-nothrow.h b/lib/msvc-nothrow.h new file mode 100644 index 0000000..2b71945 --- /dev/null +++ b/lib/msvc-nothrow.h @@ -0,0 +1,43 @@ +/* Wrappers that don't throw invalid parameter notifications + with MSVC runtime libraries. + Copyright (C) 2011-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#ifndef _MSVC_NOTHROW_H +#define _MSVC_NOTHROW_H + +/* With MSVC runtime libraries with the "invalid parameter handler" concept, + functions like fprintf(), dup2(), or close() crash when the caller passes + an invalid argument. But POSIX wants error codes (such as EINVAL or EBADF) + instead. + This file defines wrappers that turn such an invalid parameter notification + into an error code. */ + +#if (defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__ + +/* Get original declaration of _get_osfhandle. */ +# include <io.h> + +# if HAVE_MSVC_INVALID_PARAMETER_HANDLER + +/* Override _get_osfhandle. */ +extern intptr_t _gl_nothrow_get_osfhandle (int fd); +# define _get_osfhandle _gl_nothrow_get_osfhandle + +# endif + +#endif + +#endif /* _MSVC_NOTHROW_H */ diff --git a/lib/nl_langinfo.c b/lib/nl_langinfo.c new file mode 100644 index 0000000..c73c769 --- /dev/null +++ b/lib/nl_langinfo.c @@ -0,0 +1,271 @@ +/* nl_langinfo() replacement: query locale dependent information. + + Copyright (C) 2007-2012 Free Software Foundation, Inc. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +/* Specification. */ +#include <langinfo.h> + +#if REPLACE_NL_LANGINFO + +/* Override nl_langinfo with support for added nl_item values. */ + +# include <locale.h> +# include <string.h> + +# undef nl_langinfo + +char * +rpl_nl_langinfo (nl_item item) +{ + switch (item) + { +# if GNULIB_defined_CODESET + case CODESET: + { + const char *locale; + static char buf[2 + 10 + 1]; + + locale = setlocale (LC_CTYPE, NULL); + if (locale != NULL && locale[0] != '\0') + { + /* If the locale name contains an encoding after the dot, return + it. */ + const char *dot = strchr (locale, '.'); + + if (dot != NULL) + { + const char *modifier; + + dot++; + /* Look for the possible @... trailer and remove it, if any. */ + modifier = strchr (dot, '@'); + if (modifier == NULL) + return dot; + if (modifier - dot < sizeof (buf)) + { + memcpy (buf, dot, modifier - dot); + buf [modifier - dot] = '\0'; + return buf; + } + } + } + return ""; + } +# endif +# if GNULIB_defined_T_FMT_AMPM + case T_FMT_AMPM: + return "%I:%M:%S %p"; +# endif +# if GNULIB_defined_ERA + case ERA: + /* The format is not standardized. In glibc it is a sequence of strings + of the form "direction:offset:start_date:end_date:era_name:era_format" + with an empty string at the end. */ + return ""; + case ERA_D_FMT: + /* The %Ex conversion in strftime behaves like %x if the locale does not + have an alternative time format. */ + item = D_FMT; + break; + case ERA_D_T_FMT: + /* The %Ec conversion in strftime behaves like %c if the locale does not + have an alternative time format. */ + item = D_T_FMT; + break; + case ERA_T_FMT: + /* The %EX conversion in strftime behaves like %X if the locale does not + have an alternative time format. */ + item = T_FMT; + break; + case ALT_DIGITS: + /* The format is not standardized. In glibc it is a sequence of 10 + strings, appended in memory. */ + return "\0\0\0\0\0\0\0\0\0\0"; +# endif +# if GNULIB_defined_YESEXPR || !FUNC_NL_LANGINFO_YESEXPR_WORKS + case YESEXPR: + return "^[yY]"; + case NOEXPR: + return "^[nN]"; +# endif + default: + break; + } + return nl_langinfo (item); +} + +#else + +/* Provide nl_langinfo from scratch. */ + +# if (defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__ + +/* Native Windows platforms. */ + +# define WIN32_LEAN_AND_MEAN /* avoid including junk */ +# include <windows.h> + +# include <stdio.h> + +# else + +/* An old Unix platform without locales, such as Linux libc5 or BeOS. */ + +# endif + +# include <locale.h> + +char * +nl_langinfo (nl_item item) +{ + switch (item) + { + /* nl_langinfo items of the LC_CTYPE category */ + case CODESET: +# if (defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__ + { + static char buf[2 + 10 + 1]; + + /* The Windows API has a function returning the locale's codepage as + a number. */ + sprintf (buf, "CP%u", GetACP ()); + return buf; + } +# elif defined __BEOS__ + return "UTF-8"; +# else + return "ISO-8859-1"; +# endif + /* nl_langinfo items of the LC_NUMERIC category */ + case RADIXCHAR: + return localeconv () ->decimal_point; + case THOUSEP: + return localeconv () ->thousands_sep; + /* nl_langinfo items of the LC_TIME category. + TODO: Really use the locale. */ + case D_T_FMT: + case ERA_D_T_FMT: + return "%a %b %e %H:%M:%S %Y"; + case D_FMT: + case ERA_D_FMT: + return "%m/%d/%y"; + case T_FMT: + case ERA_T_FMT: + return "%H:%M:%S"; + case T_FMT_AMPM: + return "%I:%M:%S %p"; + case AM_STR: + return "AM"; + case PM_STR: + return "PM"; + case DAY_1: + return "Sunday"; + case DAY_2: + return "Monday"; + case DAY_3: + return "Tuesday"; + case DAY_4: + return "Wednesday"; + case DAY_5: + return "Thursday"; + case DAY_6: + return "Friday"; + case DAY_7: + return "Saturday"; + case ABDAY_1: + return "Sun"; + case ABDAY_2: + return "Mon"; + case ABDAY_3: + return "Tue"; + case ABDAY_4: + return "Wed"; + case ABDAY_5: + return "Thu"; + case ABDAY_6: + return "Fri"; + case ABDAY_7: + return "Sat"; + case MON_1: + return "January"; + case MON_2: + return "February"; + case MON_3: + return "March"; + case MON_4: + return "April"; + case MON_5: + return "May"; + case MON_6: + return "June"; + case MON_7: + return "July"; + case MON_8: + return "August"; + case MON_9: + return "September"; + case MON_10: + return "October"; + case MON_11: + return "November"; + case MON_12: + return "December"; + case ABMON_1: + return "Jan"; + case ABMON_2: + return "Feb"; + case ABMON_3: + return "Mar"; + case ABMON_4: + return "Apr"; + case ABMON_5: + return "May"; + case ABMON_6: + return "Jun"; + case ABMON_7: + return "Jul"; + case ABMON_8: + return "Aug"; + case ABMON_9: + return "Sep"; + case ABMON_10: + return "Oct"; + case ABMON_11: + return "Nov"; + case ABMON_12: + return "Dec"; + case ERA: + return ""; + case ALT_DIGITS: + return "\0\0\0\0\0\0\0\0\0\0"; + /* nl_langinfo items of the LC_MONETARY category + TODO: Really use the locale. */ + case CRNCYSTR: + return "-"; + /* nl_langinfo items of the LC_MESSAGES category + TODO: Really use the locale. */ + case YESEXPR: + return "^[yY]"; + case NOEXPR: + return "^[nN]"; + default: + return ""; + } +} + +#endif diff --git a/lib/ref-add.sin b/lib/ref-add.sin new file mode 100644 index 0000000..4ec043f --- /dev/null +++ b/lib/ref-add.sin @@ -0,0 +1,29 @@ +# Add this package to a list of references stored in a text file. +# +# Copyright (C) 2000, 2009-2012 Free Software Foundation, Inc. +# +# This program 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, or (at your option) +# any later version. +# +# This program 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 this program; if not, see <http://www.gnu.org/licenses/>. +# +# Written by Bruno Haible <haible@clisp.cons.org>. +# +/^# Packages using this file: / { + s/# Packages using this file:// + ta + :a + s/ @PACKAGE@ / @PACKAGE@ / + tb + s/ $/ @PACKAGE@ / + :b + s/^/# Packages using this file:/ +} diff --git a/lib/ref-del.sin b/lib/ref-del.sin new file mode 100644 index 0000000..eb03881 --- /dev/null +++ b/lib/ref-del.sin @@ -0,0 +1,24 @@ +# Remove this package from a list of references stored in a text file. +# +# Copyright (C) 2000, 2009-2012 Free Software Foundation, Inc. +# +# This program 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, or (at your option) +# any later version. +# +# This program 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 this program; if not, see <http://www.gnu.org/licenses/>. +# +# Written by Bruno Haible <haible@clisp.cons.org>. +# +/^# Packages using this file: / { + s/# Packages using this file:// + s/ @PACKAGE@ / / + s/^/# Packages using this file:/ +} diff --git a/lib/regcomp.c b/lib/regcomp.c new file mode 100644 index 0000000..033eb80 --- /dev/null +++ b/lib/regcomp.c @@ -0,0 +1,3930 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002-2012 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +static reg_errcode_t re_compile_internal (regex_t *preg, const char * pattern, + size_t length, reg_syntax_t syntax); +static void re_compile_fastmap_iter (regex_t *bufp, + const re_dfastate_t *init_state, + char *fastmap); +static reg_errcode_t init_dfa (re_dfa_t *dfa, size_t pat_len); +#ifdef RE_ENABLE_I18N +static void free_charset (re_charset_t *cset); +#endif /* RE_ENABLE_I18N */ +static void free_workarea_compile (regex_t *preg); +static reg_errcode_t create_initial_state (re_dfa_t *dfa); +#ifdef RE_ENABLE_I18N +static void optimize_utf8 (re_dfa_t *dfa); +#endif +static reg_errcode_t analyze (regex_t *preg); +static reg_errcode_t preorder (bin_tree_t *root, + reg_errcode_t (fn (void *, bin_tree_t *)), + void *extra); +static reg_errcode_t postorder (bin_tree_t *root, + reg_errcode_t (fn (void *, bin_tree_t *)), + void *extra); +static reg_errcode_t optimize_subexps (void *extra, bin_tree_t *node); +static reg_errcode_t lower_subexps (void *extra, bin_tree_t *node); +static bin_tree_t *lower_subexp (reg_errcode_t *err, regex_t *preg, + bin_tree_t *node); +static reg_errcode_t calc_first (void *extra, bin_tree_t *node); +static reg_errcode_t calc_next (void *extra, bin_tree_t *node); +static reg_errcode_t link_nfa_nodes (void *extra, bin_tree_t *node); +static Idx duplicate_node (re_dfa_t *dfa, Idx org_idx, unsigned int constraint); +static Idx search_duplicated_node (const re_dfa_t *dfa, Idx org_node, + unsigned int constraint); +static reg_errcode_t calc_eclosure (re_dfa_t *dfa); +static reg_errcode_t calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, + Idx node, bool root); +static reg_errcode_t calc_inveclosure (re_dfa_t *dfa); +static Idx fetch_number (re_string_t *input, re_token_t *token, + reg_syntax_t syntax); +static int peek_token (re_token_t *token, re_string_t *input, + reg_syntax_t syntax) internal_function; +static bin_tree_t *parse (re_string_t *regexp, regex_t *preg, + reg_syntax_t syntax, reg_errcode_t *err); +static bin_tree_t *parse_reg_exp (re_string_t *regexp, regex_t *preg, + re_token_t *token, reg_syntax_t syntax, + Idx nest, reg_errcode_t *err); +static bin_tree_t *parse_branch (re_string_t *regexp, regex_t *preg, + re_token_t *token, reg_syntax_t syntax, + Idx nest, reg_errcode_t *err); +static bin_tree_t *parse_expression (re_string_t *regexp, regex_t *preg, + re_token_t *token, reg_syntax_t syntax, + Idx nest, reg_errcode_t *err); +static bin_tree_t *parse_sub_exp (re_string_t *regexp, regex_t *preg, + re_token_t *token, reg_syntax_t syntax, + Idx nest, reg_errcode_t *err); +static bin_tree_t *parse_dup_op (bin_tree_t *dup_elem, re_string_t *regexp, + re_dfa_t *dfa, re_token_t *token, + reg_syntax_t syntax, reg_errcode_t *err); +static bin_tree_t *parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, + re_token_t *token, reg_syntax_t syntax, + reg_errcode_t *err); +static reg_errcode_t parse_bracket_element (bracket_elem_t *elem, + re_string_t *regexp, + re_token_t *token, int token_len, + re_dfa_t *dfa, + reg_syntax_t syntax, + bool accept_hyphen); +static reg_errcode_t parse_bracket_symbol (bracket_elem_t *elem, + re_string_t *regexp, + re_token_t *token); +#ifdef RE_ENABLE_I18N +static reg_errcode_t build_equiv_class (bitset_t sbcset, + re_charset_t *mbcset, + Idx *equiv_class_alloc, + const unsigned char *name); +static reg_errcode_t build_charclass (RE_TRANSLATE_TYPE trans, + bitset_t sbcset, + re_charset_t *mbcset, + Idx *char_class_alloc, + const unsigned char *class_name, + reg_syntax_t syntax); +#else /* not RE_ENABLE_I18N */ +static reg_errcode_t build_equiv_class (bitset_t sbcset, + const unsigned char *name); +static reg_errcode_t build_charclass (RE_TRANSLATE_TYPE trans, + bitset_t sbcset, + const unsigned char *class_name, + reg_syntax_t syntax); +#endif /* not RE_ENABLE_I18N */ +static bin_tree_t *build_charclass_op (re_dfa_t *dfa, + RE_TRANSLATE_TYPE trans, + const unsigned char *class_name, + const unsigned char *extra, + bool non_match, reg_errcode_t *err); +static bin_tree_t *create_tree (re_dfa_t *dfa, + bin_tree_t *left, bin_tree_t *right, + re_token_type_t type); +static bin_tree_t *create_token_tree (re_dfa_t *dfa, + bin_tree_t *left, bin_tree_t *right, + const re_token_t *token); +static bin_tree_t *duplicate_tree (const bin_tree_t *src, re_dfa_t *dfa); +static void free_token (re_token_t *node); +static reg_errcode_t free_tree (void *extra, bin_tree_t *node); +static reg_errcode_t mark_opt_subexp (void *extra, bin_tree_t *node); + +/* This table gives an error message for each of the error codes listed + in regex.h. Obviously the order here has to be same as there. + POSIX doesn't require that we do anything for REG_NOERROR, + but why not be nice? */ + +static const char __re_error_msgid[] = + { +#define REG_NOERROR_IDX 0 + gettext_noop ("Success") /* REG_NOERROR */ + "\0" +#define REG_NOMATCH_IDX (REG_NOERROR_IDX + sizeof "Success") + gettext_noop ("No match") /* REG_NOMATCH */ + "\0" +#define REG_BADPAT_IDX (REG_NOMATCH_IDX + sizeof "No match") + gettext_noop ("Invalid regular expression") /* REG_BADPAT */ + "\0" +#define REG_ECOLLATE_IDX (REG_BADPAT_IDX + sizeof "Invalid regular expression") + gettext_noop ("Invalid collation character") /* REG_ECOLLATE */ + "\0" +#define REG_ECTYPE_IDX (REG_ECOLLATE_IDX + sizeof "Invalid collation character") + gettext_noop ("Invalid character class name") /* REG_ECTYPE */ + "\0" +#define REG_EESCAPE_IDX (REG_ECTYPE_IDX + sizeof "Invalid character class name") + gettext_noop ("Trailing backslash") /* REG_EESCAPE */ + "\0" +#define REG_ESUBREG_IDX (REG_EESCAPE_IDX + sizeof "Trailing backslash") + gettext_noop ("Invalid back reference") /* REG_ESUBREG */ + "\0" +#define REG_EBRACK_IDX (REG_ESUBREG_IDX + sizeof "Invalid back reference") + gettext_noop ("Unmatched [ or [^") /* REG_EBRACK */ + "\0" +#define REG_EPAREN_IDX (REG_EBRACK_IDX + sizeof "Unmatched [ or [^") + gettext_noop ("Unmatched ( or \\(") /* REG_EPAREN */ + "\0" +#define REG_EBRACE_IDX (REG_EPAREN_IDX + sizeof "Unmatched ( or \\(") + gettext_noop ("Unmatched \\{") /* REG_EBRACE */ + "\0" +#define REG_BADBR_IDX (REG_EBRACE_IDX + sizeof "Unmatched \\{") + gettext_noop ("Invalid content of \\{\\}") /* REG_BADBR */ + "\0" +#define REG_ERANGE_IDX (REG_BADBR_IDX + sizeof "Invalid content of \\{\\}") + gettext_noop ("Invalid range end") /* REG_ERANGE */ + "\0" +#define REG_ESPACE_IDX (REG_ERANGE_IDX + sizeof "Invalid range end") + gettext_noop ("Memory exhausted") /* REG_ESPACE */ + "\0" +#define REG_BADRPT_IDX (REG_ESPACE_IDX + sizeof "Memory exhausted") + gettext_noop ("Invalid preceding regular expression") /* REG_BADRPT */ + "\0" +#define REG_EEND_IDX (REG_BADRPT_IDX + sizeof "Invalid preceding regular expression") + gettext_noop ("Premature end of regular expression") /* REG_EEND */ + "\0" +#define REG_ESIZE_IDX (REG_EEND_IDX + sizeof "Premature end of regular expression") + gettext_noop ("Regular expression too big") /* REG_ESIZE */ + "\0" +#define REG_ERPAREN_IDX (REG_ESIZE_IDX + sizeof "Regular expression too big") + gettext_noop ("Unmatched ) or \\)") /* REG_ERPAREN */ + }; + +static const size_t __re_error_msgid_idx[] = + { + REG_NOERROR_IDX, + REG_NOMATCH_IDX, + REG_BADPAT_IDX, + REG_ECOLLATE_IDX, + REG_ECTYPE_IDX, + REG_EESCAPE_IDX, + REG_ESUBREG_IDX, + REG_EBRACK_IDX, + REG_EPAREN_IDX, + REG_EBRACE_IDX, + REG_BADBR_IDX, + REG_ERANGE_IDX, + REG_ESPACE_IDX, + REG_BADRPT_IDX, + REG_EEND_IDX, + REG_ESIZE_IDX, + REG_ERPAREN_IDX + }; + +/* Entry points for GNU code. */ + +/* re_compile_pattern is the GNU regular expression compiler: it + compiles PATTERN (of length LENGTH) and puts the result in BUFP. + Returns 0 if the pattern was valid, otherwise an error string. + + Assumes the 'allocated' (and perhaps 'buffer') and 'translate' fields + are set in BUFP on entry. */ + +#ifdef _LIBC +const char * +re_compile_pattern (pattern, length, bufp) + const char *pattern; + size_t length; + struct re_pattern_buffer *bufp; +#else /* size_t might promote */ +const char * +re_compile_pattern (const char *pattern, size_t length, + struct re_pattern_buffer *bufp) +#endif +{ + reg_errcode_t ret; + + /* And GNU code determines whether or not to get register information + by passing null for the REGS argument to re_match, etc., not by + setting no_sub, unless RE_NO_SUB is set. */ + bufp->no_sub = !!(re_syntax_options & RE_NO_SUB); + + /* Match anchors at newline. */ + bufp->newline_anchor = 1; + + ret = re_compile_internal (bufp, pattern, length, re_syntax_options); + + if (!ret) + return NULL; + return gettext (__re_error_msgid + __re_error_msgid_idx[(int) ret]); +} +#ifdef _LIBC +weak_alias (__re_compile_pattern, re_compile_pattern) +#endif + +/* Set by 're_set_syntax' to the current regexp syntax to recognize. Can + also be assigned to arbitrarily: each pattern buffer stores its own + syntax, so it can be changed between regex compilations. */ +/* This has no initializer because initialized variables in Emacs + become read-only after dumping. */ +reg_syntax_t re_syntax_options; + + +/* Specify the precise syntax of regexps for compilation. This provides + for compatibility for various utilities which historically have + different, incompatible syntaxes. + + The argument SYNTAX is a bit mask comprised of the various bits + defined in regex.h. We return the old syntax. */ + +reg_syntax_t +re_set_syntax (syntax) + reg_syntax_t syntax; +{ + reg_syntax_t ret = re_syntax_options; + + re_syntax_options = syntax; + return ret; +} +#ifdef _LIBC +weak_alias (__re_set_syntax, re_set_syntax) +#endif + +int +re_compile_fastmap (bufp) + struct re_pattern_buffer *bufp; +{ + re_dfa_t *dfa = (re_dfa_t *) bufp->buffer; + char *fastmap = bufp->fastmap; + + memset (fastmap, '\0', sizeof (char) * SBC_MAX); + re_compile_fastmap_iter (bufp, dfa->init_state, fastmap); + if (dfa->init_state != dfa->init_state_word) + re_compile_fastmap_iter (bufp, dfa->init_state_word, fastmap); + if (dfa->init_state != dfa->init_state_nl) + re_compile_fastmap_iter (bufp, dfa->init_state_nl, fastmap); + if (dfa->init_state != dfa->init_state_begbuf) + re_compile_fastmap_iter (bufp, dfa->init_state_begbuf, fastmap); + bufp->fastmap_accurate = 1; + return 0; +} +#ifdef _LIBC +weak_alias (__re_compile_fastmap, re_compile_fastmap) +#endif + +static inline void +__attribute ((always_inline)) +re_set_fastmap (char *fastmap, bool icase, int ch) +{ + fastmap[ch] = 1; + if (icase) + fastmap[tolower (ch)] = 1; +} + +/* Helper function for re_compile_fastmap. + Compile fastmap for the initial_state INIT_STATE. */ + +static void +re_compile_fastmap_iter (regex_t *bufp, const re_dfastate_t *init_state, + char *fastmap) +{ + re_dfa_t *dfa = (re_dfa_t *) bufp->buffer; + Idx node_cnt; + bool icase = (dfa->mb_cur_max == 1 && (bufp->syntax & RE_ICASE)); + for (node_cnt = 0; node_cnt < init_state->nodes.nelem; ++node_cnt) + { + Idx node = init_state->nodes.elems[node_cnt]; + re_token_type_t type = dfa->nodes[node].type; + + if (type == CHARACTER) + { + re_set_fastmap (fastmap, icase, dfa->nodes[node].opr.c); +#ifdef RE_ENABLE_I18N + if ((bufp->syntax & RE_ICASE) && dfa->mb_cur_max > 1) + { + unsigned char buf[MB_LEN_MAX]; + unsigned char *p; + wchar_t wc; + mbstate_t state; + + p = buf; + *p++ = dfa->nodes[node].opr.c; + while (++node < dfa->nodes_len + && dfa->nodes[node].type == CHARACTER + && dfa->nodes[node].mb_partial) + *p++ = dfa->nodes[node].opr.c; + memset (&state, '\0', sizeof (state)); + if (__mbrtowc (&wc, (const char *) buf, p - buf, + &state) == p - buf + && (__wcrtomb ((char *) buf, towlower (wc), &state) + != (size_t) -1)) + re_set_fastmap (fastmap, false, buf[0]); + } +#endif + } + else if (type == SIMPLE_BRACKET) + { + int i, ch; + for (i = 0, ch = 0; i < BITSET_WORDS; ++i) + { + int j; + bitset_word_t w = dfa->nodes[node].opr.sbcset[i]; + for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch) + if (w & ((bitset_word_t) 1 << j)) + re_set_fastmap (fastmap, icase, ch); + } + } +#ifdef RE_ENABLE_I18N + else if (type == COMPLEX_BRACKET) + { + re_charset_t *cset = dfa->nodes[node].opr.mbcset; + Idx i; + +# ifdef _LIBC + /* See if we have to try all bytes which start multiple collation + elements. + e.g. In da_DK, we want to catch 'a' since "aa" is a valid + collation element, and don't catch 'b' since 'b' is + the only collation element which starts from 'b' (and + it is caught by SIMPLE_BRACKET). */ + if (_NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES) != 0 + && (cset->ncoll_syms || cset->nranges)) + { + const int32_t *table = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB); + for (i = 0; i < SBC_MAX; ++i) + if (table[i] < 0) + re_set_fastmap (fastmap, icase, i); + } +# endif /* _LIBC */ + + /* See if we have to start the match at all multibyte characters, + i.e. where we would not find an invalid sequence. This only + applies to multibyte character sets; for single byte character + sets, the SIMPLE_BRACKET again suffices. */ + if (dfa->mb_cur_max > 1 + && (cset->nchar_classes || cset->non_match || cset->nranges +# ifdef _LIBC + || cset->nequiv_classes +# endif /* _LIBC */ + )) + { + unsigned char c = 0; + do + { + mbstate_t mbs; + memset (&mbs, 0, sizeof (mbs)); + if (__mbrtowc (NULL, (char *) &c, 1, &mbs) == (size_t) -2) + re_set_fastmap (fastmap, false, (int) c); + } + while (++c != 0); + } + + else + { + /* ... Else catch all bytes which can start the mbchars. */ + for (i = 0; i < cset->nmbchars; ++i) + { + char buf[256]; + mbstate_t state; + memset (&state, '\0', sizeof (state)); + if (__wcrtomb (buf, cset->mbchars[i], &state) != (size_t) -1) + re_set_fastmap (fastmap, icase, *(unsigned char *) buf); + if ((bufp->syntax & RE_ICASE) && dfa->mb_cur_max > 1) + { + if (__wcrtomb (buf, towlower (cset->mbchars[i]), &state) + != (size_t) -1) + re_set_fastmap (fastmap, false, *(unsigned char *) buf); + } + } + } + } +#endif /* RE_ENABLE_I18N */ + else if (type == OP_PERIOD +#ifdef RE_ENABLE_I18N + || type == OP_UTF8_PERIOD +#endif /* RE_ENABLE_I18N */ + || type == END_OF_RE) + { + memset (fastmap, '\1', sizeof (char) * SBC_MAX); + if (type == END_OF_RE) + bufp->can_be_null = 1; + return; + } + } +} + +/* Entry point for POSIX code. */ +/* regcomp takes a regular expression as a string and compiles it. + + PREG is a regex_t *. We do not expect any fields to be initialized, + since POSIX says we shouldn't. Thus, we set + + 'buffer' to the compiled pattern; + 'used' to the length of the compiled pattern; + 'syntax' to RE_SYNTAX_POSIX_EXTENDED if the + REG_EXTENDED bit in CFLAGS is set; otherwise, to + RE_SYNTAX_POSIX_BASIC; + 'newline_anchor' to REG_NEWLINE being set in CFLAGS; + 'fastmap' to an allocated space for the fastmap; + 'fastmap_accurate' to zero; + 're_nsub' to the number of subexpressions in PATTERN. + + PATTERN is the address of the pattern string. + + CFLAGS is a series of bits which affect compilation. + + If REG_EXTENDED is set, we use POSIX extended syntax; otherwise, we + use POSIX basic syntax. + + If REG_NEWLINE is set, then . and [^...] don't match newline. + Also, regexec will try a match beginning after every newline. + + If REG_ICASE is set, then we considers upper- and lowercase + versions of letters to be equivalent when matching. + + If REG_NOSUB is set, then when PREG is passed to regexec, that + routine will report only success or failure, and nothing about the + registers. + + It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for + the return codes and their meanings.) */ + +int +regcomp (preg, pattern, cflags) + regex_t *_Restrict_ preg; + const char *_Restrict_ pattern; + int cflags; +{ + reg_errcode_t ret; + reg_syntax_t syntax = ((cflags & REG_EXTENDED) ? RE_SYNTAX_POSIX_EXTENDED + : RE_SYNTAX_POSIX_BASIC); + + preg->buffer = NULL; + preg->allocated = 0; + preg->used = 0; + + /* Try to allocate space for the fastmap. */ + preg->fastmap = re_malloc (char, SBC_MAX); + if (BE (preg->fastmap == NULL, 0)) + return REG_ESPACE; + + syntax |= (cflags & REG_ICASE) ? RE_ICASE : 0; + + /* If REG_NEWLINE is set, newlines are treated differently. */ + if (cflags & REG_NEWLINE) + { /* REG_NEWLINE implies neither . nor [^...] match newline. */ + syntax &= ~RE_DOT_NEWLINE; + syntax |= RE_HAT_LISTS_NOT_NEWLINE; + /* It also changes the matching behavior. */ + preg->newline_anchor = 1; + } + else + preg->newline_anchor = 0; + preg->no_sub = !!(cflags & REG_NOSUB); + preg->translate = NULL; + + ret = re_compile_internal (preg, pattern, strlen (pattern), syntax); + + /* POSIX doesn't distinguish between an unmatched open-group and an + unmatched close-group: both are REG_EPAREN. */ + if (ret == REG_ERPAREN) + ret = REG_EPAREN; + + /* We have already checked preg->fastmap != NULL. */ + if (BE (ret == REG_NOERROR, 1)) + /* Compute the fastmap now, since regexec cannot modify the pattern + buffer. This function never fails in this implementation. */ + (void) re_compile_fastmap (preg); + else + { + /* Some error occurred while compiling the expression. */ + re_free (preg->fastmap); + preg->fastmap = NULL; + } + + return (int) ret; +} +#ifdef _LIBC +weak_alias (__regcomp, regcomp) +#endif + +/* Returns a message corresponding to an error code, ERRCODE, returned + from either regcomp or regexec. We don't use PREG here. */ + +#ifdef _LIBC +size_t +regerror (errcode, preg, errbuf, errbuf_size) + int errcode; + const regex_t *_Restrict_ preg; + char *_Restrict_ errbuf; + size_t errbuf_size; +#else /* size_t might promote */ +size_t +regerror (int errcode, const regex_t *_Restrict_ preg, + char *_Restrict_ errbuf, size_t errbuf_size) +#endif +{ + const char *msg; + size_t msg_size; + + if (BE (errcode < 0 + || errcode >= (int) (sizeof (__re_error_msgid_idx) + / sizeof (__re_error_msgid_idx[0])), 0)) + /* Only error codes returned by the rest of the code should be passed + to this routine. If we are given anything else, or if other regex + code generates an invalid error code, then the program has a bug. + Dump core so we can fix it. */ + abort (); + + msg = gettext (__re_error_msgid + __re_error_msgid_idx[errcode]); + + msg_size = strlen (msg) + 1; /* Includes the null. */ + + if (BE (errbuf_size != 0, 1)) + { + size_t cpy_size = msg_size; + if (BE (msg_size > errbuf_size, 0)) + { + cpy_size = errbuf_size - 1; + errbuf[cpy_size] = '\0'; + } + memcpy (errbuf, msg, cpy_size); + } + + return msg_size; +} +#ifdef _LIBC +weak_alias (__regerror, regerror) +#endif + + +#ifdef RE_ENABLE_I18N +/* This static array is used for the map to single-byte characters when + UTF-8 is used. Otherwise we would allocate memory just to initialize + it the same all the time. UTF-8 is the preferred encoding so this is + a worthwhile optimization. */ +static const bitset_t utf8_sb_map = +{ + /* Set the first 128 bits. */ +# ifdef __GNUC__ + [0 ... 0x80 / BITSET_WORD_BITS - 1] = BITSET_WORD_MAX +# else +# if 4 * BITSET_WORD_BITS < ASCII_CHARS +# error "bitset_word_t is narrower than 32 bits" +# elif 3 * BITSET_WORD_BITS < ASCII_CHARS + BITSET_WORD_MAX, BITSET_WORD_MAX, BITSET_WORD_MAX, +# elif 2 * BITSET_WORD_BITS < ASCII_CHARS + BITSET_WORD_MAX, BITSET_WORD_MAX, +# elif 1 * BITSET_WORD_BITS < ASCII_CHARS + BITSET_WORD_MAX, +# endif + (BITSET_WORD_MAX + >> (SBC_MAX % BITSET_WORD_BITS == 0 + ? 0 + : BITSET_WORD_BITS - SBC_MAX % BITSET_WORD_BITS)) +# endif +}; +#endif + + +static void +free_dfa_content (re_dfa_t *dfa) +{ + Idx i, j; + + if (dfa->nodes) + for (i = 0; i < dfa->nodes_len; ++i) + free_token (dfa->nodes + i); + re_free (dfa->nexts); + for (i = 0; i < dfa->nodes_len; ++i) + { + if (dfa->eclosures != NULL) + re_node_set_free (dfa->eclosures + i); + if (dfa->inveclosures != NULL) + re_node_set_free (dfa->inveclosures + i); + if (dfa->edests != NULL) + re_node_set_free (dfa->edests + i); + } + re_free (dfa->edests); + re_free (dfa->eclosures); + re_free (dfa->inveclosures); + re_free (dfa->nodes); + + if (dfa->state_table) + for (i = 0; i <= dfa->state_hash_mask; ++i) + { + struct re_state_table_entry *entry = dfa->state_table + i; + for (j = 0; j < entry->num; ++j) + { + re_dfastate_t *state = entry->array[j]; + free_state (state); + } + re_free (entry->array); + } + re_free (dfa->state_table); +#ifdef RE_ENABLE_I18N + if (dfa->sb_char != utf8_sb_map) + re_free (dfa->sb_char); +#endif + re_free (dfa->subexp_map); +#ifdef DEBUG + re_free (dfa->re_str); +#endif + + re_free (dfa); +} + + +/* Free dynamically allocated space used by PREG. */ + +void +regfree (preg) + regex_t *preg; +{ + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; + if (BE (dfa != NULL, 1)) + free_dfa_content (dfa); + preg->buffer = NULL; + preg->allocated = 0; + + re_free (preg->fastmap); + preg->fastmap = NULL; + + re_free (preg->translate); + preg->translate = NULL; +} +#ifdef _LIBC +weak_alias (__regfree, regfree) +#endif + +/* Entry points compatible with 4.2 BSD regex library. We don't define + them unless specifically requested. */ + +#if defined _REGEX_RE_COMP || defined _LIBC + +/* BSD has one and only one pattern buffer. */ +static struct re_pattern_buffer re_comp_buf; + +char * +# ifdef _LIBC +/* Make these definitions weak in libc, so POSIX programs can redefine + these names if they don't use our functions, and still use + regcomp/regexec above without link errors. */ +weak_function +# endif +re_comp (s) + const char *s; +{ + reg_errcode_t ret; + char *fastmap; + + if (!s) + { + if (!re_comp_buf.buffer) + return gettext ("No previous regular expression"); + return 0; + } + + if (re_comp_buf.buffer) + { + fastmap = re_comp_buf.fastmap; + re_comp_buf.fastmap = NULL; + __regfree (&re_comp_buf); + memset (&re_comp_buf, '\0', sizeof (re_comp_buf)); + re_comp_buf.fastmap = fastmap; + } + + if (re_comp_buf.fastmap == NULL) + { + re_comp_buf.fastmap = (char *) malloc (SBC_MAX); + if (re_comp_buf.fastmap == NULL) + return (char *) gettext (__re_error_msgid + + __re_error_msgid_idx[(int) REG_ESPACE]); + } + + /* Since 're_exec' always passes NULL for the 'regs' argument, we + don't need to initialize the pattern buffer fields which affect it. */ + + /* Match anchors at newlines. */ + re_comp_buf.newline_anchor = 1; + + ret = re_compile_internal (&re_comp_buf, s, strlen (s), re_syntax_options); + + if (!ret) + return NULL; + + /* Yes, we're discarding 'const' here if !HAVE_LIBINTL. */ + return (char *) gettext (__re_error_msgid + __re_error_msgid_idx[(int) ret]); +} + +#ifdef _LIBC +libc_freeres_fn (free_mem) +{ + __regfree (&re_comp_buf); +} +#endif + +#endif /* _REGEX_RE_COMP */ + +/* Internal entry point. + Compile the regular expression PATTERN, whose length is LENGTH. + SYNTAX indicate regular expression's syntax. */ + +static reg_errcode_t +re_compile_internal (regex_t *preg, const char * pattern, size_t length, + reg_syntax_t syntax) +{ + reg_errcode_t err = REG_NOERROR; + re_dfa_t *dfa; + re_string_t regexp; + + /* Initialize the pattern buffer. */ + preg->fastmap_accurate = 0; + preg->syntax = syntax; + preg->not_bol = preg->not_eol = 0; + preg->used = 0; + preg->re_nsub = 0; + preg->can_be_null = 0; + preg->regs_allocated = REGS_UNALLOCATED; + + /* Initialize the dfa. */ + dfa = (re_dfa_t *) preg->buffer; + if (BE (preg->allocated < sizeof (re_dfa_t), 0)) + { + /* If zero allocated, but buffer is non-null, try to realloc + enough space. This loses if buffer's address is bogus, but + that is the user's responsibility. If ->buffer is NULL this + is a simple allocation. */ + dfa = re_realloc (preg->buffer, re_dfa_t, 1); + if (dfa == NULL) + return REG_ESPACE; + preg->allocated = sizeof (re_dfa_t); + preg->buffer = (unsigned char *) dfa; + } + preg->used = sizeof (re_dfa_t); + + err = init_dfa (dfa, length); + if (BE (err != REG_NOERROR, 0)) + { + free_dfa_content (dfa); + preg->buffer = NULL; + preg->allocated = 0; + return err; + } +#ifdef DEBUG + /* Note: length+1 will not overflow since it is checked in init_dfa. */ + dfa->re_str = re_malloc (char, length + 1); + strncpy (dfa->re_str, pattern, length + 1); +#endif + + __libc_lock_init (dfa->lock); + + err = re_string_construct (®exp, pattern, length, preg->translate, + (syntax & RE_ICASE) != 0, dfa); + if (BE (err != REG_NOERROR, 0)) + { + re_compile_internal_free_return: + free_workarea_compile (preg); + re_string_destruct (®exp); + free_dfa_content (dfa); + preg->buffer = NULL; + preg->allocated = 0; + return err; + } + + /* Parse the regular expression, and build a structure tree. */ + preg->re_nsub = 0; + dfa->str_tree = parse (®exp, preg, syntax, &err); + if (BE (dfa->str_tree == NULL, 0)) + goto re_compile_internal_free_return; + + /* Analyze the tree and create the nfa. */ + err = analyze (preg); + if (BE (err != REG_NOERROR, 0)) + goto re_compile_internal_free_return; + +#ifdef RE_ENABLE_I18N + /* If possible, do searching in single byte encoding to speed things up. */ + if (dfa->is_utf8 && !(syntax & RE_ICASE) && preg->translate == NULL) + optimize_utf8 (dfa); +#endif + + /* Then create the initial state of the dfa. */ + err = create_initial_state (dfa); + + /* Release work areas. */ + free_workarea_compile (preg); + re_string_destruct (®exp); + + if (BE (err != REG_NOERROR, 0)) + { + free_dfa_content (dfa); + preg->buffer = NULL; + preg->allocated = 0; + } + + return err; +} + +/* Initialize DFA. We use the length of the regular expression PAT_LEN + as the initial length of some arrays. */ + +static reg_errcode_t +init_dfa (re_dfa_t *dfa, size_t pat_len) +{ + __re_size_t table_size; +#ifndef _LIBC + char *codeset_name; +#endif +#ifdef RE_ENABLE_I18N + size_t max_i18n_object_size = MAX (sizeof (wchar_t), sizeof (wctype_t)); +#else + size_t max_i18n_object_size = 0; +#endif + size_t max_object_size = + MAX (sizeof (struct re_state_table_entry), + MAX (sizeof (re_token_t), + MAX (sizeof (re_node_set), + MAX (sizeof (regmatch_t), + max_i18n_object_size)))); + + memset (dfa, '\0', sizeof (re_dfa_t)); + + /* Force allocation of str_tree_storage the first time. */ + dfa->str_tree_storage_idx = BIN_TREE_STORAGE_SIZE; + + /* Avoid overflows. The extra "/ 2" is for the table_size doubling + calculation below, and for similar doubling calculations + elsewhere. And it's <= rather than <, because some of the + doubling calculations add 1 afterwards. */ + if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) / 2 <= pat_len, 0)) + return REG_ESPACE; + + dfa->nodes_alloc = pat_len + 1; + dfa->nodes = re_malloc (re_token_t, dfa->nodes_alloc); + + /* table_size = 2 ^ ceil(log pat_len) */ + for (table_size = 1; ; table_size <<= 1) + if (table_size > pat_len) + break; + + dfa->state_table = calloc (sizeof (struct re_state_table_entry), table_size); + dfa->state_hash_mask = table_size - 1; + + dfa->mb_cur_max = MB_CUR_MAX; +#ifdef _LIBC + if (dfa->mb_cur_max == 6 + && strcmp (_NL_CURRENT (LC_CTYPE, _NL_CTYPE_CODESET_NAME), "UTF-8") == 0) + dfa->is_utf8 = 1; + dfa->map_notascii = (_NL_CURRENT_WORD (LC_CTYPE, _NL_CTYPE_MAP_TO_NONASCII) + != 0); +#else + codeset_name = nl_langinfo (CODESET); + if (strcasecmp (codeset_name, "UTF-8") == 0 + || strcasecmp (codeset_name, "UTF8") == 0) + dfa->is_utf8 = 1; + + /* We check exhaustively in the loop below if this charset is a + superset of ASCII. */ + dfa->map_notascii = 0; +#endif + +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + { + if (dfa->is_utf8) + dfa->sb_char = (re_bitset_ptr_t) utf8_sb_map; + else + { + int i, j, ch; + + dfa->sb_char = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1); + if (BE (dfa->sb_char == NULL, 0)) + return REG_ESPACE; + + /* Set the bits corresponding to single byte chars. */ + for (i = 0, ch = 0; i < BITSET_WORDS; ++i) + for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch) + { + wint_t wch = __btowc (ch); + if (wch != WEOF) + dfa->sb_char[i] |= (bitset_word_t) 1 << j; +# ifndef _LIBC + if (isascii (ch) && wch != ch) + dfa->map_notascii = 1; +# endif + } + } + } +#endif + + if (BE (dfa->nodes == NULL || dfa->state_table == NULL, 0)) + return REG_ESPACE; + return REG_NOERROR; +} + +/* Initialize WORD_CHAR table, which indicate which character is + "word". In this case "word" means that it is the word construction + character used by some operators like "\<", "\>", etc. */ + +static void +internal_function +init_word_char (re_dfa_t *dfa) +{ + dfa->word_ops_used = 1; + int i = 0; + int j; + int ch = 0; + if (BE (dfa->map_notascii == 0, 1)) + { + if (BITSET_WORD_BITS == 64) + { + dfa->word_char[0] = UINT64_C (0x03ff000000000000); + dfa->word_char[1] = UINT64_C (0x07fffffe87fffffe); + i = 2; + } + else if (BITSET_WORD_BITS == 32) + { + dfa->word_char[0] = UINT32_C (0x00000000); + dfa->word_char[1] = UINT32_C (0x03ff0000); + dfa->word_char[2] = UINT32_C (0x87fffffe); + dfa->word_char[3] = UINT32_C (0x07fffffe); + i = 4; + } + else + goto general_case; + ch = 128; + + if (BE (dfa->is_utf8, 1)) + { + memset (&dfa->word_char[i], '\0', (SBC_MAX - ch) / 8); + return; + } + } + + general_case: + for (; i < BITSET_WORDS; ++i) + for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch) + if (isalnum (ch) || ch == '_') + dfa->word_char[i] |= (bitset_word_t) 1 << j; +} + +/* Free the work area which are only used while compiling. */ + +static void +free_workarea_compile (regex_t *preg) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; + bin_tree_storage_t *storage, *next; + for (storage = dfa->str_tree_storage; storage; storage = next) + { + next = storage->next; + re_free (storage); + } + dfa->str_tree_storage = NULL; + dfa->str_tree_storage_idx = BIN_TREE_STORAGE_SIZE; + dfa->str_tree = NULL; + re_free (dfa->org_indices); + dfa->org_indices = NULL; +} + +/* Create initial states for all contexts. */ + +static reg_errcode_t +create_initial_state (re_dfa_t *dfa) +{ + Idx first, i; + reg_errcode_t err; + re_node_set init_nodes; + + /* Initial states have the epsilon closure of the node which is + the first node of the regular expression. */ + first = dfa->str_tree->first->node_idx; + dfa->init_node = first; + err = re_node_set_init_copy (&init_nodes, dfa->eclosures + first); + if (BE (err != REG_NOERROR, 0)) + return err; + + /* The back-references which are in initial states can epsilon transit, + since in this case all of the subexpressions can be null. + Then we add epsilon closures of the nodes which are the next nodes of + the back-references. */ + if (dfa->nbackref > 0) + for (i = 0; i < init_nodes.nelem; ++i) + { + Idx node_idx = init_nodes.elems[i]; + re_token_type_t type = dfa->nodes[node_idx].type; + + Idx clexp_idx; + if (type != OP_BACK_REF) + continue; + for (clexp_idx = 0; clexp_idx < init_nodes.nelem; ++clexp_idx) + { + re_token_t *clexp_node; + clexp_node = dfa->nodes + init_nodes.elems[clexp_idx]; + if (clexp_node->type == OP_CLOSE_SUBEXP + && clexp_node->opr.idx == dfa->nodes[node_idx].opr.idx) + break; + } + if (clexp_idx == init_nodes.nelem) + continue; + + if (type == OP_BACK_REF) + { + Idx dest_idx = dfa->edests[node_idx].elems[0]; + if (!re_node_set_contains (&init_nodes, dest_idx)) + { + reg_errcode_t merge_err + = re_node_set_merge (&init_nodes, dfa->eclosures + dest_idx); + if (merge_err != REG_NOERROR) + return merge_err; + i = 0; + } + } + } + + /* It must be the first time to invoke acquire_state. */ + dfa->init_state = re_acquire_state_context (&err, dfa, &init_nodes, 0); + /* We don't check ERR here, since the initial state must not be NULL. */ + if (BE (dfa->init_state == NULL, 0)) + return err; + if (dfa->init_state->has_constraint) + { + dfa->init_state_word = re_acquire_state_context (&err, dfa, &init_nodes, + CONTEXT_WORD); + dfa->init_state_nl = re_acquire_state_context (&err, dfa, &init_nodes, + CONTEXT_NEWLINE); + dfa->init_state_begbuf = re_acquire_state_context (&err, dfa, + &init_nodes, + CONTEXT_NEWLINE + | CONTEXT_BEGBUF); + if (BE (dfa->init_state_word == NULL || dfa->init_state_nl == NULL + || dfa->init_state_begbuf == NULL, 0)) + return err; + } + else + dfa->init_state_word = dfa->init_state_nl + = dfa->init_state_begbuf = dfa->init_state; + + re_node_set_free (&init_nodes); + return REG_NOERROR; +} + +#ifdef RE_ENABLE_I18N +/* If it is possible to do searching in single byte encoding instead of UTF-8 + to speed things up, set dfa->mb_cur_max to 1, clear is_utf8 and change + DFA nodes where needed. */ + +static void +optimize_utf8 (re_dfa_t *dfa) +{ + Idx node; + int i; + bool mb_chars = false; + bool has_period = false; + + for (node = 0; node < dfa->nodes_len; ++node) + switch (dfa->nodes[node].type) + { + case CHARACTER: + if (dfa->nodes[node].opr.c >= ASCII_CHARS) + mb_chars = true; + break; + case ANCHOR: + switch (dfa->nodes[node].opr.ctx_type) + { + case LINE_FIRST: + case LINE_LAST: + case BUF_FIRST: + case BUF_LAST: + break; + default: + /* Word anchors etc. cannot be handled. It's okay to test + opr.ctx_type since constraints (for all DFA nodes) are + created by ORing one or more opr.ctx_type values. */ + return; + } + break; + case OP_PERIOD: + has_period = true; + break; + case OP_BACK_REF: + case OP_ALT: + case END_OF_RE: + case OP_DUP_ASTERISK: + case OP_OPEN_SUBEXP: + case OP_CLOSE_SUBEXP: + break; + case COMPLEX_BRACKET: + return; + case SIMPLE_BRACKET: + /* Just double check. */ + { + int rshift = (ASCII_CHARS % BITSET_WORD_BITS == 0 + ? 0 + : BITSET_WORD_BITS - ASCII_CHARS % BITSET_WORD_BITS); + for (i = ASCII_CHARS / BITSET_WORD_BITS; i < BITSET_WORDS; ++i) + { + if (dfa->nodes[node].opr.sbcset[i] >> rshift != 0) + return; + rshift = 0; + } + } + break; + default: + abort (); + } + + if (mb_chars || has_period) + for (node = 0; node < dfa->nodes_len; ++node) + { + if (dfa->nodes[node].type == CHARACTER + && dfa->nodes[node].opr.c >= ASCII_CHARS) + dfa->nodes[node].mb_partial = 0; + else if (dfa->nodes[node].type == OP_PERIOD) + dfa->nodes[node].type = OP_UTF8_PERIOD; + } + + /* The search can be in single byte locale. */ + dfa->mb_cur_max = 1; + dfa->is_utf8 = 0; + dfa->has_mb_node = dfa->nbackref > 0 || has_period; +} +#endif + +/* Analyze the structure tree, and calculate "first", "next", "edest", + "eclosure", and "inveclosure". */ + +static reg_errcode_t +analyze (regex_t *preg) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; + reg_errcode_t ret; + + /* Allocate arrays. */ + dfa->nexts = re_malloc (Idx, dfa->nodes_alloc); + dfa->org_indices = re_malloc (Idx, dfa->nodes_alloc); + dfa->edests = re_malloc (re_node_set, dfa->nodes_alloc); + dfa->eclosures = re_malloc (re_node_set, dfa->nodes_alloc); + if (BE (dfa->nexts == NULL || dfa->org_indices == NULL || dfa->edests == NULL + || dfa->eclosures == NULL, 0)) + return REG_ESPACE; + + dfa->subexp_map = re_malloc (Idx, preg->re_nsub); + if (dfa->subexp_map != NULL) + { + Idx i; + for (i = 0; i < preg->re_nsub; i++) + dfa->subexp_map[i] = i; + preorder (dfa->str_tree, optimize_subexps, dfa); + for (i = 0; i < preg->re_nsub; i++) + if (dfa->subexp_map[i] != i) + break; + if (i == preg->re_nsub) + { + free (dfa->subexp_map); + dfa->subexp_map = NULL; + } + } + + ret = postorder (dfa->str_tree, lower_subexps, preg); + if (BE (ret != REG_NOERROR, 0)) + return ret; + ret = postorder (dfa->str_tree, calc_first, dfa); + if (BE (ret != REG_NOERROR, 0)) + return ret; + preorder (dfa->str_tree, calc_next, dfa); + ret = preorder (dfa->str_tree, link_nfa_nodes, dfa); + if (BE (ret != REG_NOERROR, 0)) + return ret; + ret = calc_eclosure (dfa); + if (BE (ret != REG_NOERROR, 0)) + return ret; + + /* We only need this during the prune_impossible_nodes pass in regexec.c; + skip it if p_i_n will not run, as calc_inveclosure can be quadratic. */ + if ((!preg->no_sub && preg->re_nsub > 0 && dfa->has_plural_match) + || dfa->nbackref) + { + dfa->inveclosures = re_malloc (re_node_set, dfa->nodes_len); + if (BE (dfa->inveclosures == NULL, 0)) + return REG_ESPACE; + ret = calc_inveclosure (dfa); + } + + return ret; +} + +/* Our parse trees are very unbalanced, so we cannot use a stack to + implement parse tree visits. Instead, we use parent pointers and + some hairy code in these two functions. */ +static reg_errcode_t +postorder (bin_tree_t *root, reg_errcode_t (fn (void *, bin_tree_t *)), + void *extra) +{ + bin_tree_t *node, *prev; + + for (node = root; ; ) + { + /* Descend down the tree, preferably to the left (or to the right + if that's the only child). */ + while (node->left || node->right) + if (node->left) + node = node->left; + else + node = node->right; + + do + { + reg_errcode_t err = fn (extra, node); + if (BE (err != REG_NOERROR, 0)) + return err; + if (node->parent == NULL) + return REG_NOERROR; + prev = node; + node = node->parent; + } + /* Go up while we have a node that is reached from the right. */ + while (node->right == prev || node->right == NULL); + node = node->right; + } +} + +static reg_errcode_t +preorder (bin_tree_t *root, reg_errcode_t (fn (void *, bin_tree_t *)), + void *extra) +{ + bin_tree_t *node; + + for (node = root; ; ) + { + reg_errcode_t err = fn (extra, node); + if (BE (err != REG_NOERROR, 0)) + return err; + + /* Go to the left node, or up and to the right. */ + if (node->left) + node = node->left; + else + { + bin_tree_t *prev = NULL; + while (node->right == prev || node->right == NULL) + { + prev = node; + node = node->parent; + if (!node) + return REG_NOERROR; + } + node = node->right; + } + } +} + +/* Optimization pass: if a SUBEXP is entirely contained, strip it and tell + re_search_internal to map the inner one's opr.idx to this one's. Adjust + backreferences as well. Requires a preorder visit. */ +static reg_errcode_t +optimize_subexps (void *extra, bin_tree_t *node) +{ + re_dfa_t *dfa = (re_dfa_t *) extra; + + if (node->token.type == OP_BACK_REF && dfa->subexp_map) + { + int idx = node->token.opr.idx; + node->token.opr.idx = dfa->subexp_map[idx]; + dfa->used_bkref_map |= 1 << node->token.opr.idx; + } + + else if (node->token.type == SUBEXP + && node->left && node->left->token.type == SUBEXP) + { + Idx other_idx = node->left->token.opr.idx; + + node->left = node->left->left; + if (node->left) + node->left->parent = node; + + dfa->subexp_map[other_idx] = dfa->subexp_map[node->token.opr.idx]; + if (other_idx < BITSET_WORD_BITS) + dfa->used_bkref_map &= ~((bitset_word_t) 1 << other_idx); + } + + return REG_NOERROR; +} + +/* Lowering pass: Turn each SUBEXP node into the appropriate concatenation + of OP_OPEN_SUBEXP, the body of the SUBEXP (if any) and OP_CLOSE_SUBEXP. */ +static reg_errcode_t +lower_subexps (void *extra, bin_tree_t *node) +{ + regex_t *preg = (regex_t *) extra; + reg_errcode_t err = REG_NOERROR; + + if (node->left && node->left->token.type == SUBEXP) + { + node->left = lower_subexp (&err, preg, node->left); + if (node->left) + node->left->parent = node; + } + if (node->right && node->right->token.type == SUBEXP) + { + node->right = lower_subexp (&err, preg, node->right); + if (node->right) + node->right->parent = node; + } + + return err; +} + +static bin_tree_t * +lower_subexp (reg_errcode_t *err, regex_t *preg, bin_tree_t *node) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; + bin_tree_t *body = node->left; + bin_tree_t *op, *cls, *tree1, *tree; + + if (preg->no_sub + /* We do not optimize empty subexpressions, because otherwise we may + have bad CONCAT nodes with NULL children. This is obviously not + very common, so we do not lose much. An example that triggers + this case is the sed "script" /\(\)/x. */ + && node->left != NULL + && (node->token.opr.idx >= BITSET_WORD_BITS + || !(dfa->used_bkref_map + & ((bitset_word_t) 1 << node->token.opr.idx)))) + return node->left; + + /* Convert the SUBEXP node to the concatenation of an + OP_OPEN_SUBEXP, the contents, and an OP_CLOSE_SUBEXP. */ + op = create_tree (dfa, NULL, NULL, OP_OPEN_SUBEXP); + cls = create_tree (dfa, NULL, NULL, OP_CLOSE_SUBEXP); + tree1 = body ? create_tree (dfa, body, cls, CONCAT) : cls; + tree = create_tree (dfa, op, tree1, CONCAT); + if (BE (tree == NULL || tree1 == NULL || op == NULL || cls == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + + op->token.opr.idx = cls->token.opr.idx = node->token.opr.idx; + op->token.opt_subexp = cls->token.opt_subexp = node->token.opt_subexp; + return tree; +} + +/* Pass 1 in building the NFA: compute FIRST and create unlinked automaton + nodes. Requires a postorder visit. */ +static reg_errcode_t +calc_first (void *extra, bin_tree_t *node) +{ + re_dfa_t *dfa = (re_dfa_t *) extra; + if (node->token.type == CONCAT) + { + node->first = node->left->first; + node->node_idx = node->left->node_idx; + } + else + { + node->first = node; + node->node_idx = re_dfa_add_node (dfa, node->token); + if (BE (node->node_idx == REG_MISSING, 0)) + return REG_ESPACE; + if (node->token.type == ANCHOR) + dfa->nodes[node->node_idx].constraint = node->token.opr.ctx_type; + } + return REG_NOERROR; +} + +/* Pass 2: compute NEXT on the tree. Preorder visit. */ +static reg_errcode_t +calc_next (void *extra, bin_tree_t *node) +{ + switch (node->token.type) + { + case OP_DUP_ASTERISK: + node->left->next = node; + break; + case CONCAT: + node->left->next = node->right->first; + node->right->next = node->next; + break; + default: + if (node->left) + node->left->next = node->next; + if (node->right) + node->right->next = node->next; + break; + } + return REG_NOERROR; +} + +/* Pass 3: link all DFA nodes to their NEXT node (any order will do). */ +static reg_errcode_t +link_nfa_nodes (void *extra, bin_tree_t *node) +{ + re_dfa_t *dfa = (re_dfa_t *) extra; + Idx idx = node->node_idx; + reg_errcode_t err = REG_NOERROR; + + switch (node->token.type) + { + case CONCAT: + break; + + case END_OF_RE: + assert (node->next == NULL); + break; + + case OP_DUP_ASTERISK: + case OP_ALT: + { + Idx left, right; + dfa->has_plural_match = 1; + if (node->left != NULL) + left = node->left->first->node_idx; + else + left = node->next->node_idx; + if (node->right != NULL) + right = node->right->first->node_idx; + else + right = node->next->node_idx; + assert (REG_VALID_INDEX (left)); + assert (REG_VALID_INDEX (right)); + err = re_node_set_init_2 (dfa->edests + idx, left, right); + } + break; + + case ANCHOR: + case OP_OPEN_SUBEXP: + case OP_CLOSE_SUBEXP: + err = re_node_set_init_1 (dfa->edests + idx, node->next->node_idx); + break; + + case OP_BACK_REF: + dfa->nexts[idx] = node->next->node_idx; + if (node->token.type == OP_BACK_REF) + err = re_node_set_init_1 (dfa->edests + idx, dfa->nexts[idx]); + break; + + default: + assert (!IS_EPSILON_NODE (node->token.type)); + dfa->nexts[idx] = node->next->node_idx; + break; + } + + return err; +} + +/* Duplicate the epsilon closure of the node ROOT_NODE. + Note that duplicated nodes have constraint INIT_CONSTRAINT in addition + to their own constraint. */ + +static reg_errcode_t +internal_function +duplicate_node_closure (re_dfa_t *dfa, Idx top_org_node, Idx top_clone_node, + Idx root_node, unsigned int init_constraint) +{ + Idx org_node, clone_node; + bool ok; + unsigned int constraint = init_constraint; + for (org_node = top_org_node, clone_node = top_clone_node;;) + { + Idx org_dest, clone_dest; + if (dfa->nodes[org_node].type == OP_BACK_REF) + { + /* If the back reference epsilon-transit, its destination must + also have the constraint. Then duplicate the epsilon closure + of the destination of the back reference, and store it in + edests of the back reference. */ + org_dest = dfa->nexts[org_node]; + re_node_set_empty (dfa->edests + clone_node); + clone_dest = duplicate_node (dfa, org_dest, constraint); + if (BE (clone_dest == REG_MISSING, 0)) + return REG_ESPACE; + dfa->nexts[clone_node] = dfa->nexts[org_node]; + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + } + else if (dfa->edests[org_node].nelem == 0) + { + /* In case of the node can't epsilon-transit, don't duplicate the + destination and store the original destination as the + destination of the node. */ + dfa->nexts[clone_node] = dfa->nexts[org_node]; + break; + } + else if (dfa->edests[org_node].nelem == 1) + { + /* In case of the node can epsilon-transit, and it has only one + destination. */ + org_dest = dfa->edests[org_node].elems[0]; + re_node_set_empty (dfa->edests + clone_node); + /* If the node is root_node itself, it means the epsilon closure + has a loop. Then tie it to the destination of the root_node. */ + if (org_node == root_node && clone_node != org_node) + { + ok = re_node_set_insert (dfa->edests + clone_node, org_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + break; + } + /* In case the node has another constraint, append it. */ + constraint |= dfa->nodes[org_node].constraint; + clone_dest = duplicate_node (dfa, org_dest, constraint); + if (BE (clone_dest == REG_MISSING, 0)) + return REG_ESPACE; + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + } + else /* dfa->edests[org_node].nelem == 2 */ + { + /* In case of the node can epsilon-transit, and it has two + destinations. In the bin_tree_t and DFA, that's '|' and '*'. */ + org_dest = dfa->edests[org_node].elems[0]; + re_node_set_empty (dfa->edests + clone_node); + /* Search for a duplicated node which satisfies the constraint. */ + clone_dest = search_duplicated_node (dfa, org_dest, constraint); + if (clone_dest == REG_MISSING) + { + /* There is no such duplicated node, create a new one. */ + reg_errcode_t err; + clone_dest = duplicate_node (dfa, org_dest, constraint); + if (BE (clone_dest == REG_MISSING, 0)) + return REG_ESPACE; + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + err = duplicate_node_closure (dfa, org_dest, clone_dest, + root_node, constraint); + if (BE (err != REG_NOERROR, 0)) + return err; + } + else + { + /* There is a duplicated node which satisfies the constraint, + use it to avoid infinite loop. */ + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + } + + org_dest = dfa->edests[org_node].elems[1]; + clone_dest = duplicate_node (dfa, org_dest, constraint); + if (BE (clone_dest == REG_MISSING, 0)) + return REG_ESPACE; + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + } + org_node = org_dest; + clone_node = clone_dest; + } + return REG_NOERROR; +} + +/* Search for a node which is duplicated from the node ORG_NODE, and + satisfies the constraint CONSTRAINT. */ + +static Idx +search_duplicated_node (const re_dfa_t *dfa, Idx org_node, + unsigned int constraint) +{ + Idx idx; + for (idx = dfa->nodes_len - 1; dfa->nodes[idx].duplicated && idx > 0; --idx) + { + if (org_node == dfa->org_indices[idx] + && constraint == dfa->nodes[idx].constraint) + return idx; /* Found. */ + } + return REG_MISSING; /* Not found. */ +} + +/* Duplicate the node whose index is ORG_IDX and set the constraint CONSTRAINT. + Return the index of the new node, or REG_MISSING if insufficient storage is + available. */ + +static Idx +duplicate_node (re_dfa_t *dfa, Idx org_idx, unsigned int constraint) +{ + Idx dup_idx = re_dfa_add_node (dfa, dfa->nodes[org_idx]); + if (BE (dup_idx != REG_MISSING, 1)) + { + dfa->nodes[dup_idx].constraint = constraint; + dfa->nodes[dup_idx].constraint |= dfa->nodes[org_idx].constraint; + dfa->nodes[dup_idx].duplicated = 1; + + /* Store the index of the original node. */ + dfa->org_indices[dup_idx] = org_idx; + } + return dup_idx; +} + +static reg_errcode_t +calc_inveclosure (re_dfa_t *dfa) +{ + Idx src, idx; + bool ok; + for (idx = 0; idx < dfa->nodes_len; ++idx) + re_node_set_init_empty (dfa->inveclosures + idx); + + for (src = 0; src < dfa->nodes_len; ++src) + { + Idx *elems = dfa->eclosures[src].elems; + for (idx = 0; idx < dfa->eclosures[src].nelem; ++idx) + { + ok = re_node_set_insert_last (dfa->inveclosures + elems[idx], src); + if (BE (! ok, 0)) + return REG_ESPACE; + } + } + + return REG_NOERROR; +} + +/* Calculate "eclosure" for all the node in DFA. */ + +static reg_errcode_t +calc_eclosure (re_dfa_t *dfa) +{ + Idx node_idx; + bool incomplete; +#ifdef DEBUG + assert (dfa->nodes_len > 0); +#endif + incomplete = false; + /* For each nodes, calculate epsilon closure. */ + for (node_idx = 0; ; ++node_idx) + { + reg_errcode_t err; + re_node_set eclosure_elem; + if (node_idx == dfa->nodes_len) + { + if (!incomplete) + break; + incomplete = false; + node_idx = 0; + } + +#ifdef DEBUG + assert (dfa->eclosures[node_idx].nelem != REG_MISSING); +#endif + + /* If we have already calculated, skip it. */ + if (dfa->eclosures[node_idx].nelem != 0) + continue; + /* Calculate epsilon closure of 'node_idx'. */ + err = calc_eclosure_iter (&eclosure_elem, dfa, node_idx, true); + if (BE (err != REG_NOERROR, 0)) + return err; + + if (dfa->eclosures[node_idx].nelem == 0) + { + incomplete = true; + re_node_set_free (&eclosure_elem); + } + } + return REG_NOERROR; +} + +/* Calculate epsilon closure of NODE. */ + +static reg_errcode_t +calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, Idx node, bool root) +{ + reg_errcode_t err; + Idx i; + re_node_set eclosure; + bool ok; + bool incomplete = false; + err = re_node_set_alloc (&eclosure, dfa->edests[node].nelem + 1); + if (BE (err != REG_NOERROR, 0)) + return err; + + /* This indicates that we are calculating this node now. + We reference this value to avoid infinite loop. */ + dfa->eclosures[node].nelem = REG_MISSING; + + /* If the current node has constraints, duplicate all nodes + since they must inherit the constraints. */ + if (dfa->nodes[node].constraint + && dfa->edests[node].nelem + && !dfa->nodes[dfa->edests[node].elems[0]].duplicated) + { + err = duplicate_node_closure (dfa, node, node, node, + dfa->nodes[node].constraint); + if (BE (err != REG_NOERROR, 0)) + return err; + } + + /* Expand each epsilon destination nodes. */ + if (IS_EPSILON_NODE(dfa->nodes[node].type)) + for (i = 0; i < dfa->edests[node].nelem; ++i) + { + re_node_set eclosure_elem; + Idx edest = dfa->edests[node].elems[i]; + /* If calculating the epsilon closure of 'edest' is in progress, + return intermediate result. */ + if (dfa->eclosures[edest].nelem == REG_MISSING) + { + incomplete = true; + continue; + } + /* If we haven't calculated the epsilon closure of 'edest' yet, + calculate now. Otherwise use calculated epsilon closure. */ + if (dfa->eclosures[edest].nelem == 0) + { + err = calc_eclosure_iter (&eclosure_elem, dfa, edest, false); + if (BE (err != REG_NOERROR, 0)) + return err; + } + else + eclosure_elem = dfa->eclosures[edest]; + /* Merge the epsilon closure of 'edest'. */ + err = re_node_set_merge (&eclosure, &eclosure_elem); + if (BE (err != REG_NOERROR, 0)) + return err; + /* If the epsilon closure of 'edest' is incomplete, + the epsilon closure of this node is also incomplete. */ + if (dfa->eclosures[edest].nelem == 0) + { + incomplete = true; + re_node_set_free (&eclosure_elem); + } + } + + /* An epsilon closure includes itself. */ + ok = re_node_set_insert (&eclosure, node); + if (BE (! ok, 0)) + return REG_ESPACE; + if (incomplete && !root) + dfa->eclosures[node].nelem = 0; + else + dfa->eclosures[node] = eclosure; + *new_set = eclosure; + return REG_NOERROR; +} + +/* Functions for token which are used in the parser. */ + +/* Fetch a token from INPUT. + We must not use this function inside bracket expressions. */ + +static void +internal_function +fetch_token (re_token_t *result, re_string_t *input, reg_syntax_t syntax) +{ + re_string_skip_bytes (input, peek_token (result, input, syntax)); +} + +/* Peek a token from INPUT, and return the length of the token. + We must not use this function inside bracket expressions. */ + +static int +internal_function +peek_token (re_token_t *token, re_string_t *input, reg_syntax_t syntax) +{ + unsigned char c; + + if (re_string_eoi (input)) + { + token->type = END_OF_RE; + return 0; + } + + c = re_string_peek_byte (input, 0); + token->opr.c = c; + + token->word_char = 0; +#ifdef RE_ENABLE_I18N + token->mb_partial = 0; + if (input->mb_cur_max > 1 && + !re_string_first_byte (input, re_string_cur_idx (input))) + { + token->type = CHARACTER; + token->mb_partial = 1; + return 1; + } +#endif + if (c == '\\') + { + unsigned char c2; + if (re_string_cur_idx (input) + 1 >= re_string_length (input)) + { + token->type = BACK_SLASH; + return 1; + } + + c2 = re_string_peek_byte_case (input, 1); + token->opr.c = c2; + token->type = CHARACTER; +#ifdef RE_ENABLE_I18N + if (input->mb_cur_max > 1) + { + wint_t wc = re_string_wchar_at (input, + re_string_cur_idx (input) + 1); + token->word_char = IS_WIDE_WORD_CHAR (wc) != 0; + } + else +#endif + token->word_char = IS_WORD_CHAR (c2) != 0; + + switch (c2) + { + case '|': + if (!(syntax & RE_LIMITED_OPS) && !(syntax & RE_NO_BK_VBAR)) + token->type = OP_ALT; + break; + case '1': case '2': case '3': case '4': case '5': + case '6': case '7': case '8': case '9': + if (!(syntax & RE_NO_BK_REFS)) + { + token->type = OP_BACK_REF; + token->opr.idx = c2 - '1'; + } + break; + case '<': + if (!(syntax & RE_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = WORD_FIRST; + } + break; + case '>': + if (!(syntax & RE_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = WORD_LAST; + } + break; + case 'b': + if (!(syntax & RE_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = WORD_DELIM; + } + break; + case 'B': + if (!(syntax & RE_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = NOT_WORD_DELIM; + } + break; + case 'w': + if (!(syntax & RE_NO_GNU_OPS)) + token->type = OP_WORD; + break; + case 'W': + if (!(syntax & RE_NO_GNU_OPS)) + token->type = OP_NOTWORD; + break; + case 's': + if (!(syntax & RE_NO_GNU_OPS)) + token->type = OP_SPACE; + break; + case 'S': + if (!(syntax & RE_NO_GNU_OPS)) + token->type = OP_NOTSPACE; + break; + case '`': + if (!(syntax & RE_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = BUF_FIRST; + } + break; + case '\'': + if (!(syntax & RE_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = BUF_LAST; + } + break; + case '(': + if (!(syntax & RE_NO_BK_PARENS)) + token->type = OP_OPEN_SUBEXP; + break; + case ')': + if (!(syntax & RE_NO_BK_PARENS)) + token->type = OP_CLOSE_SUBEXP; + break; + case '+': + if (!(syntax & RE_LIMITED_OPS) && (syntax & RE_BK_PLUS_QM)) + token->type = OP_DUP_PLUS; + break; + case '?': + if (!(syntax & RE_LIMITED_OPS) && (syntax & RE_BK_PLUS_QM)) + token->type = OP_DUP_QUESTION; + break; + case '{': + if ((syntax & RE_INTERVALS) && (!(syntax & RE_NO_BK_BRACES))) + token->type = OP_OPEN_DUP_NUM; + break; + case '}': + if ((syntax & RE_INTERVALS) && (!(syntax & RE_NO_BK_BRACES))) + token->type = OP_CLOSE_DUP_NUM; + break; + default: + break; + } + return 2; + } + + token->type = CHARACTER; +#ifdef RE_ENABLE_I18N + if (input->mb_cur_max > 1) + { + wint_t wc = re_string_wchar_at (input, re_string_cur_idx (input)); + token->word_char = IS_WIDE_WORD_CHAR (wc) != 0; + } + else +#endif + token->word_char = IS_WORD_CHAR (token->opr.c); + + switch (c) + { + case '\n': + if (syntax & RE_NEWLINE_ALT) + token->type = OP_ALT; + break; + case '|': + if (!(syntax & RE_LIMITED_OPS) && (syntax & RE_NO_BK_VBAR)) + token->type = OP_ALT; + break; + case '*': + token->type = OP_DUP_ASTERISK; + break; + case '+': + if (!(syntax & RE_LIMITED_OPS) && !(syntax & RE_BK_PLUS_QM)) + token->type = OP_DUP_PLUS; + break; + case '?': + if (!(syntax & RE_LIMITED_OPS) && !(syntax & RE_BK_PLUS_QM)) + token->type = OP_DUP_QUESTION; + break; + case '{': + if ((syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES)) + token->type = OP_OPEN_DUP_NUM; + break; + case '}': + if ((syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES)) + token->type = OP_CLOSE_DUP_NUM; + break; + case '(': + if (syntax & RE_NO_BK_PARENS) + token->type = OP_OPEN_SUBEXP; + break; + case ')': + if (syntax & RE_NO_BK_PARENS) + token->type = OP_CLOSE_SUBEXP; + break; + case '[': + token->type = OP_OPEN_BRACKET; + break; + case '.': + token->type = OP_PERIOD; + break; + case '^': + if (!(syntax & (RE_CONTEXT_INDEP_ANCHORS | RE_CARET_ANCHORS_HERE)) && + re_string_cur_idx (input) != 0) + { + char prev = re_string_peek_byte (input, -1); + if (!(syntax & RE_NEWLINE_ALT) || prev != '\n') + break; + } + token->type = ANCHOR; + token->opr.ctx_type = LINE_FIRST; + break; + case '$': + if (!(syntax & RE_CONTEXT_INDEP_ANCHORS) && + re_string_cur_idx (input) + 1 != re_string_length (input)) + { + re_token_t next; + re_string_skip_bytes (input, 1); + peek_token (&next, input, syntax); + re_string_skip_bytes (input, -1); + if (next.type != OP_ALT && next.type != OP_CLOSE_SUBEXP) + break; + } + token->type = ANCHOR; + token->opr.ctx_type = LINE_LAST; + break; + default: + break; + } + return 1; +} + +/* Peek a token from INPUT, and return the length of the token. + We must not use this function out of bracket expressions. */ + +static int +internal_function +peek_token_bracket (re_token_t *token, re_string_t *input, reg_syntax_t syntax) +{ + unsigned char c; + if (re_string_eoi (input)) + { + token->type = END_OF_RE; + return 0; + } + c = re_string_peek_byte (input, 0); + token->opr.c = c; + +#ifdef RE_ENABLE_I18N + if (input->mb_cur_max > 1 && + !re_string_first_byte (input, re_string_cur_idx (input))) + { + token->type = CHARACTER; + return 1; + } +#endif /* RE_ENABLE_I18N */ + + if (c == '\\' && (syntax & RE_BACKSLASH_ESCAPE_IN_LISTS) + && re_string_cur_idx (input) + 1 < re_string_length (input)) + { + /* In this case, '\' escape a character. */ + unsigned char c2; + re_string_skip_bytes (input, 1); + c2 = re_string_peek_byte (input, 0); + token->opr.c = c2; + token->type = CHARACTER; + return 1; + } + if (c == '[') /* '[' is a special char in a bracket exps. */ + { + unsigned char c2; + int token_len; + if (re_string_cur_idx (input) + 1 < re_string_length (input)) + c2 = re_string_peek_byte (input, 1); + else + c2 = 0; + token->opr.c = c2; + token_len = 2; + switch (c2) + { + case '.': + token->type = OP_OPEN_COLL_ELEM; + break; + case '=': + token->type = OP_OPEN_EQUIV_CLASS; + break; + case ':': + if (syntax & RE_CHAR_CLASSES) + { + token->type = OP_OPEN_CHAR_CLASS; + break; + } + /* else fall through. */ + default: + token->type = CHARACTER; + token->opr.c = c; + token_len = 1; + break; + } + return token_len; + } + switch (c) + { + case '-': + token->type = OP_CHARSET_RANGE; + break; + case ']': + token->type = OP_CLOSE_BRACKET; + break; + case '^': + token->type = OP_NON_MATCH_LIST; + break; + default: + token->type = CHARACTER; + } + return 1; +} + +/* Functions for parser. */ + +/* Entry point of the parser. + Parse the regular expression REGEXP and return the structure tree. + If an error occurs, ERR is set by error code, and return NULL. + This function build the following tree, from regular expression <reg_exp>: + CAT + / \ + / \ + <reg_exp> EOR + + CAT means concatenation. + EOR means end of regular expression. */ + +static bin_tree_t * +parse (re_string_t *regexp, regex_t *preg, reg_syntax_t syntax, + reg_errcode_t *err) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; + bin_tree_t *tree, *eor, *root; + re_token_t current_token; + dfa->syntax = syntax; + fetch_token (¤t_token, regexp, syntax | RE_CARET_ANCHORS_HERE); + tree = parse_reg_exp (regexp, preg, ¤t_token, syntax, 0, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + eor = create_tree (dfa, NULL, NULL, END_OF_RE); + if (tree != NULL) + root = create_tree (dfa, tree, eor, CONCAT); + else + root = eor; + if (BE (eor == NULL || root == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + return root; +} + +/* This function build the following tree, from regular expression + <branch1>|<branch2>: + ALT + / \ + / \ + <branch1> <branch2> + + ALT means alternative, which represents the operator '|'. */ + +static bin_tree_t * +parse_reg_exp (re_string_t *regexp, regex_t *preg, re_token_t *token, + reg_syntax_t syntax, Idx nest, reg_errcode_t *err) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; + bin_tree_t *tree, *branch = NULL; + tree = parse_branch (regexp, preg, token, syntax, nest, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + + while (token->type == OP_ALT) + { + fetch_token (token, regexp, syntax | RE_CARET_ANCHORS_HERE); + if (token->type != OP_ALT && token->type != END_OF_RE + && (nest == 0 || token->type != OP_CLOSE_SUBEXP)) + { + branch = parse_branch (regexp, preg, token, syntax, nest, err); + if (BE (*err != REG_NOERROR && branch == NULL, 0)) + return NULL; + } + else + branch = NULL; + tree = create_tree (dfa, tree, branch, OP_ALT); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + } + return tree; +} + +/* This function build the following tree, from regular expression + <exp1><exp2>: + CAT + / \ + / \ + <exp1> <exp2> + + CAT means concatenation. */ + +static bin_tree_t * +parse_branch (re_string_t *regexp, regex_t *preg, re_token_t *token, + reg_syntax_t syntax, Idx nest, reg_errcode_t *err) +{ + bin_tree_t *tree, *expr; + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; + tree = parse_expression (regexp, preg, token, syntax, nest, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + + while (token->type != OP_ALT && token->type != END_OF_RE + && (nest == 0 || token->type != OP_CLOSE_SUBEXP)) + { + expr = parse_expression (regexp, preg, token, syntax, nest, err); + if (BE (*err != REG_NOERROR && expr == NULL, 0)) + { + if (tree != NULL) + postorder (tree, free_tree, NULL); + return NULL; + } + if (tree != NULL && expr != NULL) + { + bin_tree_t *newtree = create_tree (dfa, tree, expr, CONCAT); + if (newtree == NULL) + { + postorder (expr, free_tree, NULL); + postorder (tree, free_tree, NULL); + *err = REG_ESPACE; + return NULL; + } + tree = newtree; + } + else if (tree == NULL) + tree = expr; + /* Otherwise expr == NULL, we don't need to create new tree. */ + } + return tree; +} + +/* This function build the following tree, from regular expression a*: + * + | + a +*/ + +static bin_tree_t * +parse_expression (re_string_t *regexp, regex_t *preg, re_token_t *token, + reg_syntax_t syntax, Idx nest, reg_errcode_t *err) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; + bin_tree_t *tree; + switch (token->type) + { + case CHARACTER: + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + { + while (!re_string_eoi (regexp) + && !re_string_first_byte (regexp, re_string_cur_idx (regexp))) + { + bin_tree_t *mbc_remain; + fetch_token (token, regexp, syntax); + mbc_remain = create_token_tree (dfa, NULL, NULL, token); + tree = create_tree (dfa, tree, mbc_remain, CONCAT); + if (BE (mbc_remain == NULL || tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + } + } +#endif + break; + case OP_OPEN_SUBEXP: + tree = parse_sub_exp (regexp, preg, token, syntax, nest + 1, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + break; + case OP_OPEN_BRACKET: + tree = parse_bracket_exp (regexp, dfa, token, syntax, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + break; + case OP_BACK_REF: + if (!BE (dfa->completed_bkref_map & (1 << token->opr.idx), 1)) + { + *err = REG_ESUBREG; + return NULL; + } + dfa->used_bkref_map |= 1 << token->opr.idx; + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + ++dfa->nbackref; + dfa->has_mb_node = 1; + break; + case OP_OPEN_DUP_NUM: + if (syntax & RE_CONTEXT_INVALID_DUP) + { + *err = REG_BADRPT; + return NULL; + } + /* FALLTHROUGH */ + case OP_DUP_ASTERISK: + case OP_DUP_PLUS: + case OP_DUP_QUESTION: + if (syntax & RE_CONTEXT_INVALID_OPS) + { + *err = REG_BADRPT; + return NULL; + } + else if (syntax & RE_CONTEXT_INDEP_OPS) + { + fetch_token (token, regexp, syntax); + return parse_expression (regexp, preg, token, syntax, nest, err); + } + /* else fall through */ + case OP_CLOSE_SUBEXP: + if ((token->type == OP_CLOSE_SUBEXP) && + !(syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)) + { + *err = REG_ERPAREN; + return NULL; + } + /* else fall through */ + case OP_CLOSE_DUP_NUM: + /* We treat it as a normal character. */ + + /* Then we can these characters as normal characters. */ + token->type = CHARACTER; + /* mb_partial and word_char bits should be initialized already + by peek_token. */ + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + break; + case ANCHOR: + if ((token->opr.ctx_type + & (WORD_DELIM | NOT_WORD_DELIM | WORD_FIRST | WORD_LAST)) + && dfa->word_ops_used == 0) + init_word_char (dfa); + if (token->opr.ctx_type == WORD_DELIM + || token->opr.ctx_type == NOT_WORD_DELIM) + { + bin_tree_t *tree_first, *tree_last; + if (token->opr.ctx_type == WORD_DELIM) + { + token->opr.ctx_type = WORD_FIRST; + tree_first = create_token_tree (dfa, NULL, NULL, token); + token->opr.ctx_type = WORD_LAST; + } + else + { + token->opr.ctx_type = INSIDE_WORD; + tree_first = create_token_tree (dfa, NULL, NULL, token); + token->opr.ctx_type = INSIDE_NOTWORD; + } + tree_last = create_token_tree (dfa, NULL, NULL, token); + tree = create_tree (dfa, tree_first, tree_last, OP_ALT); + if (BE (tree_first == NULL || tree_last == NULL || tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + } + else + { + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + } + /* We must return here, since ANCHORs can't be followed + by repetition operators. + eg. RE"^*" is invalid or "<ANCHOR(^)><CHAR(*)>", + it must not be "<ANCHOR(^)><REPEAT(*)>". */ + fetch_token (token, regexp, syntax); + return tree; + case OP_PERIOD: + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + if (dfa->mb_cur_max > 1) + dfa->has_mb_node = 1; + break; + case OP_WORD: + case OP_NOTWORD: + tree = build_charclass_op (dfa, regexp->trans, + (const unsigned char *) "alnum", + (const unsigned char *) "_", + token->type == OP_NOTWORD, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + break; + case OP_SPACE: + case OP_NOTSPACE: + tree = build_charclass_op (dfa, regexp->trans, + (const unsigned char *) "space", + (const unsigned char *) "", + token->type == OP_NOTSPACE, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + break; + case OP_ALT: + case END_OF_RE: + return NULL; + case BACK_SLASH: + *err = REG_EESCAPE; + return NULL; + default: + /* Must not happen? */ +#ifdef DEBUG + assert (0); +#endif + return NULL; + } + fetch_token (token, regexp, syntax); + + while (token->type == OP_DUP_ASTERISK || token->type == OP_DUP_PLUS + || token->type == OP_DUP_QUESTION || token->type == OP_OPEN_DUP_NUM) + { + tree = parse_dup_op (tree, regexp, dfa, token, syntax, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + /* In BRE consecutive duplications are not allowed. */ + if ((syntax & RE_CONTEXT_INVALID_DUP) + && (token->type == OP_DUP_ASTERISK + || token->type == OP_OPEN_DUP_NUM)) + { + *err = REG_BADRPT; + return NULL; + } + } + + return tree; +} + +/* This function build the following tree, from regular expression + (<reg_exp>): + SUBEXP + | + <reg_exp> +*/ + +static bin_tree_t * +parse_sub_exp (re_string_t *regexp, regex_t *preg, re_token_t *token, + reg_syntax_t syntax, Idx nest, reg_errcode_t *err) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; + bin_tree_t *tree; + size_t cur_nsub; + cur_nsub = preg->re_nsub++; + + fetch_token (token, regexp, syntax | RE_CARET_ANCHORS_HERE); + + /* The subexpression may be a null string. */ + if (token->type == OP_CLOSE_SUBEXP) + tree = NULL; + else + { + tree = parse_reg_exp (regexp, preg, token, syntax, nest, err); + if (BE (*err == REG_NOERROR && token->type != OP_CLOSE_SUBEXP, 0)) + { + if (tree != NULL) + postorder (tree, free_tree, NULL); + *err = REG_EPAREN; + } + if (BE (*err != REG_NOERROR, 0)) + return NULL; + } + + if (cur_nsub <= '9' - '1') + dfa->completed_bkref_map |= 1 << cur_nsub; + + tree = create_tree (dfa, tree, NULL, SUBEXP); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + tree->token.opr.idx = cur_nsub; + return tree; +} + +/* This function parse repetition operators like "*", "+", "{1,3}" etc. */ + +static bin_tree_t * +parse_dup_op (bin_tree_t *elem, re_string_t *regexp, re_dfa_t *dfa, + re_token_t *token, reg_syntax_t syntax, reg_errcode_t *err) +{ + bin_tree_t *tree = NULL, *old_tree = NULL; + Idx i, start, end, start_idx = re_string_cur_idx (regexp); + re_token_t start_token = *token; + + if (token->type == OP_OPEN_DUP_NUM) + { + end = 0; + start = fetch_number (regexp, token, syntax); + if (start == REG_MISSING) + { + if (token->type == CHARACTER && token->opr.c == ',') + start = 0; /* We treat "{,m}" as "{0,m}". */ + else + { + *err = REG_BADBR; /* <re>{} is invalid. */ + return NULL; + } + } + if (BE (start != REG_ERROR, 1)) + { + /* We treat "{n}" as "{n,n}". */ + end = ((token->type == OP_CLOSE_DUP_NUM) ? start + : ((token->type == CHARACTER && token->opr.c == ',') + ? fetch_number (regexp, token, syntax) : REG_ERROR)); + } + if (BE (start == REG_ERROR || end == REG_ERROR, 0)) + { + /* Invalid sequence. */ + if (BE (!(syntax & RE_INVALID_INTERVAL_ORD), 0)) + { + if (token->type == END_OF_RE) + *err = REG_EBRACE; + else + *err = REG_BADBR; + + return NULL; + } + + /* If the syntax bit is set, rollback. */ + re_string_set_index (regexp, start_idx); + *token = start_token; + token->type = CHARACTER; + /* mb_partial and word_char bits should be already initialized by + peek_token. */ + return elem; + } + + if (BE ((end != REG_MISSING && start > end) + || token->type != OP_CLOSE_DUP_NUM, 0)) + { + /* First number greater than second. */ + *err = REG_BADBR; + return NULL; + } + + if (BE (RE_DUP_MAX < (end == REG_MISSING ? start : end), 0)) + { + *err = REG_ESIZE; + return NULL; + } + } + else + { + start = (token->type == OP_DUP_PLUS) ? 1 : 0; + end = (token->type == OP_DUP_QUESTION) ? 1 : REG_MISSING; + } + + fetch_token (token, regexp, syntax); + + if (BE (elem == NULL, 0)) + return NULL; + if (BE (start == 0 && end == 0, 0)) + { + postorder (elem, free_tree, NULL); + return NULL; + } + + /* Extract "<re>{n,m}" to "<re><re>...<re><re>{0,<m-n>}". */ + if (BE (start > 0, 0)) + { + tree = elem; + for (i = 2; i <= start; ++i) + { + elem = duplicate_tree (elem, dfa); + tree = create_tree (dfa, tree, elem, CONCAT); + if (BE (elem == NULL || tree == NULL, 0)) + goto parse_dup_op_espace; + } + + if (start == end) + return tree; + + /* Duplicate ELEM before it is marked optional. */ + elem = duplicate_tree (elem, dfa); + old_tree = tree; + } + else + old_tree = NULL; + + if (elem->token.type == SUBEXP) + postorder (elem, mark_opt_subexp, (void *) (long) elem->token.opr.idx); + + tree = create_tree (dfa, elem, NULL, + (end == REG_MISSING ? OP_DUP_ASTERISK : OP_ALT)); + if (BE (tree == NULL, 0)) + goto parse_dup_op_espace; + +/* From gnulib's "intprops.h": + True if the arithmetic type T is signed. */ +#define TYPE_SIGNED(t) (! ((t) 0 < (t) -1)) + + /* This loop is actually executed only when end != REG_MISSING, + to rewrite <re>{0,n} as (<re>(<re>...<re>?)?)?... We have + already created the start+1-th copy. */ + if (TYPE_SIGNED (Idx) || end != REG_MISSING) + for (i = start + 2; i <= end; ++i) + { + elem = duplicate_tree (elem, dfa); + tree = create_tree (dfa, tree, elem, CONCAT); + if (BE (elem == NULL || tree == NULL, 0)) + goto parse_dup_op_espace; + + tree = create_tree (dfa, tree, NULL, OP_ALT); + if (BE (tree == NULL, 0)) + goto parse_dup_op_espace; + } + + if (old_tree) + tree = create_tree (dfa, old_tree, tree, CONCAT); + + return tree; + + parse_dup_op_espace: + *err = REG_ESPACE; + return NULL; +} + +/* Size of the names for collating symbol/equivalence_class/character_class. + I'm not sure, but maybe enough. */ +#define BRACKET_NAME_BUF_SIZE 32 + +#ifndef _LIBC + /* Local function for parse_bracket_exp only used in case of NOT _LIBC. + Build the range expression which starts from START_ELEM, and ends + at END_ELEM. The result are written to MBCSET and SBCSET. + RANGE_ALLOC is the allocated size of mbcset->range_starts, and + mbcset->range_ends, is a pointer argument since we may + update it. */ + +static reg_errcode_t +internal_function +# ifdef RE_ENABLE_I18N +build_range_exp (const reg_syntax_t syntax, + bitset_t sbcset, + re_charset_t *mbcset, + Idx *range_alloc, + const bracket_elem_t *start_elem, + const bracket_elem_t *end_elem) +# else /* not RE_ENABLE_I18N */ +build_range_exp (const reg_syntax_t syntax, + bitset_t sbcset, + const bracket_elem_t *start_elem, + const bracket_elem_t *end_elem) +# endif /* not RE_ENABLE_I18N */ +{ + unsigned int start_ch, end_ch; + /* Equivalence Classes and Character Classes can't be a range start/end. */ + if (BE (start_elem->type == EQUIV_CLASS || start_elem->type == CHAR_CLASS + || end_elem->type == EQUIV_CLASS || end_elem->type == CHAR_CLASS, + 0)) + return REG_ERANGE; + + /* We can handle no multi character collating elements without libc + support. */ + if (BE ((start_elem->type == COLL_SYM + && strlen ((char *) start_elem->opr.name) > 1) + || (end_elem->type == COLL_SYM + && strlen ((char *) end_elem->opr.name) > 1), 0)) + return REG_ECOLLATE; + +# ifdef RE_ENABLE_I18N + { + wchar_t wc; + wint_t start_wc; + wint_t end_wc; + wchar_t cmp_buf[6] = {L'\0', L'\0', L'\0', L'\0', L'\0', L'\0'}; + + start_ch = ((start_elem->type == SB_CHAR) ? start_elem->opr.ch + : ((start_elem->type == COLL_SYM) ? start_elem->opr.name[0] + : 0)); + end_ch = ((end_elem->type == SB_CHAR) ? end_elem->opr.ch + : ((end_elem->type == COLL_SYM) ? end_elem->opr.name[0] + : 0)); + start_wc = ((start_elem->type == SB_CHAR || start_elem->type == COLL_SYM) + ? __btowc (start_ch) : start_elem->opr.wch); + end_wc = ((end_elem->type == SB_CHAR || end_elem->type == COLL_SYM) + ? __btowc (end_ch) : end_elem->opr.wch); + if (start_wc == WEOF || end_wc == WEOF) + return REG_ECOLLATE; + cmp_buf[0] = start_wc; + cmp_buf[4] = end_wc; + + if (BE ((syntax & RE_NO_EMPTY_RANGES) + && wcscoll (cmp_buf, cmp_buf + 4) > 0, 0)) + return REG_ERANGE; + + /* Got valid collation sequence values, add them as a new entry. + However, for !_LIBC we have no collation elements: if the + character set is single byte, the single byte character set + that we build below suffices. parse_bracket_exp passes + no MBCSET if dfa->mb_cur_max == 1. */ + if (mbcset) + { + /* Check the space of the arrays. */ + if (BE (*range_alloc == mbcset->nranges, 0)) + { + /* There is not enough space, need realloc. */ + wchar_t *new_array_start, *new_array_end; + Idx new_nranges; + + /* +1 in case of mbcset->nranges is 0. */ + new_nranges = 2 * mbcset->nranges + 1; + /* Use realloc since mbcset->range_starts and mbcset->range_ends + are NULL if *range_alloc == 0. */ + new_array_start = re_realloc (mbcset->range_starts, wchar_t, + new_nranges); + new_array_end = re_realloc (mbcset->range_ends, wchar_t, + new_nranges); + + if (BE (new_array_start == NULL || new_array_end == NULL, 0)) + return REG_ESPACE; + + mbcset->range_starts = new_array_start; + mbcset->range_ends = new_array_end; + *range_alloc = new_nranges; + } + + mbcset->range_starts[mbcset->nranges] = start_wc; + mbcset->range_ends[mbcset->nranges++] = end_wc; + } + + /* Build the table for single byte characters. */ + for (wc = 0; wc < SBC_MAX; ++wc) + { + cmp_buf[2] = wc; + if (wcscoll (cmp_buf, cmp_buf + 2) <= 0 + && wcscoll (cmp_buf + 2, cmp_buf + 4) <= 0) + bitset_set (sbcset, wc); + } + } +# else /* not RE_ENABLE_I18N */ + { + unsigned int ch; + start_ch = ((start_elem->type == SB_CHAR ) ? start_elem->opr.ch + : ((start_elem->type == COLL_SYM) ? start_elem->opr.name[0] + : 0)); + end_ch = ((end_elem->type == SB_CHAR ) ? end_elem->opr.ch + : ((end_elem->type == COLL_SYM) ? end_elem->opr.name[0] + : 0)); + if (start_ch > end_ch) + return REG_ERANGE; + /* Build the table for single byte characters. */ + for (ch = 0; ch < SBC_MAX; ++ch) + if (start_ch <= ch && ch <= end_ch) + bitset_set (sbcset, ch); + } +# endif /* not RE_ENABLE_I18N */ + return REG_NOERROR; +} +#endif /* not _LIBC */ + +#ifndef _LIBC +/* Helper function for parse_bracket_exp only used in case of NOT _LIBC.. + Build the collating element which is represented by NAME. + The result are written to MBCSET and SBCSET. + COLL_SYM_ALLOC is the allocated size of mbcset->coll_sym, is a + pointer argument since we may update it. */ + +static reg_errcode_t +internal_function +# ifdef RE_ENABLE_I18N +build_collating_symbol (bitset_t sbcset, re_charset_t *mbcset, + Idx *coll_sym_alloc, const unsigned char *name) +# else /* not RE_ENABLE_I18N */ +build_collating_symbol (bitset_t sbcset, const unsigned char *name) +# endif /* not RE_ENABLE_I18N */ +{ + size_t name_len = strlen ((const char *) name); + if (BE (name_len != 1, 0)) + return REG_ECOLLATE; + else + { + bitset_set (sbcset, name[0]); + return REG_NOERROR; + } +} +#endif /* not _LIBC */ + +/* This function parse bracket expression like "[abc]", "[a-c]", + "[[.a-a.]]" etc. */ + +static bin_tree_t * +parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, + reg_syntax_t syntax, reg_errcode_t *err) +{ +#ifdef _LIBC + const unsigned char *collseqmb; + const char *collseqwc; + uint32_t nrules; + int32_t table_size; + const int32_t *symb_table; + const unsigned char *extra; + + /* Local function for parse_bracket_exp used in _LIBC environment. + Seek the collating symbol entry corresponding to NAME. + Return the index of the symbol in the SYMB_TABLE. */ + + auto inline int32_t + __attribute ((always_inline)) + seek_collating_symbol_entry (name, name_len) + const unsigned char *name; + size_t name_len; + { + int32_t hash = elem_hash ((const char *) name, name_len); + int32_t elem = hash % table_size; + if (symb_table[2 * elem] != 0) + { + int32_t second = hash % (table_size - 2) + 1; + + do + { + /* First compare the hashing value. */ + if (symb_table[2 * elem] == hash + /* Compare the length of the name. */ + && name_len == extra[symb_table[2 * elem + 1]] + /* Compare the name. */ + && memcmp (name, &extra[symb_table[2 * elem + 1] + 1], + name_len) == 0) + { + /* Yep, this is the entry. */ + break; + } + + /* Next entry. */ + elem += second; + } + while (symb_table[2 * elem] != 0); + } + return elem; + } + + /* Local function for parse_bracket_exp used in _LIBC environment. + Look up the collation sequence value of BR_ELEM. + Return the value if succeeded, UINT_MAX otherwise. */ + + auto inline unsigned int + __attribute ((always_inline)) + lookup_collation_sequence_value (br_elem) + bracket_elem_t *br_elem; + { + if (br_elem->type == SB_CHAR) + { + /* + if (MB_CUR_MAX == 1) + */ + if (nrules == 0) + return collseqmb[br_elem->opr.ch]; + else + { + wint_t wc = __btowc (br_elem->opr.ch); + return __collseq_table_lookup (collseqwc, wc); + } + } + else if (br_elem->type == MB_CHAR) + { + if (nrules != 0) + return __collseq_table_lookup (collseqwc, br_elem->opr.wch); + } + else if (br_elem->type == COLL_SYM) + { + size_t sym_name_len = strlen ((char *) br_elem->opr.name); + if (nrules != 0) + { + int32_t elem, idx; + elem = seek_collating_symbol_entry (br_elem->opr.name, + sym_name_len); + if (symb_table[2 * elem] != 0) + { + /* We found the entry. */ + idx = symb_table[2 * elem + 1]; + /* Skip the name of collating element name. */ + idx += 1 + extra[idx]; + /* Skip the byte sequence of the collating element. */ + idx += 1 + extra[idx]; + /* Adjust for the alignment. */ + idx = (idx + 3) & ~3; + /* Skip the multibyte collation sequence value. */ + idx += sizeof (unsigned int); + /* Skip the wide char sequence of the collating element. */ + idx += sizeof (unsigned int) * + (1 + *(unsigned int *) (extra + idx)); + /* Return the collation sequence value. */ + return *(unsigned int *) (extra + idx); + } + else if (symb_table[2 * elem] == 0 && sym_name_len == 1) + { + /* No valid character. Match it as a single byte + character. */ + return collseqmb[br_elem->opr.name[0]]; + } + } + else if (sym_name_len == 1) + return collseqmb[br_elem->opr.name[0]]; + } + return UINT_MAX; + } + + /* Local function for parse_bracket_exp used in _LIBC environment. + Build the range expression which starts from START_ELEM, and ends + at END_ELEM. The result are written to MBCSET and SBCSET. + RANGE_ALLOC is the allocated size of mbcset->range_starts, and + mbcset->range_ends, is a pointer argument since we may + update it. */ + + auto inline reg_errcode_t + __attribute ((always_inline)) + build_range_exp (sbcset, mbcset, range_alloc, start_elem, end_elem) + re_charset_t *mbcset; + Idx *range_alloc; + bitset_t sbcset; + bracket_elem_t *start_elem, *end_elem; + { + unsigned int ch; + uint32_t start_collseq; + uint32_t end_collseq; + + /* Equivalence Classes and Character Classes can't be a range + start/end. */ + if (BE (start_elem->type == EQUIV_CLASS || start_elem->type == CHAR_CLASS + || end_elem->type == EQUIV_CLASS || end_elem->type == CHAR_CLASS, + 0)) + return REG_ERANGE; + + start_collseq = lookup_collation_sequence_value (start_elem); + end_collseq = lookup_collation_sequence_value (end_elem); + /* Check start/end collation sequence values. */ + if (BE (start_collseq == UINT_MAX || end_collseq == UINT_MAX, 0)) + return REG_ECOLLATE; + if (BE ((syntax & RE_NO_EMPTY_RANGES) && start_collseq > end_collseq, 0)) + return REG_ERANGE; + + /* Got valid collation sequence values, add them as a new entry. + However, if we have no collation elements, and the character set + is single byte, the single byte character set that we + build below suffices. */ + if (nrules > 0 || dfa->mb_cur_max > 1) + { + /* Check the space of the arrays. */ + if (BE (*range_alloc == mbcset->nranges, 0)) + { + /* There is not enough space, need realloc. */ + uint32_t *new_array_start; + uint32_t *new_array_end; + Idx new_nranges; + + /* +1 in case of mbcset->nranges is 0. */ + new_nranges = 2 * mbcset->nranges + 1; + new_array_start = re_realloc (mbcset->range_starts, uint32_t, + new_nranges); + new_array_end = re_realloc (mbcset->range_ends, uint32_t, + new_nranges); + + if (BE (new_array_start == NULL || new_array_end == NULL, 0)) + return REG_ESPACE; + + mbcset->range_starts = new_array_start; + mbcset->range_ends = new_array_end; + *range_alloc = new_nranges; + } + + mbcset->range_starts[mbcset->nranges] = start_collseq; + mbcset->range_ends[mbcset->nranges++] = end_collseq; + } + + /* Build the table for single byte characters. */ + for (ch = 0; ch < SBC_MAX; ch++) + { + uint32_t ch_collseq; + /* + if (MB_CUR_MAX == 1) + */ + if (nrules == 0) + ch_collseq = collseqmb[ch]; + else + ch_collseq = __collseq_table_lookup (collseqwc, __btowc (ch)); + if (start_collseq <= ch_collseq && ch_collseq <= end_collseq) + bitset_set (sbcset, ch); + } + return REG_NOERROR; + } + + /* Local function for parse_bracket_exp used in _LIBC environment. + Build the collating element which is represented by NAME. + The result are written to MBCSET and SBCSET. + COLL_SYM_ALLOC is the allocated size of mbcset->coll_sym, is a + pointer argument since we may update it. */ + + auto inline reg_errcode_t + __attribute ((always_inline)) + build_collating_symbol (sbcset, mbcset, coll_sym_alloc, name) + re_charset_t *mbcset; + Idx *coll_sym_alloc; + bitset_t sbcset; + const unsigned char *name; + { + int32_t elem, idx; + size_t name_len = strlen ((const char *) name); + if (nrules != 0) + { + elem = seek_collating_symbol_entry (name, name_len); + if (symb_table[2 * elem] != 0) + { + /* We found the entry. */ + idx = symb_table[2 * elem + 1]; + /* Skip the name of collating element name. */ + idx += 1 + extra[idx]; + } + else if (symb_table[2 * elem] == 0 && name_len == 1) + { + /* No valid character, treat it as a normal + character. */ + bitset_set (sbcset, name[0]); + return REG_NOERROR; + } + else + return REG_ECOLLATE; + + /* Got valid collation sequence, add it as a new entry. */ + /* Check the space of the arrays. */ + if (BE (*coll_sym_alloc == mbcset->ncoll_syms, 0)) + { + /* Not enough, realloc it. */ + /* +1 in case of mbcset->ncoll_syms is 0. */ + Idx new_coll_sym_alloc = 2 * mbcset->ncoll_syms + 1; + /* Use realloc since mbcset->coll_syms is NULL + if *alloc == 0. */ + int32_t *new_coll_syms = re_realloc (mbcset->coll_syms, int32_t, + new_coll_sym_alloc); + if (BE (new_coll_syms == NULL, 0)) + return REG_ESPACE; + mbcset->coll_syms = new_coll_syms; + *coll_sym_alloc = new_coll_sym_alloc; + } + mbcset->coll_syms[mbcset->ncoll_syms++] = idx; + return REG_NOERROR; + } + else + { + if (BE (name_len != 1, 0)) + return REG_ECOLLATE; + else + { + bitset_set (sbcset, name[0]); + return REG_NOERROR; + } + } + } +#endif + + re_token_t br_token; + re_bitset_ptr_t sbcset; +#ifdef RE_ENABLE_I18N + re_charset_t *mbcset; + Idx coll_sym_alloc = 0, range_alloc = 0, mbchar_alloc = 0; + Idx equiv_class_alloc = 0, char_class_alloc = 0; +#endif /* not RE_ENABLE_I18N */ + bool non_match = false; + bin_tree_t *work_tree; + int token_len; + bool first_round = true; +#ifdef _LIBC + collseqmb = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQMB); + nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + if (nrules) + { + /* + if (MB_CUR_MAX > 1) + */ + collseqwc = _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQWC); + table_size = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_SYMB_HASH_SIZEMB); + symb_table = (const int32_t *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_SYMB_TABLEMB); + extra = (const unsigned char *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_SYMB_EXTRAMB); + } +#endif + sbcset = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1); +#ifdef RE_ENABLE_I18N + mbcset = (re_charset_t *) calloc (sizeof (re_charset_t), 1); +#endif /* RE_ENABLE_I18N */ +#ifdef RE_ENABLE_I18N + if (BE (sbcset == NULL || mbcset == NULL, 0)) +#else + if (BE (sbcset == NULL, 0)) +#endif /* RE_ENABLE_I18N */ + { + re_free (sbcset); +#ifdef RE_ENABLE_I18N + re_free (mbcset); +#endif + *err = REG_ESPACE; + return NULL; + } + + token_len = peek_token_bracket (token, regexp, syntax); + if (BE (token->type == END_OF_RE, 0)) + { + *err = REG_BADPAT; + goto parse_bracket_exp_free_return; + } + if (token->type == OP_NON_MATCH_LIST) + { +#ifdef RE_ENABLE_I18N + mbcset->non_match = 1; +#endif /* not RE_ENABLE_I18N */ + non_match = true; + if (syntax & RE_HAT_LISTS_NOT_NEWLINE) + bitset_set (sbcset, '\n'); + re_string_skip_bytes (regexp, token_len); /* Skip a token. */ + token_len = peek_token_bracket (token, regexp, syntax); + if (BE (token->type == END_OF_RE, 0)) + { + *err = REG_BADPAT; + goto parse_bracket_exp_free_return; + } + } + + /* We treat the first ']' as a normal character. */ + if (token->type == OP_CLOSE_BRACKET) + token->type = CHARACTER; + + while (1) + { + bracket_elem_t start_elem, end_elem; + unsigned char start_name_buf[BRACKET_NAME_BUF_SIZE]; + unsigned char end_name_buf[BRACKET_NAME_BUF_SIZE]; + reg_errcode_t ret; + int token_len2 = 0; + bool is_range_exp = false; + re_token_t token2; + + start_elem.opr.name = start_name_buf; + ret = parse_bracket_element (&start_elem, regexp, token, token_len, dfa, + syntax, first_round); + if (BE (ret != REG_NOERROR, 0)) + { + *err = ret; + goto parse_bracket_exp_free_return; + } + first_round = false; + + /* Get information about the next token. We need it in any case. */ + token_len = peek_token_bracket (token, regexp, syntax); + + /* Do not check for ranges if we know they are not allowed. */ + if (start_elem.type != CHAR_CLASS && start_elem.type != EQUIV_CLASS) + { + if (BE (token->type == END_OF_RE, 0)) + { + *err = REG_EBRACK; + goto parse_bracket_exp_free_return; + } + if (token->type == OP_CHARSET_RANGE) + { + re_string_skip_bytes (regexp, token_len); /* Skip '-'. */ + token_len2 = peek_token_bracket (&token2, regexp, syntax); + if (BE (token2.type == END_OF_RE, 0)) + { + *err = REG_EBRACK; + goto parse_bracket_exp_free_return; + } + if (token2.type == OP_CLOSE_BRACKET) + { + /* We treat the last '-' as a normal character. */ + re_string_skip_bytes (regexp, -token_len); + token->type = CHARACTER; + } + else + is_range_exp = true; + } + } + + if (is_range_exp == true) + { + end_elem.opr.name = end_name_buf; + ret = parse_bracket_element (&end_elem, regexp, &token2, token_len2, + dfa, syntax, true); + if (BE (ret != REG_NOERROR, 0)) + { + *err = ret; + goto parse_bracket_exp_free_return; + } + + token_len = peek_token_bracket (token, regexp, syntax); + +#ifdef _LIBC + *err = build_range_exp (sbcset, mbcset, &range_alloc, + &start_elem, &end_elem); +#else +# ifdef RE_ENABLE_I18N + *err = build_range_exp (syntax, sbcset, + dfa->mb_cur_max > 1 ? mbcset : NULL, + &range_alloc, &start_elem, &end_elem); +# else + *err = build_range_exp (syntax, sbcset, &start_elem, &end_elem); +# endif +#endif /* RE_ENABLE_I18N */ + if (BE (*err != REG_NOERROR, 0)) + goto parse_bracket_exp_free_return; + } + else + { + switch (start_elem.type) + { + case SB_CHAR: + bitset_set (sbcset, start_elem.opr.ch); + break; +#ifdef RE_ENABLE_I18N + case MB_CHAR: + /* Check whether the array has enough space. */ + if (BE (mbchar_alloc == mbcset->nmbchars, 0)) + { + wchar_t *new_mbchars; + /* Not enough, realloc it. */ + /* +1 in case of mbcset->nmbchars is 0. */ + mbchar_alloc = 2 * mbcset->nmbchars + 1; + /* Use realloc since array is NULL if *alloc == 0. */ + new_mbchars = re_realloc (mbcset->mbchars, wchar_t, + mbchar_alloc); + if (BE (new_mbchars == NULL, 0)) + goto parse_bracket_exp_espace; + mbcset->mbchars = new_mbchars; + } + mbcset->mbchars[mbcset->nmbchars++] = start_elem.opr.wch; + break; +#endif /* RE_ENABLE_I18N */ + case EQUIV_CLASS: + *err = build_equiv_class (sbcset, +#ifdef RE_ENABLE_I18N + mbcset, &equiv_class_alloc, +#endif /* RE_ENABLE_I18N */ + start_elem.opr.name); + if (BE (*err != REG_NOERROR, 0)) + goto parse_bracket_exp_free_return; + break; + case COLL_SYM: + *err = build_collating_symbol (sbcset, +#ifdef RE_ENABLE_I18N + mbcset, &coll_sym_alloc, +#endif /* RE_ENABLE_I18N */ + start_elem.opr.name); + if (BE (*err != REG_NOERROR, 0)) + goto parse_bracket_exp_free_return; + break; + case CHAR_CLASS: + *err = build_charclass (regexp->trans, sbcset, +#ifdef RE_ENABLE_I18N + mbcset, &char_class_alloc, +#endif /* RE_ENABLE_I18N */ + start_elem.opr.name, syntax); + if (BE (*err != REG_NOERROR, 0)) + goto parse_bracket_exp_free_return; + break; + default: + assert (0); + break; + } + } + if (BE (token->type == END_OF_RE, 0)) + { + *err = REG_EBRACK; + goto parse_bracket_exp_free_return; + } + if (token->type == OP_CLOSE_BRACKET) + break; + } + + re_string_skip_bytes (regexp, token_len); /* Skip a token. */ + + /* If it is non-matching list. */ + if (non_match) + bitset_not (sbcset); + +#ifdef RE_ENABLE_I18N + /* Ensure only single byte characters are set. */ + if (dfa->mb_cur_max > 1) + bitset_mask (sbcset, dfa->sb_char); + + if (mbcset->nmbchars || mbcset->ncoll_syms || mbcset->nequiv_classes + || mbcset->nranges || (dfa->mb_cur_max > 1 && (mbcset->nchar_classes + || mbcset->non_match))) + { + bin_tree_t *mbc_tree; + int sbc_idx; + /* Build a tree for complex bracket. */ + dfa->has_mb_node = 1; + br_token.type = COMPLEX_BRACKET; + br_token.opr.mbcset = mbcset; + mbc_tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (mbc_tree == NULL, 0)) + goto parse_bracket_exp_espace; + for (sbc_idx = 0; sbc_idx < BITSET_WORDS; ++sbc_idx) + if (sbcset[sbc_idx]) + break; + /* If there are no bits set in sbcset, there is no point + of having both SIMPLE_BRACKET and COMPLEX_BRACKET. */ + if (sbc_idx < BITSET_WORDS) + { + /* Build a tree for simple bracket. */ + br_token.type = SIMPLE_BRACKET; + br_token.opr.sbcset = sbcset; + work_tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (work_tree == NULL, 0)) + goto parse_bracket_exp_espace; + + /* Then join them by ALT node. */ + work_tree = create_tree (dfa, work_tree, mbc_tree, OP_ALT); + if (BE (work_tree == NULL, 0)) + goto parse_bracket_exp_espace; + } + else + { + re_free (sbcset); + work_tree = mbc_tree; + } + } + else +#endif /* not RE_ENABLE_I18N */ + { +#ifdef RE_ENABLE_I18N + free_charset (mbcset); +#endif + /* Build a tree for simple bracket. */ + br_token.type = SIMPLE_BRACKET; + br_token.opr.sbcset = sbcset; + work_tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (work_tree == NULL, 0)) + goto parse_bracket_exp_espace; + } + return work_tree; + + parse_bracket_exp_espace: + *err = REG_ESPACE; + parse_bracket_exp_free_return: + re_free (sbcset); +#ifdef RE_ENABLE_I18N + free_charset (mbcset); +#endif /* RE_ENABLE_I18N */ + return NULL; +} + +/* Parse an element in the bracket expression. */ + +static reg_errcode_t +parse_bracket_element (bracket_elem_t *elem, re_string_t *regexp, + re_token_t *token, int token_len, re_dfa_t *dfa, + reg_syntax_t syntax, bool accept_hyphen) +{ +#ifdef RE_ENABLE_I18N + int cur_char_size; + cur_char_size = re_string_char_size_at (regexp, re_string_cur_idx (regexp)); + if (cur_char_size > 1) + { + elem->type = MB_CHAR; + elem->opr.wch = re_string_wchar_at (regexp, re_string_cur_idx (regexp)); + re_string_skip_bytes (regexp, cur_char_size); + return REG_NOERROR; + } +#endif /* RE_ENABLE_I18N */ + re_string_skip_bytes (regexp, token_len); /* Skip a token. */ + if (token->type == OP_OPEN_COLL_ELEM || token->type == OP_OPEN_CHAR_CLASS + || token->type == OP_OPEN_EQUIV_CLASS) + return parse_bracket_symbol (elem, regexp, token); + if (BE (token->type == OP_CHARSET_RANGE, 0) && !accept_hyphen) + { + /* A '-' must only appear as anything but a range indicator before + the closing bracket. Everything else is an error. */ + re_token_t token2; + (void) peek_token_bracket (&token2, regexp, syntax); + if (token2.type != OP_CLOSE_BRACKET) + /* The actual error value is not standardized since this whole + case is undefined. But ERANGE makes good sense. */ + return REG_ERANGE; + } + elem->type = SB_CHAR; + elem->opr.ch = token->opr.c; + return REG_NOERROR; +} + +/* Parse a bracket symbol in the bracket expression. Bracket symbols are + such as [:<character_class>:], [.<collating_element>.], and + [=<equivalent_class>=]. */ + +static reg_errcode_t +parse_bracket_symbol (bracket_elem_t *elem, re_string_t *regexp, + re_token_t *token) +{ + unsigned char ch, delim = token->opr.c; + int i = 0; + if (re_string_eoi(regexp)) + return REG_EBRACK; + for (;; ++i) + { + if (i >= BRACKET_NAME_BUF_SIZE) + return REG_EBRACK; + if (token->type == OP_OPEN_CHAR_CLASS) + ch = re_string_fetch_byte_case (regexp); + else + ch = re_string_fetch_byte (regexp); + if (re_string_eoi(regexp)) + return REG_EBRACK; + if (ch == delim && re_string_peek_byte (regexp, 0) == ']') + break; + elem->opr.name[i] = ch; + } + re_string_skip_bytes (regexp, 1); + elem->opr.name[i] = '\0'; + switch (token->type) + { + case OP_OPEN_COLL_ELEM: + elem->type = COLL_SYM; + break; + case OP_OPEN_EQUIV_CLASS: + elem->type = EQUIV_CLASS; + break; + case OP_OPEN_CHAR_CLASS: + elem->type = CHAR_CLASS; + break; + default: + break; + } + return REG_NOERROR; +} + + /* Helper function for parse_bracket_exp. + Build the equivalence class which is represented by NAME. + The result are written to MBCSET and SBCSET. + EQUIV_CLASS_ALLOC is the allocated size of mbcset->equiv_classes, + is a pointer argument since we may update it. */ + +static reg_errcode_t +#ifdef RE_ENABLE_I18N +build_equiv_class (bitset_t sbcset, re_charset_t *mbcset, + Idx *equiv_class_alloc, const unsigned char *name) +#else /* not RE_ENABLE_I18N */ +build_equiv_class (bitset_t sbcset, const unsigned char *name) +#endif /* not RE_ENABLE_I18N */ +{ +#ifdef _LIBC + uint32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + if (nrules != 0) + { + const int32_t *table, *indirect; + const unsigned char *weights, *extra, *cp; + unsigned char char_buf[2]; + int32_t idx1, idx2; + unsigned int ch; + size_t len; + /* This #include defines a local function! */ +# include <locale/weight.h> + /* Calculate the index for equivalence class. */ + cp = name; + table = (const int32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB); + weights = (const unsigned char *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_WEIGHTMB); + extra = (const unsigned char *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_EXTRAMB); + indirect = (const int32_t *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_INDIRECTMB); + idx1 = findidx (&cp, -1); + if (BE (idx1 == 0 || *cp != '\0', 0)) + /* This isn't a valid character. */ + return REG_ECOLLATE; + + /* Build single byte matching table for this equivalence class. */ + len = weights[idx1 & 0xffffff]; + for (ch = 0; ch < SBC_MAX; ++ch) + { + char_buf[0] = ch; + cp = char_buf; + idx2 = findidx (&cp, 1); +/* + idx2 = table[ch]; +*/ + if (idx2 == 0) + /* This isn't a valid character. */ + continue; + /* Compare only if the length matches and the collation rule + index is the same. */ + if (len == weights[idx2 & 0xffffff] && (idx1 >> 24) == (idx2 >> 24)) + { + int cnt = 0; + + while (cnt <= len && + weights[(idx1 & 0xffffff) + 1 + cnt] + == weights[(idx2 & 0xffffff) + 1 + cnt]) + ++cnt; + + if (cnt > len) + bitset_set (sbcset, ch); + } + } + /* Check whether the array has enough space. */ + if (BE (*equiv_class_alloc == mbcset->nequiv_classes, 0)) + { + /* Not enough, realloc it. */ + /* +1 in case of mbcset->nequiv_classes is 0. */ + Idx new_equiv_class_alloc = 2 * mbcset->nequiv_classes + 1; + /* Use realloc since the array is NULL if *alloc == 0. */ + int32_t *new_equiv_classes = re_realloc (mbcset->equiv_classes, + int32_t, + new_equiv_class_alloc); + if (BE (new_equiv_classes == NULL, 0)) + return REG_ESPACE; + mbcset->equiv_classes = new_equiv_classes; + *equiv_class_alloc = new_equiv_class_alloc; + } + mbcset->equiv_classes[mbcset->nequiv_classes++] = idx1; + } + else +#endif /* _LIBC */ + { + if (BE (strlen ((const char *) name) != 1, 0)) + return REG_ECOLLATE; + bitset_set (sbcset, *name); + } + return REG_NOERROR; +} + + /* Helper function for parse_bracket_exp. + Build the character class which is represented by NAME. + The result are written to MBCSET and SBCSET. + CHAR_CLASS_ALLOC is the allocated size of mbcset->char_classes, + is a pointer argument since we may update it. */ + +static reg_errcode_t +#ifdef RE_ENABLE_I18N +build_charclass (RE_TRANSLATE_TYPE trans, bitset_t sbcset, + re_charset_t *mbcset, Idx *char_class_alloc, + const unsigned char *class_name, reg_syntax_t syntax) +#else /* not RE_ENABLE_I18N */ +build_charclass (RE_TRANSLATE_TYPE trans, bitset_t sbcset, + const unsigned char *class_name, reg_syntax_t syntax) +#endif /* not RE_ENABLE_I18N */ +{ + int i; + const char *name = (const char *) class_name; + + /* In case of REG_ICASE "upper" and "lower" match the both of + upper and lower cases. */ + if ((syntax & RE_ICASE) + && (strcmp (name, "upper") == 0 || strcmp (name, "lower") == 0)) + name = "alpha"; + +#ifdef RE_ENABLE_I18N + /* Check the space of the arrays. */ + if (BE (*char_class_alloc == mbcset->nchar_classes, 0)) + { + /* Not enough, realloc it. */ + /* +1 in case of mbcset->nchar_classes is 0. */ + Idx new_char_class_alloc = 2 * mbcset->nchar_classes + 1; + /* Use realloc since array is NULL if *alloc == 0. */ + wctype_t *new_char_classes = re_realloc (mbcset->char_classes, wctype_t, + new_char_class_alloc); + if (BE (new_char_classes == NULL, 0)) + return REG_ESPACE; + mbcset->char_classes = new_char_classes; + *char_class_alloc = new_char_class_alloc; + } + mbcset->char_classes[mbcset->nchar_classes++] = __wctype (name); +#endif /* RE_ENABLE_I18N */ + +#define BUILD_CHARCLASS_LOOP(ctype_func) \ + do { \ + if (BE (trans != NULL, 0)) \ + { \ + for (i = 0; i < SBC_MAX; ++i) \ + if (ctype_func (i)) \ + bitset_set (sbcset, trans[i]); \ + } \ + else \ + { \ + for (i = 0; i < SBC_MAX; ++i) \ + if (ctype_func (i)) \ + bitset_set (sbcset, i); \ + } \ + } while (0) + + if (strcmp (name, "alnum") == 0) + BUILD_CHARCLASS_LOOP (isalnum); + else if (strcmp (name, "cntrl") == 0) + BUILD_CHARCLASS_LOOP (iscntrl); + else if (strcmp (name, "lower") == 0) + BUILD_CHARCLASS_LOOP (islower); + else if (strcmp (name, "space") == 0) + BUILD_CHARCLASS_LOOP (isspace); + else if (strcmp (name, "alpha") == 0) + BUILD_CHARCLASS_LOOP (isalpha); + else if (strcmp (name, "digit") == 0) + BUILD_CHARCLASS_LOOP (isdigit); + else if (strcmp (name, "print") == 0) + BUILD_CHARCLASS_LOOP (isprint); + else if (strcmp (name, "upper") == 0) + BUILD_CHARCLASS_LOOP (isupper); + else if (strcmp (name, "blank") == 0) + BUILD_CHARCLASS_LOOP (isblank); + else if (strcmp (name, "graph") == 0) + BUILD_CHARCLASS_LOOP (isgraph); + else if (strcmp (name, "punct") == 0) + BUILD_CHARCLASS_LOOP (ispunct); + else if (strcmp (name, "xdigit") == 0) + BUILD_CHARCLASS_LOOP (isxdigit); + else + return REG_ECTYPE; + + return REG_NOERROR; +} + +static bin_tree_t * +build_charclass_op (re_dfa_t *dfa, RE_TRANSLATE_TYPE trans, + const unsigned char *class_name, + const unsigned char *extra, bool non_match, + reg_errcode_t *err) +{ + re_bitset_ptr_t sbcset; +#ifdef RE_ENABLE_I18N + re_charset_t *mbcset; + Idx alloc = 0; +#endif /* not RE_ENABLE_I18N */ + reg_errcode_t ret; + re_token_t br_token; + bin_tree_t *tree; + + sbcset = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1); +#ifdef RE_ENABLE_I18N + mbcset = (re_charset_t *) calloc (sizeof (re_charset_t), 1); +#endif /* RE_ENABLE_I18N */ + +#ifdef RE_ENABLE_I18N + if (BE (sbcset == NULL || mbcset == NULL, 0)) +#else /* not RE_ENABLE_I18N */ + if (BE (sbcset == NULL, 0)) +#endif /* not RE_ENABLE_I18N */ + { + *err = REG_ESPACE; + return NULL; + } + + if (non_match) + { +#ifdef RE_ENABLE_I18N + mbcset->non_match = 1; +#endif /* not RE_ENABLE_I18N */ + } + + /* We don't care the syntax in this case. */ + ret = build_charclass (trans, sbcset, +#ifdef RE_ENABLE_I18N + mbcset, &alloc, +#endif /* RE_ENABLE_I18N */ + class_name, 0); + + if (BE (ret != REG_NOERROR, 0)) + { + re_free (sbcset); +#ifdef RE_ENABLE_I18N + free_charset (mbcset); +#endif /* RE_ENABLE_I18N */ + *err = ret; + return NULL; + } + /* \w match '_' also. */ + for (; *extra; extra++) + bitset_set (sbcset, *extra); + + /* If it is non-matching list. */ + if (non_match) + bitset_not (sbcset); + +#ifdef RE_ENABLE_I18N + /* Ensure only single byte characters are set. */ + if (dfa->mb_cur_max > 1) + bitset_mask (sbcset, dfa->sb_char); +#endif + + /* Build a tree for simple bracket. */ + br_token.type = SIMPLE_BRACKET; + br_token.opr.sbcset = sbcset; + tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (tree == NULL, 0)) + goto build_word_op_espace; + +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + { + bin_tree_t *mbc_tree; + /* Build a tree for complex bracket. */ + br_token.type = COMPLEX_BRACKET; + br_token.opr.mbcset = mbcset; + dfa->has_mb_node = 1; + mbc_tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (mbc_tree == NULL, 0)) + goto build_word_op_espace; + /* Then join them by ALT node. */ + tree = create_tree (dfa, tree, mbc_tree, OP_ALT); + if (BE (mbc_tree != NULL, 1)) + return tree; + } + else + { + free_charset (mbcset); + return tree; + } +#else /* not RE_ENABLE_I18N */ + return tree; +#endif /* not RE_ENABLE_I18N */ + + build_word_op_espace: + re_free (sbcset); +#ifdef RE_ENABLE_I18N + free_charset (mbcset); +#endif /* RE_ENABLE_I18N */ + *err = REG_ESPACE; + return NULL; +} + +/* This is intended for the expressions like "a{1,3}". + Fetch a number from 'input', and return the number. + Return REG_MISSING if the number field is empty like "{,1}". + Return RE_DUP_MAX + 1 if the number field is too large. + Return REG_ERROR if an error occurred. */ + +static Idx +fetch_number (re_string_t *input, re_token_t *token, reg_syntax_t syntax) +{ + Idx num = REG_MISSING; + unsigned char c; + while (1) + { + fetch_token (token, input, syntax); + c = token->opr.c; + if (BE (token->type == END_OF_RE, 0)) + return REG_ERROR; + if (token->type == OP_CLOSE_DUP_NUM || c == ',') + break; + num = ((token->type != CHARACTER || c < '0' || '9' < c + || num == REG_ERROR) + ? REG_ERROR + : num == REG_MISSING + ? c - '0' + : MIN (RE_DUP_MAX + 1, num * 10 + c - '0')); + } + return num; +} + +#ifdef RE_ENABLE_I18N +static void +free_charset (re_charset_t *cset) +{ + re_free (cset->mbchars); +# ifdef _LIBC + re_free (cset->coll_syms); + re_free (cset->equiv_classes); + re_free (cset->range_starts); + re_free (cset->range_ends); +# endif + re_free (cset->char_classes); + re_free (cset); +} +#endif /* RE_ENABLE_I18N */ + +/* Functions for binary tree operation. */ + +/* Create a tree node. */ + +static bin_tree_t * +create_tree (re_dfa_t *dfa, bin_tree_t *left, bin_tree_t *right, + re_token_type_t type) +{ + re_token_t t; + t.type = type; + return create_token_tree (dfa, left, right, &t); +} + +static bin_tree_t * +create_token_tree (re_dfa_t *dfa, bin_tree_t *left, bin_tree_t *right, + const re_token_t *token) +{ + bin_tree_t *tree; + if (BE (dfa->str_tree_storage_idx == BIN_TREE_STORAGE_SIZE, 0)) + { + bin_tree_storage_t *storage = re_malloc (bin_tree_storage_t, 1); + + if (storage == NULL) + return NULL; + storage->next = dfa->str_tree_storage; + dfa->str_tree_storage = storage; + dfa->str_tree_storage_idx = 0; + } + tree = &dfa->str_tree_storage->data[dfa->str_tree_storage_idx++]; + + tree->parent = NULL; + tree->left = left; + tree->right = right; + tree->token = *token; + tree->token.duplicated = 0; + tree->token.opt_subexp = 0; + tree->first = NULL; + tree->next = NULL; + tree->node_idx = REG_MISSING; + + if (left != NULL) + left->parent = tree; + if (right != NULL) + right->parent = tree; + return tree; +} + +/* Mark the tree SRC as an optional subexpression. + To be called from preorder or postorder. */ + +static reg_errcode_t +mark_opt_subexp (void *extra, bin_tree_t *node) +{ + Idx idx = (Idx) (long) extra; + if (node->token.type == SUBEXP && node->token.opr.idx == idx) + node->token.opt_subexp = 1; + + return REG_NOERROR; +} + +/* Free the allocated memory inside NODE. */ + +static void +free_token (re_token_t *node) +{ +#ifdef RE_ENABLE_I18N + if (node->type == COMPLEX_BRACKET && node->duplicated == 0) + free_charset (node->opr.mbcset); + else +#endif /* RE_ENABLE_I18N */ + if (node->type == SIMPLE_BRACKET && node->duplicated == 0) + re_free (node->opr.sbcset); +} + +/* Worker function for tree walking. Free the allocated memory inside NODE + and its children. */ + +static reg_errcode_t +free_tree (void *extra, bin_tree_t *node) +{ + free_token (&node->token); + return REG_NOERROR; +} + + +/* Duplicate the node SRC, and return new node. This is a preorder + visit similar to the one implemented by the generic visitor, but + we need more infrastructure to maintain two parallel trees --- so, + it's easier to duplicate. */ + +static bin_tree_t * +duplicate_tree (const bin_tree_t *root, re_dfa_t *dfa) +{ + const bin_tree_t *node; + bin_tree_t *dup_root; + bin_tree_t **p_new = &dup_root, *dup_node = root->parent; + + for (node = root; ; ) + { + /* Create a new tree and link it back to the current parent. */ + *p_new = create_token_tree (dfa, NULL, NULL, &node->token); + if (*p_new == NULL) + return NULL; + (*p_new)->parent = dup_node; + (*p_new)->token.duplicated = 1; + dup_node = *p_new; + + /* Go to the left node, or up and to the right. */ + if (node->left) + { + node = node->left; + p_new = &dup_node->left; + } + else + { + const bin_tree_t *prev = NULL; + while (node->right == prev || node->right == NULL) + { + prev = node; + node = node->parent; + dup_node = dup_node->parent; + if (!node) + return dup_root; + } + node = node->right; + p_new = &dup_node->right; + } + } +} diff --git a/lib/regex.c b/lib/regex.c new file mode 100644 index 0000000..bf95254 --- /dev/null +++ b/lib/regex.c @@ -0,0 +1,77 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002-2003, 2005-2006, 2009-2012 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#ifndef _LIBC +# include <config.h> + +# if (__GNUC__ == 4 && 3 <= __GNUC_MINOR__) || 4 < __GNUC__ +# pragma GCC diagnostic ignored "-Wsuggest-attribute=pure" +# 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 diff --git a/lib/regex.h b/lib/regex.h new file mode 100644 index 0000000..2320c4b --- /dev/null +++ b/lib/regex.h @@ -0,0 +1,667 @@ +/* Definitions for data structures and routines for the regular + expression library. + Copyright (C) 1985, 1989-1993, 1995-1998, 2000-2003, 2005-2012 + Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://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 nonnegative object indexes. Traditionally, GNU regex + uses 'int' for these. Code that uses __re_idx_t should work + regardless of whether the type is signed. */ +typedef size_t __re_idx_t; + +/* 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 int __re_idx_t; +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_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. 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. 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. */ + __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. */ +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 initialised 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, __re_idx_t __length, + __re_idx_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, __re_idx_t __length1, + const char *__string2, __re_idx_t __length2, + __re_idx_t __start, regoff_t __range, + struct re_registers *__regs, + __re_idx_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, __re_idx_t __length, + __re_idx_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, __re_idx_t __length1, + const char *__string2, __re_idx_t __length2, + __re_idx_t __start, struct re_registers *__regs, + __re_idx_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_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". + Other compilers use __restrict, __restrict__, and _Restrict, and + 'configure' might #define 'restrict' to those words, so pick a + different name. */ +#ifndef _Restrict_ +# if 199901L <= __STDC_VERSION__ +# define _Restrict_ restrict +# elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__) +# define _Restrict_ __restrict +# else +# define _Restrict_ +# endif +#endif +/* gcc 3.1 and up support the [restrict] syntax. Don't trust + sys/cdefs.h's definition of __restrict_arr, though, as it + mishandles gcc -ansi -pedantic. */ +#ifndef _Restrict_arr_ +# if ((199901L <= __STDC_VERSION__ \ + || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__)) \ + && !defined __STRICT_ANSI__)) \ + && !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 */ diff --git a/lib/regex_internal.c b/lib/regex_internal.c new file mode 100644 index 0000000..5ecabc0 --- /dev/null +++ b/lib/regex_internal.c @@ -0,0 +1,1742 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002-2012 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +static void re_string_construct_common (const char *str, Idx len, + re_string_t *pstr, + RE_TRANSLATE_TYPE trans, bool 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, + re_hashval_t hash) internal_function; +static re_dfastate_t *create_cd_newstate (const re_dfa_t *dfa, + const re_node_set *nodes, + unsigned int context, + re_hashval_t hash) internal_function; + +/* 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 __attribute_warn_unused_result__ +re_string_allocate (re_string_t *pstr, const char *str, Idx len, Idx init_len, + RE_TRANSLATE_TYPE trans, bool icase, const re_dfa_t *dfa) +{ + reg_errcode_t ret; + Idx 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 __attribute_warn_unused_result__ +re_string_construct (re_string_t *pstr, const char *str, Idx len, + RE_TRANSLATE_TYPE trans, bool 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 __attribute_warn_unused_result__ +re_string_realloc_buffers (re_string_t *pstr, Idx 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 (Idx)); + if (BE (MIN (IDX_MAX, 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) + { + Idx *new_offsets = re_realloc (pstr->offsets, Idx, 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, Idx len, re_string_t *pstr, + RE_TRANSLATE_TYPE trans, bool 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; + 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; + Idx 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) -1 || mbclen == 0 + || (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len), 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; + } + else if (BE (mbclen == (size_t) -2, 0)) + { + /* The buffer doesn't have enough space, finish to build. */ + pstr->cur_state = prev_st; + break; + } + + /* 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 __attribute_warn_unused_result__ +build_wcs_upper_buffer (re_string_t *pstr) +{ + mbstate_t prev_st; + Idx 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 < (size_t) -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 + || (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len)) + { + /* It is an invalid character, an incomplete character + at the end of the string, 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 < (size_t) -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 (Idx, 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 + || (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len)) + { + /* 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 Idx +internal_function +re_string_skip_chars (re_string_t *pstr, Idx new_raw_idx, wint_t *last_wc) +{ + mbstate_t prev_st; + Idx 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; + Idx remain_len = pstr->raw_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 = wc2; + /* Then proceed the next character. */ + rawbuf_idx += mbclen; + } + *last_wc = 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) +{ + Idx 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) +{ + Idx 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 __attribute_warn_unused_result__ +re_string_reconstruct (re_string_t *pstr, Idx idx, int eflags) +{ + Idx offset; + + if (BE (pstr->raw_mbs_idx <= idx, 0)) + offset = idx - pstr->raw_mbs_idx; + else + { + /* 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)) + { + Idx 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. */ + Idx 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) + { + Idx 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; + Idx mlen = raw + pstr->len - p; + unsigned char buf[6]; + size_t mbclen; + + const unsigned char *pp = p; + if (BE (pstr->trans != NULL, 0)) + { + int i = mlen < 6 ? mlen : 6; + while (--i >= 0) + buf[i] = pstr->trans[p[i]]; + pp = buf; + } + /* 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 *) pp, 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, Idx idx) +{ + int ch; + Idx 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 +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) + { + Idx off; + int 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, Idx idx, int eflags) +{ + int c; + if (BE (! REG_VALID_INDEX (idx), 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; + Idx wc_idx = idx; + while(input->wcs[wc_idx] == WEOF) + { +#ifdef DEBUG + /* It must not happen. */ + assert (REG_VALID_INDEX (wc_idx)); +#endif + --wc_idx; + if (! REG_VALID_INDEX (wc_idx)) + 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 __attribute_warn_unused_result__ +re_node_set_alloc (re_node_set *set, Idx size) +{ + set->alloc = size; + set->nelem = 0; + set->elems = re_malloc (Idx, size); + if (BE (set->elems == NULL, 0)) + return REG_ESPACE; + return REG_NOERROR; +} + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +re_node_set_init_1 (re_node_set *set, Idx elem) +{ + set->alloc = 1; + set->nelem = 1; + set->elems = re_malloc (Idx, 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 __attribute_warn_unused_result__ +re_node_set_init_2 (re_node_set *set, Idx elem1, Idx elem2) +{ + set->alloc = 2; + set->elems = re_malloc (Idx, 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 __attribute_warn_unused_result__ +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 (Idx, dest->alloc); + if (BE (dest->elems == NULL, 0)) + { + dest->alloc = dest->nelem = 0; + return REG_ESPACE; + } + memcpy (dest->elems, src->elems, src->nelem * sizeof (Idx)); + } + 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 __attribute_warn_unused_result__ +re_node_set_add_intersect (re_node_set *dest, const re_node_set *src1, + const re_node_set *src2) +{ + Idx 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) + { + Idx new_alloc = src1->nelem + src2->nelem + dest->alloc; + Idx *new_elems = re_realloc (dest->elems, Idx, 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 (REG_VALID_INDEX (id) && dest->elems[id] > src1->elems[i1]) + --id; + + if (! REG_VALID_INDEX (id) || dest->elems[id] != src1->elems[i1]) + dest->elems[--sbase] = src1->elems[i1]; + + if (! REG_VALID_INDEX (--i1) || ! REG_VALID_INDEX (--i2)) + break; + } + + /* Lower the highest of the two items. */ + else if (src1->elems[i1] < src2->elems[i2]) + { + if (! REG_VALID_INDEX (--i2)) + break; + } + else + { + if (! REG_VALID_INDEX (--i1)) + 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 && REG_VALID_INDEX (id)) + 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 (! REG_VALID_INDEX (--id)) + break; + } + } + + /* Copy remaining SRC elements. */ + memcpy (dest->elems, dest->elems + sbase, delta * sizeof (Idx)); + + 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 __attribute_warn_unused_result__ +re_node_set_init_union (re_node_set *dest, const re_node_set *src1, + const re_node_set *src2) +{ + Idx i1, i2, id; + if (src1 != NULL && src1->nelem > 0 && src2 != NULL && src2->nelem > 0) + { + dest->alloc = src1->nelem + src2->nelem; + dest->elems = re_malloc (Idx, 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 (Idx)); + id += src1->nelem - i1; + } + else if (i2 < src2->nelem) + { + memcpy (dest->elems + id, src2->elems + i2, + (src2->nelem - i2) * sizeof (Idx)); + 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 __attribute_warn_unused_result__ +re_node_set_merge (re_node_set *dest, const re_node_set *src) +{ + Idx is, id, sbase, delta; + if (src == NULL || src->nelem == 0) + return REG_NOERROR; + if (dest->alloc < 2 * src->nelem + dest->nelem) + { + Idx new_alloc = 2 * (src->nelem + dest->alloc); + Idx *new_buffer = re_realloc (dest->elems, Idx, 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 (Idx)); + 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; + REG_VALID_INDEX (is) && REG_VALID_INDEX (id); ) + { + 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 (REG_VALID_INDEX (is)) + { + /* If DEST is exhausted, the remaining items of SRC must be unique. */ + sbase -= is + 1; + memcpy (dest->elems + sbase, src->elems, (is + 1) * sizeof (Idx)); + } + + 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 (! REG_VALID_INDEX (--id)) + { + /* Copy remaining SRC elements. */ + memcpy (dest->elems, dest->elems + sbase, + delta * sizeof (Idx)); + break; + } + } + } + + return REG_NOERROR; +} + +/* Insert the new element ELEM to the re_node_set* SET. + SET should not already have ELEM. + Return true if successful. */ + +static bool +internal_function __attribute_warn_unused_result__ +re_node_set_insert (re_node_set *set, Idx elem) +{ + Idx idx; + /* In case the set is empty. */ + if (set->alloc == 0) + return BE (re_node_set_init_1 (set, elem) == REG_NOERROR, 1); + + if (BE (set->nelem, 0) == 0) + { + /* We already guaranteed above that set->alloc != 0. */ + set->elems[0] = elem; + ++set->nelem; + return true; + } + + /* Realloc if we need. */ + if (set->alloc == set->nelem) + { + Idx *new_elems; + set->alloc = set->alloc * 2; + new_elems = re_realloc (set->elems, Idx, set->alloc); + if (BE (new_elems == NULL, 0)) + return false; + 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 true; +} + +/* 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 true if successful. */ + +static bool +internal_function __attribute_warn_unused_result__ +re_node_set_insert_last (re_node_set *set, Idx elem) +{ + /* Realloc if we need. */ + if (set->alloc == set->nelem) + { + Idx *new_elems; + set->alloc = (set->alloc + 1) * 2; + new_elems = re_realloc (set->elems, Idx, set->alloc); + if (BE (new_elems == NULL, 0)) + return false; + set->elems = new_elems; + } + + /* Insert the new element. */ + set->elems[set->nelem++] = elem; + return true; +} + +/* Compare two node sets SET1 and SET2. + Return true if SET1 and SET2 are equivalent. */ + +static bool +internal_function __attribute ((pure)) +re_node_set_compare (const re_node_set *set1, const re_node_set *set2) +{ + Idx i; + if (set1 == NULL || set2 == NULL || set1->nelem != set2->nelem) + return false; + for (i = set1->nelem ; REG_VALID_INDEX (--i) ; ) + if (set1->elems[i] != set2->elems[i]) + return false; + return true; +} + +/* Return (idx + 1) if SET contains the element ELEM, return 0 otherwise. */ + +static Idx +internal_function __attribute ((pure)) +re_node_set_contains (const re_node_set *set, Idx elem) +{ + __re_size_t idx, right, mid; + if (! REG_VALID_NONZERO_INDEX (set->nelem)) + 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, Idx 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 REG_MISSING if an error occurred. */ + +static Idx +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; + Idx *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 (Idx))); + if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) < new_nodes_alloc, 0)) + return REG_MISSING; + + new_nodes = re_realloc (dfa->nodes, re_token_t, new_nodes_alloc); + if (BE (new_nodes == NULL, 0)) + return REG_MISSING; + dfa->nodes = new_nodes; + new_nexts = re_realloc (dfa->nexts, Idx, new_nodes_alloc); + new_indices = re_realloc (dfa->org_indices, Idx, 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 REG_MISSING; + 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 + { + int type = token.type; + dfa->nodes[dfa->nodes_len].accept_mb = + (type == OP_PERIOD && dfa->mb_cur_max > 1) || type == COMPLEX_BRACKET; + } +#endif + dfa->nexts[dfa->nodes_len] = REG_MISSING; + 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 re_hashval_t +internal_function +calc_state_hash (const re_node_set *nodes, unsigned int context) +{ + re_hashval_t hash = nodes->nelem + context; + Idx 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 __attribute_warn_unused_result__ +re_acquire_state (reg_errcode_t *err, const re_dfa_t *dfa, + const re_node_set *nodes) +{ + re_hashval_t hash; + re_dfastate_t *new_state; + struct re_state_table_entry *spot; + Idx i; +#ifdef lint + /* Suppress bogus uninitialized-variable warnings. */ + *err = REG_NOERROR; +#endif + 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 __attribute_warn_unused_result__ +re_acquire_state_context (reg_errcode_t *err, const re_dfa_t *dfa, + const re_node_set *nodes, unsigned int context) +{ + re_hashval_t hash; + re_dfastate_t *new_state; + struct re_state_table_entry *spot; + Idx i; +#ifdef lint + /* Suppress bogus uninitialized-variable warnings. */ + *err = REG_NOERROR; +#endif + 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 +__attribute_warn_unused_result__ +register_state (const re_dfa_t *dfa, re_dfastate_t *newstate, + re_hashval_t hash) +{ + struct re_state_table_entry *spot; + reg_errcode_t err; + Idx 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++) + { + Idx elem = newstate->nodes.elems[i]; + if (!IS_EPSILON_NODE (dfa->nodes[elem].type)) + if (! re_node_set_insert_last (&newstate->non_eps_nodes, elem)) + return REG_ESPACE; + } + + spot = dfa->state_table + (hash & dfa->state_hash_mask); + if (BE (spot->alloc <= spot->num, 0)) + { + Idx 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 independent of contexts. + Return the new state if succeeded, otherwise return NULL. */ + +static re_dfastate_t * +internal_function __attribute_warn_unused_result__ +create_ci_newstate (const re_dfa_t *dfa, const re_node_set *nodes, + re_hashval_t hash) +{ + Idx 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 __attribute_warn_unused_result__ +create_cd_newstate (const re_dfa_t *dfa, const re_node_set *nodes, + unsigned int context, re_hashval_t hash) +{ + Idx 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; +} diff --git a/lib/regex_internal.h b/lib/regex_internal.h new file mode 100644 index 0000000..196605d --- /dev/null +++ b/lib/regex_internal.h @@ -0,0 +1,869 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002-2012 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://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> +#ifndef _LIBC +# include "localcharset.h" +#endif +#include <locale.h> +#include <wchar.h> +#include <wctype.h> +#include <stdbool.h> +#include <stdint.h> +#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 + +/* 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/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 + +#if (defined MB_CUR_MAX && HAVE_WCTYPE_H && HAVE_ISWCTYPE && HAVE_WCSCOLL) || _LIBC +# define RE_ENABLE_I18N +#endif + +#if __GNUC__ >= 3 +# define BE(expr, val) __builtin_expect (expr, val) +#else +# define BE(expr, val) (expr) +# ifdef _LIBC +# define inline +# endif +#endif + +/* 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 +# define __wctype wctype +# define __iswctype iswctype +# define __btowc btowc +# define __mbrtowc mbrtowc +# define __wcrtomb wcrtomb +# define __regfree regfree +# define attribute_hidden +#endif /* not _LIBC */ + +#if __GNUC__ >= 4 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1) +# define __attribute(arg) __attribute__ (arg) +#else +# define __attribute(arg) +#endif + +typedef __re_idx_t Idx; +#ifdef _REGEX_LARGE_OFFSETS +# define IDX_MAX (SIZE_MAX - 2) +#else +# define IDX_MAX INT_MAX +#endif + +/* Special return value for failure to match. */ +#define REG_MISSING ((Idx) -1) + +/* Special return value for internal error. */ +#define REG_ERROR ((Idx) -2) + +/* Test whether N is a valid index, and is not one of the above. */ +#ifdef _REGEX_LARGE_OFFSETS +# define REG_VALID_INDEX(n) ((Idx) (n) < REG_ERROR) +#else +# define REG_VALID_INDEX(n) (0 <= (n)) +#endif + +/* Test whether N is a valid nonzero index. */ +#ifdef _REGEX_LARGE_OFFSETS +# define REG_VALID_NONZERO_INDEX(n) ((Idx) ((n) - 1) < (Idx) (REG_ERROR - 1)) +#else +# define REG_VALID_NONZERO_INDEX(n) (0 < (n)) +#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 internal_function +#endif + +#ifndef NOT_IN_libc +static reg_errcode_t re_string_realloc_buffers (re_string_t *pstr, + Idx 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, Idx 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)) + +#include <alloca.h> + +#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 + +#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 + __libc_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; + + +/* Inline 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 +/* Inline functions for re_string. */ +static inline int +internal_function __attribute ((pure)) +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 inline wint_t +internal_function __attribute ((pure)) +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]; +} + +# ifndef NOT_IN_libc +static int +internal_function __attribute ((pure)) +re_string_elem_size_at (const re_string_t *pstr, Idx idx) +{ +# ifdef _LIBC + const unsigned char *p, *extra; + const int32_t *table, *indirect; +# 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; + findidx (&p, pstr->len - idx); + return p - pstr->mbs - idx; + } + else +# endif /* _LIBC */ + return 1; +} +# endif +#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 + +#endif /* _REGEX_INTERNAL_H */ diff --git a/lib/regexec.c b/lib/regexec.c new file mode 100644 index 0000000..1d7719c --- /dev/null +++ b/lib/regexec.c @@ -0,0 +1,4419 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002-2012 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +static reg_errcode_t match_ctx_init (re_match_context_t *cache, int eflags, + Idx n) internal_function; +static void match_ctx_clean (re_match_context_t *mctx) internal_function; +static void match_ctx_free (re_match_context_t *cache) internal_function; +static reg_errcode_t match_ctx_add_entry (re_match_context_t *cache, Idx node, + Idx str_idx, Idx from, Idx to) + internal_function; +static Idx search_cur_bkref_entry (const re_match_context_t *mctx, Idx str_idx) + internal_function; +static reg_errcode_t match_ctx_add_subtop (re_match_context_t *mctx, Idx node, + Idx str_idx) internal_function; +static re_sub_match_last_t * match_ctx_add_sublast (re_sub_match_top_t *subtop, + Idx node, Idx str_idx) + internal_function; +static void sift_ctx_init (re_sift_context_t *sctx, re_dfastate_t **sifted_sts, + re_dfastate_t **limited_sts, Idx last_node, + Idx last_str_idx) + internal_function; +static reg_errcode_t re_search_internal (const regex_t *preg, + const char *string, Idx length, + Idx start, Idx last_start, Idx stop, + size_t nmatch, regmatch_t pmatch[], + int eflags) internal_function; +static regoff_t re_search_2_stub (struct re_pattern_buffer *bufp, + const char *string1, Idx length1, + const char *string2, Idx length2, + Idx start, regoff_t range, + struct re_registers *regs, + Idx stop, bool ret_len) internal_function; +static regoff_t re_search_stub (struct re_pattern_buffer *bufp, + const char *string, Idx length, Idx start, + regoff_t range, Idx stop, + struct re_registers *regs, + bool ret_len) internal_function; +static unsigned re_copy_regs (struct re_registers *regs, regmatch_t *pmatch, + Idx nregs, int regs_allocated) internal_function; +static reg_errcode_t prune_impossible_nodes (re_match_context_t *mctx) + internal_function; +static Idx check_matching (re_match_context_t *mctx, bool fl_longest_match, + Idx *p_match_first) internal_function; +static Idx check_halt_state_context (const re_match_context_t *mctx, + const re_dfastate_t *state, Idx idx) + internal_function; +static void update_regs (const re_dfa_t *dfa, regmatch_t *pmatch, + regmatch_t *prev_idx_match, Idx cur_node, + Idx cur_idx, Idx nmatch) internal_function; +static reg_errcode_t push_fail_stack (struct re_fail_stack_t *fs, + Idx str_idx, Idx dest_node, Idx nregs, + regmatch_t *regs, + re_node_set *eps_via_nodes) + internal_function; +static reg_errcode_t set_regs (const regex_t *preg, + const re_match_context_t *mctx, + size_t nmatch, regmatch_t *pmatch, + bool fl_backtrack) internal_function; +static reg_errcode_t free_fail_stack_return (struct re_fail_stack_t *fs) + internal_function; + +#ifdef RE_ENABLE_I18N +static int sift_states_iter_mb (const re_match_context_t *mctx, + re_sift_context_t *sctx, + Idx node_idx, Idx str_idx, Idx max_str_idx) + internal_function; +#endif /* RE_ENABLE_I18N */ +static reg_errcode_t sift_states_backward (const re_match_context_t *mctx, + re_sift_context_t *sctx) + internal_function; +static reg_errcode_t build_sifted_states (const re_match_context_t *mctx, + re_sift_context_t *sctx, Idx str_idx, + re_node_set *cur_dest) + internal_function; +static reg_errcode_t update_cur_sifted_state (const re_match_context_t *mctx, + re_sift_context_t *sctx, + Idx str_idx, + re_node_set *dest_nodes) + internal_function; +static reg_errcode_t add_epsilon_src_nodes (const re_dfa_t *dfa, + re_node_set *dest_nodes, + const re_node_set *candidates) + internal_function; +static bool check_dst_limits (const re_match_context_t *mctx, + const re_node_set *limits, + Idx dst_node, Idx dst_idx, Idx src_node, + Idx src_idx) internal_function; +static int check_dst_limits_calc_pos_1 (const re_match_context_t *mctx, + int boundaries, Idx subexp_idx, + Idx from_node, Idx bkref_idx) + internal_function; +static int check_dst_limits_calc_pos (const re_match_context_t *mctx, + Idx limit, Idx subexp_idx, + Idx node, Idx str_idx, + Idx bkref_idx) internal_function; +static reg_errcode_t check_subexp_limits (const re_dfa_t *dfa, + re_node_set *dest_nodes, + const re_node_set *candidates, + re_node_set *limits, + struct re_backref_cache_entry *bkref_ents, + Idx str_idx) internal_function; +static reg_errcode_t sift_states_bkref (const re_match_context_t *mctx, + re_sift_context_t *sctx, + Idx str_idx, const re_node_set *candidates) + internal_function; +static reg_errcode_t merge_state_array (const re_dfa_t *dfa, + re_dfastate_t **dst, + re_dfastate_t **src, Idx num) + internal_function; +static re_dfastate_t *find_recover_state (reg_errcode_t *err, + re_match_context_t *mctx) internal_function; +static re_dfastate_t *transit_state (reg_errcode_t *err, + re_match_context_t *mctx, + re_dfastate_t *state) internal_function; +static re_dfastate_t *merge_state_with_log (reg_errcode_t *err, + re_match_context_t *mctx, + re_dfastate_t *next_state) + internal_function; +static reg_errcode_t check_subexp_matching_top (re_match_context_t *mctx, + re_node_set *cur_nodes, + Idx str_idx) internal_function; +#if 0 +static re_dfastate_t *transit_state_sb (reg_errcode_t *err, + re_match_context_t *mctx, + re_dfastate_t *pstate) + internal_function; +#endif +#ifdef RE_ENABLE_I18N +static reg_errcode_t transit_state_mb (re_match_context_t *mctx, + re_dfastate_t *pstate) + internal_function; +#endif /* RE_ENABLE_I18N */ +static reg_errcode_t transit_state_bkref (re_match_context_t *mctx, + const re_node_set *nodes) + internal_function; +static reg_errcode_t get_subexp (re_match_context_t *mctx, + Idx bkref_node, Idx bkref_str_idx) + internal_function; +static reg_errcode_t get_subexp_sub (re_match_context_t *mctx, + const re_sub_match_top_t *sub_top, + re_sub_match_last_t *sub_last, + Idx bkref_node, Idx bkref_str) + internal_function; +static Idx find_subexp_node (const re_dfa_t *dfa, const re_node_set *nodes, + Idx subexp_idx, int type) internal_function; +static reg_errcode_t check_arrival (re_match_context_t *mctx, + state_array_t *path, Idx top_node, + Idx top_str, Idx last_node, Idx last_str, + int type) internal_function; +static reg_errcode_t check_arrival_add_next_nodes (re_match_context_t *mctx, + Idx str_idx, + re_node_set *cur_nodes, + re_node_set *next_nodes) + internal_function; +static reg_errcode_t check_arrival_expand_ecl (const re_dfa_t *dfa, + re_node_set *cur_nodes, + Idx ex_subexp, int type) + internal_function; +static reg_errcode_t check_arrival_expand_ecl_sub (const re_dfa_t *dfa, + re_node_set *dst_nodes, + Idx target, Idx ex_subexp, + int type) internal_function; +static reg_errcode_t expand_bkref_cache (re_match_context_t *mctx, + re_node_set *cur_nodes, Idx cur_str, + Idx subexp_num, int type) + internal_function; +static bool build_trtable (const re_dfa_t *dfa, + re_dfastate_t *state) internal_function; +#ifdef RE_ENABLE_I18N +static int check_node_accept_bytes (const re_dfa_t *dfa, Idx node_idx, + const re_string_t *input, Idx idx) + internal_function; +# ifdef _LIBC +static unsigned int find_collation_sequence_value (const unsigned char *mbs, + size_t name_len) + internal_function; +# endif /* _LIBC */ +#endif /* RE_ENABLE_I18N */ +static Idx group_nodes_into_DFAstates (const re_dfa_t *dfa, + const re_dfastate_t *state, + re_node_set *states_node, + bitset_t *states_ch) internal_function; +static bool check_node_accept (const re_match_context_t *mctx, + const re_token_t *node, Idx idx) + internal_function; +static reg_errcode_t extend_buffers (re_match_context_t *mctx) + internal_function; + +/* Entry point for POSIX code. */ + +/* regexec searches for a given pattern, specified by PREG, in the + string STRING. + + If NMATCH is zero or REG_NOSUB was set in the cflags argument to + 'regcomp', we ignore PMATCH. Otherwise, we assume PMATCH has at + least NMATCH elements, and we set them to the offsets of the + corresponding matched substrings. + + EFLAGS specifies "execution flags" which affect matching: if + REG_NOTBOL is set, then ^ does not match at the beginning of the + string; if REG_NOTEOL is set, then $ does not match at the end. + + We return 0 if we find a match and REG_NOMATCH if not. */ + +int +regexec (preg, string, nmatch, pmatch, eflags) + const regex_t *_Restrict_ preg; + const char *_Restrict_ string; + size_t nmatch; + regmatch_t pmatch[_Restrict_arr_]; + int eflags; +{ + reg_errcode_t err; + Idx start, length; +#ifdef _LIBC + re_dfa_t *dfa = (re_dfa_t *) preg->buffer; +#endif + + if (eflags & ~(REG_NOTBOL | REG_NOTEOL | REG_STARTEND)) + return REG_BADPAT; + + if (eflags & REG_STARTEND) + { + start = pmatch[0].rm_so; + length = pmatch[0].rm_eo; + } + else + { + start = 0; + length = strlen (string); + } + + __libc_lock_lock (dfa->lock); + if (preg->no_sub) + err = re_search_internal (preg, string, length, start, length, + length, 0, NULL, eflags); + else + err = re_search_internal (preg, string, length, start, length, + length, nmatch, pmatch, eflags); + __libc_lock_unlock (dfa->lock); + return err != REG_NOERROR; +} + +#ifdef _LIBC +# include <shlib-compat.h> +versioned_symbol (libc, __regexec, regexec, GLIBC_2_3_4); + +# if SHLIB_COMPAT (libc, GLIBC_2_0, GLIBC_2_3_4) +__typeof__ (__regexec) __compat_regexec; + +int +attribute_compat_text_section +__compat_regexec (const regex_t *_Restrict_ preg, + const char *_Restrict_ string, size_t nmatch, + regmatch_t pmatch[], int eflags) +{ + return regexec (preg, string, nmatch, pmatch, + eflags & (REG_NOTBOL | REG_NOTEOL)); +} +compat_symbol (libc, __compat_regexec, regexec, GLIBC_2_0); +# endif +#endif + +/* Entry points for GNU code. */ + +/* re_match, re_search, re_match_2, re_search_2 + + The former two functions operate on STRING with length LENGTH, + while the later two operate on concatenation of STRING1 and STRING2 + with lengths LENGTH1 and LENGTH2, respectively. + + re_match() matches the compiled pattern in BUFP against the string, + starting at index START. + + re_search() first tries matching at index START, then it tries to match + starting from index START + 1, and so on. The last start position tried + is START + RANGE. (Thus RANGE = 0 forces re_search to operate the same + way as re_match().) + + The parameter STOP of re_{match,search}_2 specifies that no match exceeding + the first STOP characters of the concatenation of the strings should be + concerned. + + If REGS is not NULL, and BUFP->no_sub is not set, the offsets of the match + and all groups is stored in REGS. (For the "_2" variants, the offsets are + computed relative to the concatenation, not relative to the individual + strings.) + + On success, re_match* functions return the length of the match, re_search* + return the position of the start of the match. Return value -1 means no + match was found and -2 indicates an internal error. */ + +regoff_t +re_match (bufp, string, length, start, regs) + struct re_pattern_buffer *bufp; + const char *string; + Idx length, start; + struct re_registers *regs; +{ + return re_search_stub (bufp, string, length, start, 0, length, regs, true); +} +#ifdef _LIBC +weak_alias (__re_match, re_match) +#endif + +regoff_t +re_search (bufp, string, length, start, range, regs) + struct re_pattern_buffer *bufp; + const char *string; + Idx length, start; + regoff_t range; + struct re_registers *regs; +{ + return re_search_stub (bufp, string, length, start, range, length, regs, + false); +} +#ifdef _LIBC +weak_alias (__re_search, re_search) +#endif + +regoff_t +re_match_2 (bufp, string1, length1, string2, length2, start, regs, stop) + struct re_pattern_buffer *bufp; + const char *string1, *string2; + Idx length1, length2, start, stop; + struct re_registers *regs; +{ + return re_search_2_stub (bufp, string1, length1, string2, length2, + start, 0, regs, stop, true); +} +#ifdef _LIBC +weak_alias (__re_match_2, re_match_2) +#endif + +regoff_t +re_search_2 (bufp, string1, length1, string2, length2, start, range, regs, stop) + struct re_pattern_buffer *bufp; + const char *string1, *string2; + Idx length1, length2, start, stop; + regoff_t range; + struct re_registers *regs; +{ + return re_search_2_stub (bufp, string1, length1, string2, length2, + start, range, regs, stop, false); +} +#ifdef _LIBC +weak_alias (__re_search_2, re_search_2) +#endif + +static regoff_t +re_search_2_stub (struct re_pattern_buffer *bufp, + const char *string1, Idx length1, + const char *string2, Idx length2, + Idx start, regoff_t range, struct re_registers *regs, + Idx stop, bool ret_len) +{ + const char *str; + regoff_t rval; + Idx len = length1 + length2; + char *s = NULL; + + if (BE (length1 < 0 || length2 < 0 || stop < 0 || len < length1, 0)) + return -2; + + /* Concatenate the strings. */ + if (length2 > 0) + if (length1 > 0) + { + s = re_malloc (char, len); + + if (BE (s == NULL, 0)) + return -2; +#ifdef _LIBC + memcpy (__mempcpy (s, string1, length1), string2, length2); +#else + memcpy (s, string1, length1); + memcpy (s + length1, string2, length2); +#endif + str = s; + } + else + str = string2; + else + str = string1; + + rval = re_search_stub (bufp, str, len, start, range, stop, regs, + ret_len); + re_free (s); + return rval; +} + +/* The parameters have the same meaning as those of re_search. + Additional parameters: + If RET_LEN is true the length of the match is returned (re_match style); + otherwise the position of the match is returned. */ + +static regoff_t +re_search_stub (struct re_pattern_buffer *bufp, + const char *string, Idx length, + Idx start, regoff_t range, Idx stop, struct re_registers *regs, + bool ret_len) +{ + reg_errcode_t result; + regmatch_t *pmatch; + Idx nregs; + regoff_t rval; + int eflags = 0; +#ifdef _LIBC + re_dfa_t *dfa = (re_dfa_t *) bufp->buffer; +#endif + Idx last_start = start + range; + + /* Check for out-of-range. */ + if (BE (start < 0 || start > length, 0)) + return -1; + if (BE (length < last_start || (0 <= range && last_start < start), 0)) + last_start = length; + else if (BE (last_start < 0 || (range < 0 && start <= last_start), 0)) + last_start = 0; + + __libc_lock_lock (dfa->lock); + + eflags |= (bufp->not_bol) ? REG_NOTBOL : 0; + eflags |= (bufp->not_eol) ? REG_NOTEOL : 0; + + /* Compile fastmap if we haven't yet. */ + if (start < last_start && bufp->fastmap != NULL && !bufp->fastmap_accurate) + re_compile_fastmap (bufp); + + if (BE (bufp->no_sub, 0)) + regs = NULL; + + /* We need at least 1 register. */ + if (regs == NULL) + nregs = 1; + else if (BE (bufp->regs_allocated == REGS_FIXED + && regs->num_regs <= bufp->re_nsub, 0)) + { + nregs = regs->num_regs; + if (BE (nregs < 1, 0)) + { + /* Nothing can be copied to regs. */ + regs = NULL; + nregs = 1; + } + } + else + nregs = bufp->re_nsub + 1; + pmatch = re_malloc (regmatch_t, nregs); + if (BE (pmatch == NULL, 0)) + { + rval = -2; + goto out; + } + + result = re_search_internal (bufp, string, length, start, last_start, stop, + nregs, pmatch, eflags); + + rval = 0; + + /* I hope we needn't fill ther regs with -1's when no match was found. */ + if (result != REG_NOERROR) + rval = result == REG_NOMATCH ? -1 : -2; + else if (regs != NULL) + { + /* If caller wants register contents data back, copy them. */ + bufp->regs_allocated = re_copy_regs (regs, pmatch, nregs, + bufp->regs_allocated); + if (BE (bufp->regs_allocated == REGS_UNALLOCATED, 0)) + rval = -2; + } + + if (BE (rval == 0, 1)) + { + if (ret_len) + { + assert (pmatch[0].rm_so == start); + rval = pmatch[0].rm_eo - start; + } + else + rval = pmatch[0].rm_so; + } + re_free (pmatch); + out: + __libc_lock_unlock (dfa->lock); + return rval; +} + +static unsigned +re_copy_regs (struct re_registers *regs, regmatch_t *pmatch, Idx nregs, + int regs_allocated) +{ + int rval = REGS_REALLOCATE; + Idx i; + Idx need_regs = nregs + 1; + /* We need one extra element beyond 'num_regs' for the '-1' marker GNU code + uses. */ + + /* Have the register data arrays been allocated? */ + if (regs_allocated == REGS_UNALLOCATED) + { /* No. So allocate them with malloc. */ + regs->start = re_malloc (regoff_t, need_regs); + if (BE (regs->start == NULL, 0)) + return REGS_UNALLOCATED; + regs->end = re_malloc (regoff_t, need_regs); + if (BE (regs->end == NULL, 0)) + { + re_free (regs->start); + return REGS_UNALLOCATED; + } + regs->num_regs = need_regs; + } + else if (regs_allocated == REGS_REALLOCATE) + { /* Yes. If we need more elements than were already + allocated, reallocate them. If we need fewer, just + leave it alone. */ + if (BE (need_regs > regs->num_regs, 0)) + { + regoff_t *new_start = re_realloc (regs->start, regoff_t, need_regs); + regoff_t *new_end; + if (BE (new_start == NULL, 0)) + return REGS_UNALLOCATED; + new_end = re_realloc (regs->end, regoff_t, need_regs); + if (BE (new_end == NULL, 0)) + { + re_free (new_start); + return REGS_UNALLOCATED; + } + regs->start = new_start; + regs->end = new_end; + regs->num_regs = need_regs; + } + } + else + { + assert (regs_allocated == REGS_FIXED); + /* This function may not be called with REGS_FIXED and nregs too big. */ + assert (regs->num_regs >= nregs); + rval = REGS_FIXED; + } + + /* Copy the regs. */ + for (i = 0; i < nregs; ++i) + { + regs->start[i] = pmatch[i].rm_so; + regs->end[i] = pmatch[i].rm_eo; + } + for ( ; i < regs->num_regs; ++i) + regs->start[i] = regs->end[i] = -1; + + return rval; +} + +/* Set REGS to hold NUM_REGS registers, storing them in STARTS and + ENDS. Subsequent matches using PATTERN_BUFFER and REGS will use + this memory for recording register information. STARTS and ENDS + must be allocated using the malloc library routine, 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. */ + +void +re_set_registers (bufp, regs, num_regs, starts, ends) + struct re_pattern_buffer *bufp; + struct re_registers *regs; + __re_size_t num_regs; + regoff_t *starts, *ends; +{ + if (num_regs) + { + bufp->regs_allocated = REGS_REALLOCATE; + regs->num_regs = num_regs; + regs->start = starts; + regs->end = ends; + } + else + { + bufp->regs_allocated = REGS_UNALLOCATED; + regs->num_regs = 0; + regs->start = regs->end = NULL; + } +} +#ifdef _LIBC +weak_alias (__re_set_registers, re_set_registers) +#endif + +/* Entry points compatible with 4.2 BSD regex library. We don't define + them unless specifically requested. */ + +#if defined _REGEX_RE_COMP || defined _LIBC +int +# ifdef _LIBC +weak_function +# endif +re_exec (s) + const char *s; +{ + return 0 == regexec (&re_comp_buf, s, 0, NULL, 0); +} +#endif /* _REGEX_RE_COMP */ + +/* Internal entry point. */ + +/* Searches for a compiled pattern PREG in the string STRING, whose + length is LENGTH. NMATCH, PMATCH, and EFLAGS have the same + meaning as with regexec. LAST_START is START + RANGE, where + START and RANGE have the same meaning as with re_search. + Return REG_NOERROR if we find a match, and REG_NOMATCH if not, + otherwise return the error code. + Note: We assume front end functions already check ranges. + (0 <= LAST_START && LAST_START <= LENGTH) */ + +static reg_errcode_t +__attribute_warn_unused_result__ +re_search_internal (const regex_t *preg, + const char *string, Idx length, + Idx start, Idx last_start, Idx stop, + size_t nmatch, regmatch_t pmatch[], + int eflags) +{ + reg_errcode_t err; + const re_dfa_t *dfa = (const re_dfa_t *) preg->buffer; + Idx left_lim, right_lim; + int incr; + bool fl_longest_match; + int match_kind; + Idx match_first; + Idx match_last = REG_MISSING; + Idx extra_nmatch; + bool sb; + int ch; +#if defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L) + re_match_context_t mctx = { .dfa = dfa }; +#else + re_match_context_t mctx; +#endif + char *fastmap = ((preg->fastmap != NULL && preg->fastmap_accurate + && start != last_start && !preg->can_be_null) + ? preg->fastmap : NULL); + RE_TRANSLATE_TYPE t = preg->translate; + +#if !(defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L)) + memset (&mctx, '\0', sizeof (re_match_context_t)); + mctx.dfa = dfa; +#endif + + extra_nmatch = (nmatch > preg->re_nsub) ? nmatch - (preg->re_nsub + 1) : 0; + nmatch -= extra_nmatch; + + /* Check if the DFA haven't been compiled. */ + if (BE (preg->used == 0 || dfa->init_state == NULL + || dfa->init_state_word == NULL || dfa->init_state_nl == NULL + || dfa->init_state_begbuf == NULL, 0)) + return REG_NOMATCH; + +#ifdef DEBUG + /* We assume front-end functions already check them. */ + assert (0 <= last_start && last_start <= length); +#endif + + /* If initial states with non-begbuf contexts have no elements, + the regex must be anchored. If preg->newline_anchor is set, + we'll never use init_state_nl, so do not check it. */ + if (dfa->init_state->nodes.nelem == 0 + && dfa->init_state_word->nodes.nelem == 0 + && (dfa->init_state_nl->nodes.nelem == 0 + || !preg->newline_anchor)) + { + if (start != 0 && last_start != 0) + return REG_NOMATCH; + start = last_start = 0; + } + + /* We must check the longest matching, if nmatch > 0. */ + fl_longest_match = (nmatch != 0 || dfa->nbackref); + + err = re_string_allocate (&mctx.input, string, length, dfa->nodes_len + 1, + preg->translate, (preg->syntax & RE_ICASE) != 0, + dfa); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + mctx.input.stop = stop; + mctx.input.raw_stop = stop; + mctx.input.newline_anchor = preg->newline_anchor; + + err = match_ctx_init (&mctx, eflags, dfa->nbackref * 2); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + + /* We will log all the DFA states through which the dfa pass, + if nmatch > 1, or this dfa has "multibyte node", which is a + back-reference or a node which can accept multibyte character or + multi character collating element. */ + if (nmatch > 1 || dfa->has_mb_node) + { + /* Avoid overflow. */ + if (BE ((MIN (IDX_MAX, SIZE_MAX / sizeof (re_dfastate_t *)) + <= mctx.input.bufs_len), 0)) + { + err = REG_ESPACE; + goto free_return; + } + + mctx.state_log = re_malloc (re_dfastate_t *, mctx.input.bufs_len + 1); + if (BE (mctx.state_log == NULL, 0)) + { + err = REG_ESPACE; + goto free_return; + } + } + else + mctx.state_log = NULL; + + match_first = start; + mctx.input.tip_context = (eflags & REG_NOTBOL) ? CONTEXT_BEGBUF + : CONTEXT_NEWLINE | CONTEXT_BEGBUF; + + /* Check incrementally whether of not the input string match. */ + incr = (last_start < start) ? -1 : 1; + left_lim = (last_start < start) ? last_start : start; + right_lim = (last_start < start) ? start : last_start; + sb = dfa->mb_cur_max == 1; + match_kind = + (fastmap + ? ((sb || !(preg->syntax & RE_ICASE || t) ? 4 : 0) + | (start <= last_start ? 2 : 0) + | (t != NULL ? 1 : 0)) + : 8); + + for (;; match_first += incr) + { + err = REG_NOMATCH; + if (match_first < left_lim || right_lim < match_first) + goto free_return; + + /* Advance as rapidly as possible through the string, until we + find a plausible place to start matching. This may be done + with varying efficiency, so there are various possibilities: + only the most common of them are specialized, in order to + save on code size. We use a switch statement for speed. */ + switch (match_kind) + { + case 8: + /* No fastmap. */ + break; + + case 7: + /* Fastmap with single-byte translation, match forward. */ + while (BE (match_first < right_lim, 1) + && !fastmap[t[(unsigned char) string[match_first]]]) + ++match_first; + goto forward_match_found_start_or_reached_end; + + case 6: + /* Fastmap without translation, match forward. */ + while (BE (match_first < right_lim, 1) + && !fastmap[(unsigned char) string[match_first]]) + ++match_first; + + forward_match_found_start_or_reached_end: + if (BE (match_first == right_lim, 0)) + { + ch = match_first >= length + ? 0 : (unsigned char) string[match_first]; + if (!fastmap[t ? t[ch] : ch]) + goto free_return; + } + break; + + case 4: + case 5: + /* Fastmap without multi-byte translation, match backwards. */ + while (match_first >= left_lim) + { + ch = match_first >= length + ? 0 : (unsigned char) string[match_first]; + if (fastmap[t ? t[ch] : ch]) + break; + --match_first; + } + if (match_first < left_lim) + goto free_return; + break; + + default: + /* In this case, we can't determine easily the current byte, + since it might be a component byte of a multibyte + character. Then we use the constructed buffer instead. */ + for (;;) + { + /* If MATCH_FIRST is out of the valid range, reconstruct the + buffers. */ + __re_size_t offset = match_first - mctx.input.raw_mbs_idx; + if (BE (offset >= (__re_size_t) mctx.input.valid_raw_len, 0)) + { + err = re_string_reconstruct (&mctx.input, match_first, + eflags); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + + offset = match_first - mctx.input.raw_mbs_idx; + } + /* If MATCH_FIRST is out of the buffer, leave it as '\0'. + Note that MATCH_FIRST must not be smaller than 0. */ + ch = (match_first >= length + ? 0 : re_string_byte_at (&mctx.input, offset)); + if (fastmap[ch]) + break; + match_first += incr; + if (match_first < left_lim || match_first > right_lim) + { + err = REG_NOMATCH; + goto free_return; + } + } + break; + } + + /* Reconstruct the buffers so that the matcher can assume that + the matching starts from the beginning of the buffer. */ + err = re_string_reconstruct (&mctx.input, match_first, eflags); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + +#ifdef RE_ENABLE_I18N + /* Don't consider this char as a possible match start if it part, + yet isn't the head, of a multibyte character. */ + if (!sb && !re_string_first_byte (&mctx.input, 0)) + continue; +#endif + + /* It seems to be appropriate one, then use the matcher. */ + /* We assume that the matching starts from 0. */ + mctx.state_log_top = mctx.nbkref_ents = mctx.max_mb_elem_len = 0; + match_last = check_matching (&mctx, fl_longest_match, + start <= last_start ? &match_first : NULL); + if (match_last != REG_MISSING) + { + if (BE (match_last == REG_ERROR, 0)) + { + err = REG_ESPACE; + goto free_return; + } + else + { + mctx.match_last = match_last; + if ((!preg->no_sub && nmatch > 1) || dfa->nbackref) + { + re_dfastate_t *pstate = mctx.state_log[match_last]; + mctx.last_node = check_halt_state_context (&mctx, pstate, + match_last); + } + if ((!preg->no_sub && nmatch > 1 && dfa->has_plural_match) + || dfa->nbackref) + { + err = prune_impossible_nodes (&mctx); + if (err == REG_NOERROR) + break; + if (BE (err != REG_NOMATCH, 0)) + goto free_return; + match_last = REG_MISSING; + } + else + break; /* We found a match. */ + } + } + + match_ctx_clean (&mctx); + } + +#ifdef DEBUG + assert (match_last != REG_MISSING); + assert (err == REG_NOERROR); +#endif + + /* Set pmatch[] if we need. */ + if (nmatch > 0) + { + Idx reg_idx; + + /* Initialize registers. */ + for (reg_idx = 1; reg_idx < nmatch; ++reg_idx) + pmatch[reg_idx].rm_so = pmatch[reg_idx].rm_eo = -1; + + /* Set the points where matching start/end. */ + pmatch[0].rm_so = 0; + pmatch[0].rm_eo = mctx.match_last; + /* FIXME: This function should fail if mctx.match_last exceeds + the maximum possible regoff_t value. We need a new error + code REG_OVERFLOW. */ + + if (!preg->no_sub && nmatch > 1) + { + err = set_regs (preg, &mctx, nmatch, pmatch, + dfa->has_plural_match && dfa->nbackref > 0); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + + /* At last, add the offset to each register, since we slid + the buffers so that we could assume that the matching starts + from 0. */ + for (reg_idx = 0; reg_idx < nmatch; ++reg_idx) + if (pmatch[reg_idx].rm_so != -1) + { +#ifdef RE_ENABLE_I18N + if (BE (mctx.input.offsets_needed != 0, 0)) + { + pmatch[reg_idx].rm_so = + (pmatch[reg_idx].rm_so == mctx.input.valid_len + ? mctx.input.valid_raw_len + : mctx.input.offsets[pmatch[reg_idx].rm_so]); + pmatch[reg_idx].rm_eo = + (pmatch[reg_idx].rm_eo == mctx.input.valid_len + ? mctx.input.valid_raw_len + : mctx.input.offsets[pmatch[reg_idx].rm_eo]); + } +#else + assert (mctx.input.offsets_needed == 0); +#endif + pmatch[reg_idx].rm_so += match_first; + pmatch[reg_idx].rm_eo += match_first; + } + for (reg_idx = 0; reg_idx < extra_nmatch; ++reg_idx) + { + pmatch[nmatch + reg_idx].rm_so = -1; + pmatch[nmatch + reg_idx].rm_eo = -1; + } + + if (dfa->subexp_map) + for (reg_idx = 0; reg_idx + 1 < nmatch; reg_idx++) + if (dfa->subexp_map[reg_idx] != reg_idx) + { + pmatch[reg_idx + 1].rm_so + = pmatch[dfa->subexp_map[reg_idx] + 1].rm_so; + pmatch[reg_idx + 1].rm_eo + = pmatch[dfa->subexp_map[reg_idx] + 1].rm_eo; + } + } + + free_return: + re_free (mctx.state_log); + if (dfa->nbackref) + match_ctx_free (&mctx); + re_string_destruct (&mctx.input); + return err; +} + +static reg_errcode_t +__attribute_warn_unused_result__ +prune_impossible_nodes (re_match_context_t *mctx) +{ + const re_dfa_t *const dfa = mctx->dfa; + Idx halt_node, match_last; + reg_errcode_t ret; + re_dfastate_t **sifted_states; + re_dfastate_t **lim_states = NULL; + re_sift_context_t sctx; +#ifdef DEBUG + assert (mctx->state_log != NULL); +#endif + match_last = mctx->match_last; + halt_node = mctx->last_node; + + /* Avoid overflow. */ + if (BE (MIN (IDX_MAX, SIZE_MAX / sizeof (re_dfastate_t *)) <= match_last, 0)) + return REG_ESPACE; + + sifted_states = re_malloc (re_dfastate_t *, match_last + 1); + if (BE (sifted_states == NULL, 0)) + { + ret = REG_ESPACE; + goto free_return; + } + if (dfa->nbackref) + { + lim_states = re_malloc (re_dfastate_t *, match_last + 1); + if (BE (lim_states == NULL, 0)) + { + ret = REG_ESPACE; + goto free_return; + } + while (1) + { + memset (lim_states, '\0', + sizeof (re_dfastate_t *) * (match_last + 1)); + sift_ctx_init (&sctx, sifted_states, lim_states, halt_node, + match_last); + ret = sift_states_backward (mctx, &sctx); + re_node_set_free (&sctx.limits); + if (BE (ret != REG_NOERROR, 0)) + goto free_return; + if (sifted_states[0] != NULL || lim_states[0] != NULL) + break; + do + { + --match_last; + if (! REG_VALID_INDEX (match_last)) + { + ret = REG_NOMATCH; + goto free_return; + } + } while (mctx->state_log[match_last] == NULL + || !mctx->state_log[match_last]->halt); + halt_node = check_halt_state_context (mctx, + mctx->state_log[match_last], + match_last); + } + ret = merge_state_array (dfa, sifted_states, lim_states, + match_last + 1); + re_free (lim_states); + lim_states = NULL; + if (BE (ret != REG_NOERROR, 0)) + goto free_return; + } + else + { + sift_ctx_init (&sctx, sifted_states, lim_states, halt_node, match_last); + ret = sift_states_backward (mctx, &sctx); + re_node_set_free (&sctx.limits); + if (BE (ret != REG_NOERROR, 0)) + goto free_return; + if (sifted_states[0] == NULL) + { + ret = REG_NOMATCH; + goto free_return; + } + } + re_free (mctx->state_log); + mctx->state_log = sifted_states; + sifted_states = NULL; + mctx->last_node = halt_node; + mctx->match_last = match_last; + ret = REG_NOERROR; + free_return: + re_free (sifted_states); + re_free (lim_states); + return ret; +} + +/* Acquire an initial state and return it. + We must select appropriate initial state depending on the context, + since initial states may have constraints like "\<", "^", etc.. */ + +static inline re_dfastate_t * +__attribute ((always_inline)) internal_function +acquire_init_state_context (reg_errcode_t *err, const re_match_context_t *mctx, + Idx idx) +{ + const re_dfa_t *const dfa = mctx->dfa; + if (dfa->init_state->has_constraint) + { + unsigned int context; + context = re_string_context_at (&mctx->input, idx - 1, mctx->eflags); + if (IS_WORD_CONTEXT (context)) + return dfa->init_state_word; + else if (IS_ORDINARY_CONTEXT (context)) + return dfa->init_state; + else if (IS_BEGBUF_CONTEXT (context) && IS_NEWLINE_CONTEXT (context)) + return dfa->init_state_begbuf; + else if (IS_NEWLINE_CONTEXT (context)) + return dfa->init_state_nl; + else if (IS_BEGBUF_CONTEXT (context)) + { + /* It is relatively rare case, then calculate on demand. */ + return re_acquire_state_context (err, dfa, + dfa->init_state->entrance_nodes, + context); + } + else + /* Must not happen? */ + return dfa->init_state; + } + else + return dfa->init_state; +} + +/* Check whether the regular expression match input string INPUT or not, + and return the index where the matching end. Return REG_MISSING if + there is no match, and return REG_ERROR in case of an error. + FL_LONGEST_MATCH means we want the POSIX longest matching. + If P_MATCH_FIRST is not NULL, and the match fails, it is set to the + next place where we may want to try matching. + Note that the matcher assumes that the matching starts from the current + index of the buffer. */ + +static Idx +internal_function __attribute_warn_unused_result__ +check_matching (re_match_context_t *mctx, bool fl_longest_match, + Idx *p_match_first) +{ + const re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx match = 0; + Idx match_last = REG_MISSING; + Idx cur_str_idx = re_string_cur_idx (&mctx->input); + re_dfastate_t *cur_state; + bool at_init_state = p_match_first != NULL; + Idx next_start_idx = cur_str_idx; + + err = REG_NOERROR; + cur_state = acquire_init_state_context (&err, mctx, cur_str_idx); + /* An initial state must not be NULL (invalid). */ + if (BE (cur_state == NULL, 0)) + { + assert (err == REG_ESPACE); + return REG_ERROR; + } + + if (mctx->state_log != NULL) + { + mctx->state_log[cur_str_idx] = cur_state; + + /* Check OP_OPEN_SUBEXP in the initial state in case that we use them + later. E.g. Processing back references. */ + if (BE (dfa->nbackref, 0)) + { + at_init_state = false; + err = check_subexp_matching_top (mctx, &cur_state->nodes, 0); + if (BE (err != REG_NOERROR, 0)) + return err; + + if (cur_state->has_backref) + { + err = transit_state_bkref (mctx, &cur_state->nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + } + + /* If the RE accepts NULL string. */ + if (BE (cur_state->halt, 0)) + { + if (!cur_state->has_constraint + || check_halt_state_context (mctx, cur_state, cur_str_idx)) + { + if (!fl_longest_match) + return cur_str_idx; + else + { + match_last = cur_str_idx; + match = 1; + } + } + } + + while (!re_string_eoi (&mctx->input)) + { + re_dfastate_t *old_state = cur_state; + Idx next_char_idx = re_string_cur_idx (&mctx->input) + 1; + + if ((BE (next_char_idx >= mctx->input.bufs_len, 0) + && mctx->input.bufs_len < mctx->input.len) + || (BE (next_char_idx >= mctx->input.valid_len, 0) + && mctx->input.valid_len < mctx->input.len)) + { + err = extend_buffers (mctx); + if (BE (err != REG_NOERROR, 0)) + { + assert (err == REG_ESPACE); + return REG_ERROR; + } + } + + cur_state = transit_state (&err, mctx, cur_state); + if (mctx->state_log != NULL) + cur_state = merge_state_with_log (&err, mctx, cur_state); + + if (cur_state == NULL) + { + /* Reached the invalid state or an error. Try to recover a valid + state using the state log, if available and if we have not + already found a valid (even if not the longest) match. */ + if (BE (err != REG_NOERROR, 0)) + return REG_ERROR; + + if (mctx->state_log == NULL + || (match && !fl_longest_match) + || (cur_state = find_recover_state (&err, mctx)) == NULL) + break; + } + + if (BE (at_init_state, 0)) + { + if (old_state == cur_state) + next_start_idx = next_char_idx; + else + at_init_state = false; + } + + if (cur_state->halt) + { + /* Reached a halt state. + Check the halt state can satisfy the current context. */ + if (!cur_state->has_constraint + || check_halt_state_context (mctx, cur_state, + re_string_cur_idx (&mctx->input))) + { + /* We found an appropriate halt state. */ + match_last = re_string_cur_idx (&mctx->input); + match = 1; + + /* We found a match, do not modify match_first below. */ + p_match_first = NULL; + if (!fl_longest_match) + break; + } + } + } + + if (p_match_first) + *p_match_first += next_start_idx; + + return match_last; +} + +/* Check NODE match the current context. */ + +static bool +internal_function +check_halt_node_context (const re_dfa_t *dfa, Idx node, unsigned int context) +{ + re_token_type_t type = dfa->nodes[node].type; + unsigned int constraint = dfa->nodes[node].constraint; + if (type != END_OF_RE) + return false; + if (!constraint) + return true; + if (NOT_SATISFY_NEXT_CONSTRAINT (constraint, context)) + return false; + return true; +} + +/* Check the halt state STATE match the current context. + Return 0 if not match, if the node, STATE has, is a halt node and + match the context, return the node. */ + +static Idx +internal_function +check_halt_state_context (const re_match_context_t *mctx, + const re_dfastate_t *state, Idx idx) +{ + Idx i; + unsigned int context; +#ifdef DEBUG + assert (state->halt); +#endif + context = re_string_context_at (&mctx->input, idx, mctx->eflags); + for (i = 0; i < state->nodes.nelem; ++i) + if (check_halt_node_context (mctx->dfa, state->nodes.elems[i], context)) + return state->nodes.elems[i]; + return 0; +} + +/* Compute the next node to which "NFA" transit from NODE("NFA" is a NFA + corresponding to the DFA). + Return the destination node, and update EPS_VIA_NODES; + return REG_MISSING in case of errors. */ + +static Idx +internal_function +proceed_next_node (const re_match_context_t *mctx, Idx nregs, regmatch_t *regs, + Idx *pidx, Idx node, re_node_set *eps_via_nodes, + struct re_fail_stack_t *fs) +{ + const re_dfa_t *const dfa = mctx->dfa; + Idx i; + bool ok; + if (IS_EPSILON_NODE (dfa->nodes[node].type)) + { + re_node_set *cur_nodes = &mctx->state_log[*pidx]->nodes; + re_node_set *edests = &dfa->edests[node]; + Idx dest_node; + ok = re_node_set_insert (eps_via_nodes, node); + if (BE (! ok, 0)) + return REG_ERROR; + /* Pick up a valid destination, or return REG_MISSING if none + is found. */ + for (dest_node = REG_MISSING, i = 0; i < edests->nelem; ++i) + { + Idx candidate = edests->elems[i]; + if (!re_node_set_contains (cur_nodes, candidate)) + continue; + if (dest_node == REG_MISSING) + dest_node = candidate; + + else + { + /* In order to avoid infinite loop like "(a*)*", return the second + epsilon-transition if the first was already considered. */ + if (re_node_set_contains (eps_via_nodes, dest_node)) + return candidate; + + /* Otherwise, push the second epsilon-transition on the fail stack. */ + else if (fs != NULL + && push_fail_stack (fs, *pidx, candidate, nregs, regs, + eps_via_nodes)) + return REG_ERROR; + + /* We know we are going to exit. */ + break; + } + } + return dest_node; + } + else + { + Idx naccepted = 0; + re_token_type_t type = dfa->nodes[node].type; + +#ifdef RE_ENABLE_I18N + if (dfa->nodes[node].accept_mb) + naccepted = check_node_accept_bytes (dfa, node, &mctx->input, *pidx); + else +#endif /* RE_ENABLE_I18N */ + if (type == OP_BACK_REF) + { + Idx subexp_idx = dfa->nodes[node].opr.idx + 1; + naccepted = regs[subexp_idx].rm_eo - regs[subexp_idx].rm_so; + if (fs != NULL) + { + if (regs[subexp_idx].rm_so == -1 || regs[subexp_idx].rm_eo == -1) + return REG_MISSING; + else if (naccepted) + { + char *buf = (char *) re_string_get_buffer (&mctx->input); + if (memcmp (buf + regs[subexp_idx].rm_so, buf + *pidx, + naccepted) != 0) + return REG_MISSING; + } + } + + if (naccepted == 0) + { + Idx dest_node; + ok = re_node_set_insert (eps_via_nodes, node); + if (BE (! ok, 0)) + return REG_ERROR; + dest_node = dfa->edests[node].elems[0]; + if (re_node_set_contains (&mctx->state_log[*pidx]->nodes, + dest_node)) + return dest_node; + } + } + + if (naccepted != 0 + || check_node_accept (mctx, dfa->nodes + node, *pidx)) + { + Idx dest_node = dfa->nexts[node]; + *pidx = (naccepted == 0) ? *pidx + 1 : *pidx + naccepted; + if (fs && (*pidx > mctx->match_last || mctx->state_log[*pidx] == NULL + || !re_node_set_contains (&mctx->state_log[*pidx]->nodes, + dest_node))) + return REG_MISSING; + re_node_set_empty (eps_via_nodes); + return dest_node; + } + } + return REG_MISSING; +} + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +push_fail_stack (struct re_fail_stack_t *fs, Idx str_idx, Idx dest_node, + Idx nregs, regmatch_t *regs, re_node_set *eps_via_nodes) +{ + reg_errcode_t err; + Idx num = fs->num++; + if (fs->num == fs->alloc) + { + struct re_fail_stack_ent_t *new_array; + new_array = realloc (fs->stack, (sizeof (struct re_fail_stack_ent_t) + * fs->alloc * 2)); + if (new_array == NULL) + return REG_ESPACE; + fs->alloc *= 2; + fs->stack = new_array; + } + fs->stack[num].idx = str_idx; + fs->stack[num].node = dest_node; + fs->stack[num].regs = re_malloc (regmatch_t, nregs); + if (fs->stack[num].regs == NULL) + return REG_ESPACE; + memcpy (fs->stack[num].regs, regs, sizeof (regmatch_t) * nregs); + err = re_node_set_init_copy (&fs->stack[num].eps_via_nodes, eps_via_nodes); + return err; +} + +static Idx +internal_function +pop_fail_stack (struct re_fail_stack_t *fs, Idx *pidx, Idx nregs, + regmatch_t *regs, re_node_set *eps_via_nodes) +{ + Idx num = --fs->num; + assert (REG_VALID_INDEX (num)); + *pidx = fs->stack[num].idx; + memcpy (regs, fs->stack[num].regs, sizeof (regmatch_t) * nregs); + re_node_set_free (eps_via_nodes); + re_free (fs->stack[num].regs); + *eps_via_nodes = fs->stack[num].eps_via_nodes; + return fs->stack[num].node; +} + +/* Set the positions where the subexpressions are starts/ends to registers + PMATCH. + Note: We assume that pmatch[0] is already set, and + pmatch[i].rm_so == pmatch[i].rm_eo == -1 for 0 < i < nmatch. */ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +set_regs (const regex_t *preg, const re_match_context_t *mctx, size_t nmatch, + regmatch_t *pmatch, bool fl_backtrack) +{ + const re_dfa_t *dfa = (const re_dfa_t *) preg->buffer; + Idx idx, cur_node; + re_node_set eps_via_nodes; + struct re_fail_stack_t *fs; + struct re_fail_stack_t fs_body = { 0, 2, NULL }; + regmatch_t *prev_idx_match; + bool prev_idx_match_malloced = false; + +#ifdef DEBUG + assert (nmatch > 1); + assert (mctx->state_log != NULL); +#endif + if (fl_backtrack) + { + fs = &fs_body; + fs->stack = re_malloc (struct re_fail_stack_ent_t, fs->alloc); + if (fs->stack == NULL) + return REG_ESPACE; + } + else + fs = NULL; + + cur_node = dfa->init_node; + re_node_set_init_empty (&eps_via_nodes); + + if (__libc_use_alloca (nmatch * sizeof (regmatch_t))) + prev_idx_match = (regmatch_t *) alloca (nmatch * sizeof (regmatch_t)); + else + { + prev_idx_match = re_malloc (regmatch_t, nmatch); + if (prev_idx_match == NULL) + { + free_fail_stack_return (fs); + return REG_ESPACE; + } + prev_idx_match_malloced = true; + } + memcpy (prev_idx_match, pmatch, sizeof (regmatch_t) * nmatch); + + for (idx = pmatch[0].rm_so; idx <= pmatch[0].rm_eo ;) + { + update_regs (dfa, pmatch, prev_idx_match, cur_node, idx, nmatch); + + if (idx == pmatch[0].rm_eo && cur_node == mctx->last_node) + { + Idx reg_idx; + if (fs) + { + for (reg_idx = 0; reg_idx < nmatch; ++reg_idx) + if (pmatch[reg_idx].rm_so > -1 && pmatch[reg_idx].rm_eo == -1) + break; + if (reg_idx == nmatch) + { + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + return free_fail_stack_return (fs); + } + cur_node = pop_fail_stack (fs, &idx, nmatch, pmatch, + &eps_via_nodes); + } + else + { + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + return REG_NOERROR; + } + } + + /* Proceed to next node. */ + cur_node = proceed_next_node (mctx, nmatch, pmatch, &idx, cur_node, + &eps_via_nodes, fs); + + if (BE (! REG_VALID_INDEX (cur_node), 0)) + { + if (BE (cur_node == REG_ERROR, 0)) + { + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + free_fail_stack_return (fs); + return REG_ESPACE; + } + if (fs) + cur_node = pop_fail_stack (fs, &idx, nmatch, pmatch, + &eps_via_nodes); + else + { + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + return REG_NOMATCH; + } + } + } + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + return free_fail_stack_return (fs); +} + +static reg_errcode_t +internal_function +free_fail_stack_return (struct re_fail_stack_t *fs) +{ + if (fs) + { + Idx fs_idx; + for (fs_idx = 0; fs_idx < fs->num; ++fs_idx) + { + re_node_set_free (&fs->stack[fs_idx].eps_via_nodes); + re_free (fs->stack[fs_idx].regs); + } + re_free (fs->stack); + } + return REG_NOERROR; +} + +static void +internal_function +update_regs (const re_dfa_t *dfa, regmatch_t *pmatch, + regmatch_t *prev_idx_match, Idx cur_node, Idx cur_idx, Idx nmatch) +{ + int type = dfa->nodes[cur_node].type; + if (type == OP_OPEN_SUBEXP) + { + Idx reg_num = dfa->nodes[cur_node].opr.idx + 1; + + /* We are at the first node of this sub expression. */ + if (reg_num < nmatch) + { + pmatch[reg_num].rm_so = cur_idx; + pmatch[reg_num].rm_eo = -1; + } + } + else if (type == OP_CLOSE_SUBEXP) + { + Idx reg_num = dfa->nodes[cur_node].opr.idx + 1; + if (reg_num < nmatch) + { + /* We are at the last node of this sub expression. */ + if (pmatch[reg_num].rm_so < cur_idx) + { + pmatch[reg_num].rm_eo = cur_idx; + /* This is a non-empty match or we are not inside an optional + subexpression. Accept this right away. */ + memcpy (prev_idx_match, pmatch, sizeof (regmatch_t) * nmatch); + } + else + { + if (dfa->nodes[cur_node].opt_subexp + && prev_idx_match[reg_num].rm_so != -1) + /* We transited through an empty match for an optional + subexpression, like (a?)*, and this is not the subexp's + first match. Copy back the old content of the registers + so that matches of an inner subexpression are undone as + well, like in ((a?))*. */ + memcpy (pmatch, prev_idx_match, sizeof (regmatch_t) * nmatch); + else + /* We completed a subexpression, but it may be part of + an optional one, so do not update PREV_IDX_MATCH. */ + pmatch[reg_num].rm_eo = cur_idx; + } + } + } +} + +/* This function checks the STATE_LOG from the SCTX->last_str_idx to 0 + and sift the nodes in each states according to the following rules. + Updated state_log will be wrote to STATE_LOG. + + Rules: We throw away the Node 'a' in the STATE_LOG[STR_IDX] if... + 1. When STR_IDX == MATCH_LAST(the last index in the state_log): + If 'a' isn't the LAST_NODE and 'a' can't epsilon transit to + the LAST_NODE, we throw away the node 'a'. + 2. When 0 <= STR_IDX < MATCH_LAST and 'a' accepts + string 's' and transit to 'b': + i. If 'b' isn't in the STATE_LOG[STR_IDX+strlen('s')], we throw + away the node 'a'. + ii. If 'b' is in the STATE_LOG[STR_IDX+strlen('s')] but 'b' is + thrown away, we throw away the node 'a'. + 3. When 0 <= STR_IDX < MATCH_LAST and 'a' epsilon transit to 'b': + i. If 'b' isn't in the STATE_LOG[STR_IDX], we throw away the + node 'a'. + ii. If 'b' is in the STATE_LOG[STR_IDX] but 'b' is thrown away, + we throw away the node 'a'. */ + +#define STATE_NODE_CONTAINS(state,node) \ + ((state) != NULL && re_node_set_contains (&(state)->nodes, node)) + +static reg_errcode_t +internal_function +sift_states_backward (const re_match_context_t *mctx, re_sift_context_t *sctx) +{ + reg_errcode_t err; + int null_cnt = 0; + Idx str_idx = sctx->last_str_idx; + re_node_set cur_dest; + +#ifdef DEBUG + assert (mctx->state_log != NULL && mctx->state_log[str_idx] != NULL); +#endif + + /* Build sifted state_log[str_idx]. It has the nodes which can epsilon + transit to the last_node and the last_node itself. */ + err = re_node_set_init_1 (&cur_dest, sctx->last_node); + if (BE (err != REG_NOERROR, 0)) + return err; + err = update_cur_sifted_state (mctx, sctx, str_idx, &cur_dest); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + + /* Then check each states in the state_log. */ + while (str_idx > 0) + { + /* Update counters. */ + null_cnt = (sctx->sifted_states[str_idx] == NULL) ? null_cnt + 1 : 0; + if (null_cnt > mctx->max_mb_elem_len) + { + memset (sctx->sifted_states, '\0', + sizeof (re_dfastate_t *) * str_idx); + re_node_set_free (&cur_dest); + return REG_NOERROR; + } + re_node_set_empty (&cur_dest); + --str_idx; + + if (mctx->state_log[str_idx]) + { + err = build_sifted_states (mctx, sctx, str_idx, &cur_dest); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + + /* Add all the nodes which satisfy the following conditions: + - It can epsilon transit to a node in CUR_DEST. + - It is in CUR_SRC. + And update state_log. */ + err = update_cur_sifted_state (mctx, sctx, str_idx, &cur_dest); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + err = REG_NOERROR; + free_return: + re_node_set_free (&cur_dest); + return err; +} + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +build_sifted_states (const re_match_context_t *mctx, re_sift_context_t *sctx, + Idx str_idx, re_node_set *cur_dest) +{ + const re_dfa_t *const dfa = mctx->dfa; + const re_node_set *cur_src = &mctx->state_log[str_idx]->non_eps_nodes; + Idx i; + + /* Then build the next sifted state. + We build the next sifted state on 'cur_dest', and update + 'sifted_states[str_idx]' with 'cur_dest'. + Note: + 'cur_dest' is the sifted state from 'state_log[str_idx + 1]'. + 'cur_src' points the node_set of the old 'state_log[str_idx]' + (with the epsilon nodes pre-filtered out). */ + for (i = 0; i < cur_src->nelem; i++) + { + Idx prev_node = cur_src->elems[i]; + int naccepted = 0; + bool ok; + +#ifdef DEBUG + re_token_type_t type = dfa->nodes[prev_node].type; + assert (!IS_EPSILON_NODE (type)); +#endif +#ifdef RE_ENABLE_I18N + /* If the node may accept "multi byte". */ + if (dfa->nodes[prev_node].accept_mb) + naccepted = sift_states_iter_mb (mctx, sctx, prev_node, + str_idx, sctx->last_str_idx); +#endif /* RE_ENABLE_I18N */ + + /* We don't check backreferences here. + See update_cur_sifted_state(). */ + if (!naccepted + && check_node_accept (mctx, dfa->nodes + prev_node, str_idx) + && STATE_NODE_CONTAINS (sctx->sifted_states[str_idx + 1], + dfa->nexts[prev_node])) + naccepted = 1; + + if (naccepted == 0) + continue; + + if (sctx->limits.nelem) + { + Idx to_idx = str_idx + naccepted; + if (check_dst_limits (mctx, &sctx->limits, + dfa->nexts[prev_node], to_idx, + prev_node, str_idx)) + continue; + } + ok = re_node_set_insert (cur_dest, prev_node); + if (BE (! ok, 0)) + return REG_ESPACE; + } + + return REG_NOERROR; +} + +/* Helper functions. */ + +static reg_errcode_t +internal_function +clean_state_log_if_needed (re_match_context_t *mctx, Idx next_state_log_idx) +{ + Idx top = mctx->state_log_top; + + if ((next_state_log_idx >= mctx->input.bufs_len + && mctx->input.bufs_len < mctx->input.len) + || (next_state_log_idx >= mctx->input.valid_len + && mctx->input.valid_len < mctx->input.len)) + { + reg_errcode_t err; + err = extend_buffers (mctx); + if (BE (err != REG_NOERROR, 0)) + return err; + } + + if (top < next_state_log_idx) + { + memset (mctx->state_log + top + 1, '\0', + sizeof (re_dfastate_t *) * (next_state_log_idx - top)); + mctx->state_log_top = next_state_log_idx; + } + return REG_NOERROR; +} + +static reg_errcode_t +internal_function +merge_state_array (const re_dfa_t *dfa, re_dfastate_t **dst, + re_dfastate_t **src, Idx num) +{ + Idx st_idx; + reg_errcode_t err; + for (st_idx = 0; st_idx < num; ++st_idx) + { + if (dst[st_idx] == NULL) + dst[st_idx] = src[st_idx]; + else if (src[st_idx] != NULL) + { + re_node_set merged_set; + err = re_node_set_init_union (&merged_set, &dst[st_idx]->nodes, + &src[st_idx]->nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + dst[st_idx] = re_acquire_state (&err, dfa, &merged_set); + re_node_set_free (&merged_set); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + return REG_NOERROR; +} + +static reg_errcode_t +internal_function +update_cur_sifted_state (const re_match_context_t *mctx, + re_sift_context_t *sctx, Idx str_idx, + re_node_set *dest_nodes) +{ + const re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err = REG_NOERROR; + const re_node_set *candidates; + candidates = ((mctx->state_log[str_idx] == NULL) ? NULL + : &mctx->state_log[str_idx]->nodes); + + if (dest_nodes->nelem == 0) + sctx->sifted_states[str_idx] = NULL; + else + { + if (candidates) + { + /* At first, add the nodes which can epsilon transit to a node in + DEST_NODE. */ + err = add_epsilon_src_nodes (dfa, dest_nodes, candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + + /* Then, check the limitations in the current sift_context. */ + if (sctx->limits.nelem) + { + err = check_subexp_limits (dfa, dest_nodes, candidates, &sctx->limits, + mctx->bkref_ents, str_idx); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + + sctx->sifted_states[str_idx] = re_acquire_state (&err, dfa, dest_nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + } + + if (candidates && mctx->state_log[str_idx]->has_backref) + { + err = sift_states_bkref (mctx, sctx, str_idx, candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + } + return REG_NOERROR; +} + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +add_epsilon_src_nodes (const re_dfa_t *dfa, re_node_set *dest_nodes, + const re_node_set *candidates) +{ + reg_errcode_t err = REG_NOERROR; + Idx i; + + re_dfastate_t *state = re_acquire_state (&err, dfa, dest_nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + + if (!state->inveclosure.alloc) + { + err = re_node_set_alloc (&state->inveclosure, dest_nodes->nelem); + if (BE (err != REG_NOERROR, 0)) + return REG_ESPACE; + for (i = 0; i < dest_nodes->nelem; i++) + { + err = re_node_set_merge (&state->inveclosure, + dfa->inveclosures + dest_nodes->elems[i]); + if (BE (err != REG_NOERROR, 0)) + return REG_ESPACE; + } + } + return re_node_set_add_intersect (dest_nodes, candidates, + &state->inveclosure); +} + +static reg_errcode_t +internal_function +sub_epsilon_src_nodes (const re_dfa_t *dfa, Idx node, re_node_set *dest_nodes, + const re_node_set *candidates) +{ + Idx ecl_idx; + reg_errcode_t err; + re_node_set *inv_eclosure = dfa->inveclosures + node; + re_node_set except_nodes; + re_node_set_init_empty (&except_nodes); + for (ecl_idx = 0; ecl_idx < inv_eclosure->nelem; ++ecl_idx) + { + Idx cur_node = inv_eclosure->elems[ecl_idx]; + if (cur_node == node) + continue; + if (IS_EPSILON_NODE (dfa->nodes[cur_node].type)) + { + Idx edst1 = dfa->edests[cur_node].elems[0]; + Idx edst2 = ((dfa->edests[cur_node].nelem > 1) + ? dfa->edests[cur_node].elems[1] : REG_MISSING); + if ((!re_node_set_contains (inv_eclosure, edst1) + && re_node_set_contains (dest_nodes, edst1)) + || (REG_VALID_NONZERO_INDEX (edst2) + && !re_node_set_contains (inv_eclosure, edst2) + && re_node_set_contains (dest_nodes, edst2))) + { + err = re_node_set_add_intersect (&except_nodes, candidates, + dfa->inveclosures + cur_node); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&except_nodes); + return err; + } + } + } + } + for (ecl_idx = 0; ecl_idx < inv_eclosure->nelem; ++ecl_idx) + { + Idx cur_node = inv_eclosure->elems[ecl_idx]; + if (!re_node_set_contains (&except_nodes, cur_node)) + { + Idx idx = re_node_set_contains (dest_nodes, cur_node) - 1; + re_node_set_remove_at (dest_nodes, idx); + } + } + re_node_set_free (&except_nodes); + return REG_NOERROR; +} + +static bool +internal_function +check_dst_limits (const re_match_context_t *mctx, const re_node_set *limits, + Idx dst_node, Idx dst_idx, Idx src_node, Idx src_idx) +{ + const re_dfa_t *const dfa = mctx->dfa; + Idx lim_idx, src_pos, dst_pos; + + Idx dst_bkref_idx = search_cur_bkref_entry (mctx, dst_idx); + Idx src_bkref_idx = search_cur_bkref_entry (mctx, src_idx); + for (lim_idx = 0; lim_idx < limits->nelem; ++lim_idx) + { + Idx subexp_idx; + struct re_backref_cache_entry *ent; + ent = mctx->bkref_ents + limits->elems[lim_idx]; + subexp_idx = dfa->nodes[ent->node].opr.idx; + + dst_pos = check_dst_limits_calc_pos (mctx, limits->elems[lim_idx], + subexp_idx, dst_node, dst_idx, + dst_bkref_idx); + src_pos = check_dst_limits_calc_pos (mctx, limits->elems[lim_idx], + subexp_idx, src_node, src_idx, + src_bkref_idx); + + /* In case of: + <src> <dst> ( <subexp> ) + ( <subexp> ) <src> <dst> + ( <subexp1> <src> <subexp2> <dst> <subexp3> ) */ + if (src_pos == dst_pos) + continue; /* This is unrelated limitation. */ + else + return true; + } + return false; +} + +static int +internal_function +check_dst_limits_calc_pos_1 (const re_match_context_t *mctx, int boundaries, + Idx subexp_idx, Idx from_node, Idx bkref_idx) +{ + const re_dfa_t *const dfa = mctx->dfa; + const re_node_set *eclosures = dfa->eclosures + from_node; + Idx node_idx; + + /* Else, we are on the boundary: examine the nodes on the epsilon + closure. */ + for (node_idx = 0; node_idx < eclosures->nelem; ++node_idx) + { + Idx node = eclosures->elems[node_idx]; + switch (dfa->nodes[node].type) + { + case OP_BACK_REF: + if (bkref_idx != REG_MISSING) + { + struct re_backref_cache_entry *ent = mctx->bkref_ents + bkref_idx; + do + { + Idx dst; + int cpos; + + if (ent->node != node) + continue; + + if (subexp_idx < BITSET_WORD_BITS + && !(ent->eps_reachable_subexps_map + & ((bitset_word_t) 1 << subexp_idx))) + continue; + + /* Recurse trying to reach the OP_OPEN_SUBEXP and + OP_CLOSE_SUBEXP cases below. But, if the + destination node is the same node as the source + node, don't recurse because it would cause an + infinite loop: a regex that exhibits this behavior + is ()\1*\1* */ + dst = dfa->edests[node].elems[0]; + if (dst == from_node) + { + if (boundaries & 1) + return -1; + else /* if (boundaries & 2) */ + return 0; + } + + cpos = + check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx, + dst, bkref_idx); + if (cpos == -1 /* && (boundaries & 1) */) + return -1; + if (cpos == 0 && (boundaries & 2)) + return 0; + + if (subexp_idx < BITSET_WORD_BITS) + ent->eps_reachable_subexps_map + &= ~((bitset_word_t) 1 << subexp_idx); + } + while (ent++->more); + } + break; + + case OP_OPEN_SUBEXP: + if ((boundaries & 1) && subexp_idx == dfa->nodes[node].opr.idx) + return -1; + break; + + case OP_CLOSE_SUBEXP: + if ((boundaries & 2) && subexp_idx == dfa->nodes[node].opr.idx) + return 0; + break; + + default: + break; + } + } + + return (boundaries & 2) ? 1 : 0; +} + +static int +internal_function +check_dst_limits_calc_pos (const re_match_context_t *mctx, Idx limit, + Idx subexp_idx, Idx from_node, Idx str_idx, + Idx bkref_idx) +{ + struct re_backref_cache_entry *lim = mctx->bkref_ents + limit; + int boundaries; + + /* If we are outside the range of the subexpression, return -1 or 1. */ + if (str_idx < lim->subexp_from) + return -1; + + if (lim->subexp_to < str_idx) + return 1; + + /* If we are within the subexpression, return 0. */ + boundaries = (str_idx == lim->subexp_from); + boundaries |= (str_idx == lim->subexp_to) << 1; + if (boundaries == 0) + return 0; + + /* Else, examine epsilon closure. */ + return check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx, + from_node, bkref_idx); +} + +/* Check the limitations of sub expressions LIMITS, and remove the nodes + which are against limitations from DEST_NODES. */ + +static reg_errcode_t +internal_function +check_subexp_limits (const re_dfa_t *dfa, re_node_set *dest_nodes, + const re_node_set *candidates, re_node_set *limits, + struct re_backref_cache_entry *bkref_ents, Idx str_idx) +{ + reg_errcode_t err; + Idx node_idx, lim_idx; + + for (lim_idx = 0; lim_idx < limits->nelem; ++lim_idx) + { + Idx subexp_idx; + struct re_backref_cache_entry *ent; + ent = bkref_ents + limits->elems[lim_idx]; + + if (str_idx <= ent->subexp_from || ent->str_idx < str_idx) + continue; /* This is unrelated limitation. */ + + subexp_idx = dfa->nodes[ent->node].opr.idx; + if (ent->subexp_to == str_idx) + { + Idx ops_node = REG_MISSING; + Idx cls_node = REG_MISSING; + for (node_idx = 0; node_idx < dest_nodes->nelem; ++node_idx) + { + Idx node = dest_nodes->elems[node_idx]; + re_token_type_t type = dfa->nodes[node].type; + if (type == OP_OPEN_SUBEXP + && subexp_idx == dfa->nodes[node].opr.idx) + ops_node = node; + else if (type == OP_CLOSE_SUBEXP + && subexp_idx == dfa->nodes[node].opr.idx) + cls_node = node; + } + + /* Check the limitation of the open subexpression. */ + /* Note that (ent->subexp_to = str_idx != ent->subexp_from). */ + if (REG_VALID_INDEX (ops_node)) + { + err = sub_epsilon_src_nodes (dfa, ops_node, dest_nodes, + candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + } + + /* Check the limitation of the close subexpression. */ + if (REG_VALID_INDEX (cls_node)) + for (node_idx = 0; node_idx < dest_nodes->nelem; ++node_idx) + { + Idx node = dest_nodes->elems[node_idx]; + if (!re_node_set_contains (dfa->inveclosures + node, + cls_node) + && !re_node_set_contains (dfa->eclosures + node, + cls_node)) + { + /* It is against this limitation. + Remove it form the current sifted state. */ + err = sub_epsilon_src_nodes (dfa, node, dest_nodes, + candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + --node_idx; + } + } + } + else /* (ent->subexp_to != str_idx) */ + { + for (node_idx = 0; node_idx < dest_nodes->nelem; ++node_idx) + { + Idx node = dest_nodes->elems[node_idx]; + re_token_type_t type = dfa->nodes[node].type; + if (type == OP_CLOSE_SUBEXP || type == OP_OPEN_SUBEXP) + { + if (subexp_idx != dfa->nodes[node].opr.idx) + continue; + /* It is against this limitation. + Remove it form the current sifted state. */ + err = sub_epsilon_src_nodes (dfa, node, dest_nodes, + candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + } + } + return REG_NOERROR; +} + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +sift_states_bkref (const re_match_context_t *mctx, re_sift_context_t *sctx, + Idx str_idx, const re_node_set *candidates) +{ + const re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx node_idx, node; + re_sift_context_t local_sctx; + Idx first_idx = search_cur_bkref_entry (mctx, str_idx); + + if (first_idx == REG_MISSING) + return REG_NOERROR; + + local_sctx.sifted_states = NULL; /* Mark that it hasn't been initialized. */ + + for (node_idx = 0; node_idx < candidates->nelem; ++node_idx) + { + Idx enabled_idx; + re_token_type_t type; + struct re_backref_cache_entry *entry; + node = candidates->elems[node_idx]; + type = dfa->nodes[node].type; + /* Avoid infinite loop for the REs like "()\1+". */ + if (node == sctx->last_node && str_idx == sctx->last_str_idx) + continue; + if (type != OP_BACK_REF) + continue; + + entry = mctx->bkref_ents + first_idx; + enabled_idx = first_idx; + do + { + Idx subexp_len; + Idx to_idx; + Idx dst_node; + bool ok; + re_dfastate_t *cur_state; + + if (entry->node != node) + continue; + subexp_len = entry->subexp_to - entry->subexp_from; + to_idx = str_idx + subexp_len; + dst_node = (subexp_len ? dfa->nexts[node] + : dfa->edests[node].elems[0]); + + if (to_idx > sctx->last_str_idx + || sctx->sifted_states[to_idx] == NULL + || !STATE_NODE_CONTAINS (sctx->sifted_states[to_idx], dst_node) + || check_dst_limits (mctx, &sctx->limits, node, + str_idx, dst_node, to_idx)) + continue; + + if (local_sctx.sifted_states == NULL) + { + local_sctx = *sctx; + err = re_node_set_init_copy (&local_sctx.limits, &sctx->limits); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + local_sctx.last_node = node; + local_sctx.last_str_idx = str_idx; + ok = re_node_set_insert (&local_sctx.limits, enabled_idx); + if (BE (! ok, 0)) + { + err = REG_ESPACE; + goto free_return; + } + cur_state = local_sctx.sifted_states[str_idx]; + err = sift_states_backward (mctx, &local_sctx); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + if (sctx->limited_states != NULL) + { + err = merge_state_array (dfa, sctx->limited_states, + local_sctx.sifted_states, + str_idx + 1); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + local_sctx.sifted_states[str_idx] = cur_state; + re_node_set_remove (&local_sctx.limits, enabled_idx); + + /* mctx->bkref_ents may have changed, reload the pointer. */ + entry = mctx->bkref_ents + enabled_idx; + } + while (enabled_idx++, entry++->more); + } + err = REG_NOERROR; + free_return: + if (local_sctx.sifted_states != NULL) + { + re_node_set_free (&local_sctx.limits); + } + + return err; +} + + +#ifdef RE_ENABLE_I18N +static int +internal_function +sift_states_iter_mb (const re_match_context_t *mctx, re_sift_context_t *sctx, + Idx node_idx, Idx str_idx, Idx max_str_idx) +{ + const re_dfa_t *const dfa = mctx->dfa; + int naccepted; + /* Check the node can accept "multi byte". */ + naccepted = check_node_accept_bytes (dfa, node_idx, &mctx->input, str_idx); + if (naccepted > 0 && str_idx + naccepted <= max_str_idx && + !STATE_NODE_CONTAINS (sctx->sifted_states[str_idx + naccepted], + dfa->nexts[node_idx])) + /* The node can't accept the "multi byte", or the + destination was already thrown away, then the node + could't accept the current input "multi byte". */ + naccepted = 0; + /* Otherwise, it is sure that the node could accept + 'naccepted' bytes input. */ + return naccepted; +} +#endif /* RE_ENABLE_I18N */ + + +/* Functions for state transition. */ + +/* Return the next state to which the current state STATE will transit by + accepting the current input byte, and update STATE_LOG if necessary. + If STATE can accept a multibyte char/collating element/back reference + update the destination of STATE_LOG. */ + +static re_dfastate_t * +internal_function __attribute_warn_unused_result__ +transit_state (reg_errcode_t *err, re_match_context_t *mctx, + re_dfastate_t *state) +{ + re_dfastate_t **trtable; + unsigned char ch; + +#ifdef RE_ENABLE_I18N + /* If the current state can accept multibyte. */ + if (BE (state->accept_mb, 0)) + { + *err = transit_state_mb (mctx, state); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + } +#endif /* RE_ENABLE_I18N */ + + /* Then decide the next state with the single byte. */ +#if 0 + if (0) + /* don't use transition table */ + return transit_state_sb (err, mctx, state); +#endif + + /* Use transition table */ + ch = re_string_fetch_byte (&mctx->input); + for (;;) + { + trtable = state->trtable; + if (BE (trtable != NULL, 1)) + return trtable[ch]; + + trtable = state->word_trtable; + if (BE (trtable != NULL, 1)) + { + unsigned int context; + context + = re_string_context_at (&mctx->input, + re_string_cur_idx (&mctx->input) - 1, + mctx->eflags); + if (IS_WORD_CONTEXT (context)) + return trtable[ch + SBC_MAX]; + else + return trtable[ch]; + } + + if (!build_trtable (mctx->dfa, state)) + { + *err = REG_ESPACE; + return NULL; + } + + /* Retry, we now have a transition table. */ + } +} + +/* Update the state_log if we need */ +static re_dfastate_t * +internal_function +merge_state_with_log (reg_errcode_t *err, re_match_context_t *mctx, + re_dfastate_t *next_state) +{ + const re_dfa_t *const dfa = mctx->dfa; + Idx cur_idx = re_string_cur_idx (&mctx->input); + + if (cur_idx > mctx->state_log_top) + { + mctx->state_log[cur_idx] = next_state; + mctx->state_log_top = cur_idx; + } + else if (mctx->state_log[cur_idx] == 0) + { + mctx->state_log[cur_idx] = next_state; + } + else + { + re_dfastate_t *pstate; + unsigned int context; + re_node_set next_nodes, *log_nodes, *table_nodes = NULL; + /* If (state_log[cur_idx] != 0), it implies that cur_idx is + the destination of a multibyte char/collating element/ + back reference. Then the next state is the union set of + these destinations and the results of the transition table. */ + pstate = mctx->state_log[cur_idx]; + log_nodes = pstate->entrance_nodes; + if (next_state != NULL) + { + table_nodes = next_state->entrance_nodes; + *err = re_node_set_init_union (&next_nodes, table_nodes, + log_nodes); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + } + else + next_nodes = *log_nodes; + /* Note: We already add the nodes of the initial state, + then we don't need to add them here. */ + + context = re_string_context_at (&mctx->input, + re_string_cur_idx (&mctx->input) - 1, + mctx->eflags); + next_state = mctx->state_log[cur_idx] + = re_acquire_state_context (err, dfa, &next_nodes, context); + /* We don't need to check errors here, since the return value of + this function is next_state and ERR is already set. */ + + if (table_nodes != NULL) + re_node_set_free (&next_nodes); + } + + if (BE (dfa->nbackref, 0) && next_state != NULL) + { + /* Check OP_OPEN_SUBEXP in the current state in case that we use them + later. We must check them here, since the back references in the + next state might use them. */ + *err = check_subexp_matching_top (mctx, &next_state->nodes, + cur_idx); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + + /* If the next state has back references. */ + if (next_state->has_backref) + { + *err = transit_state_bkref (mctx, &next_state->nodes); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + next_state = mctx->state_log[cur_idx]; + } + } + + return next_state; +} + +/* Skip bytes in the input that correspond to part of a + multi-byte match, then look in the log for a state + from which to restart matching. */ +static re_dfastate_t * +internal_function +find_recover_state (reg_errcode_t *err, re_match_context_t *mctx) +{ + re_dfastate_t *cur_state; + do + { + Idx max = mctx->state_log_top; + Idx cur_str_idx = re_string_cur_idx (&mctx->input); + + do + { + if (++cur_str_idx > max) + return NULL; + re_string_skip_bytes (&mctx->input, 1); + } + while (mctx->state_log[cur_str_idx] == NULL); + + cur_state = merge_state_with_log (err, mctx, NULL); + } + while (*err == REG_NOERROR && cur_state == NULL); + return cur_state; +} + +/* Helper functions for transit_state. */ + +/* From the node set CUR_NODES, pick up the nodes whose types are + OP_OPEN_SUBEXP and which have corresponding back references in the regular + expression. And register them to use them later for evaluating the + corresponding back references. */ + +static reg_errcode_t +internal_function +check_subexp_matching_top (re_match_context_t *mctx, re_node_set *cur_nodes, + Idx str_idx) +{ + const re_dfa_t *const dfa = mctx->dfa; + Idx node_idx; + reg_errcode_t err; + + /* TODO: This isn't efficient. + Because there might be more than one nodes whose types are + OP_OPEN_SUBEXP and whose index is SUBEXP_IDX, we must check all + nodes. + E.g. RE: (a){2} */ + for (node_idx = 0; node_idx < cur_nodes->nelem; ++node_idx) + { + Idx node = cur_nodes->elems[node_idx]; + if (dfa->nodes[node].type == OP_OPEN_SUBEXP + && dfa->nodes[node].opr.idx < BITSET_WORD_BITS + && (dfa->used_bkref_map + & ((bitset_word_t) 1 << dfa->nodes[node].opr.idx))) + { + err = match_ctx_add_subtop (mctx, node, str_idx); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + return REG_NOERROR; +} + +#if 0 +/* Return the next state to which the current state STATE will transit by + accepting the current input byte. */ + +static re_dfastate_t * +transit_state_sb (reg_errcode_t *err, re_match_context_t *mctx, + re_dfastate_t *state) +{ + const re_dfa_t *const dfa = mctx->dfa; + re_node_set next_nodes; + re_dfastate_t *next_state; + Idx node_cnt, cur_str_idx = re_string_cur_idx (&mctx->input); + unsigned int context; + + *err = re_node_set_alloc (&next_nodes, state->nodes.nelem + 1); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + for (node_cnt = 0; node_cnt < state->nodes.nelem; ++node_cnt) + { + Idx cur_node = state->nodes.elems[node_cnt]; + if (check_node_accept (mctx, dfa->nodes + cur_node, cur_str_idx)) + { + *err = re_node_set_merge (&next_nodes, + dfa->eclosures + dfa->nexts[cur_node]); + if (BE (*err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return NULL; + } + } + } + context = re_string_context_at (&mctx->input, cur_str_idx, mctx->eflags); + next_state = re_acquire_state_context (err, dfa, &next_nodes, context); + /* We don't need to check errors here, since the return value of + this function is next_state and ERR is already set. */ + + re_node_set_free (&next_nodes); + re_string_skip_bytes (&mctx->input, 1); + return next_state; +} +#endif + +#ifdef RE_ENABLE_I18N +static reg_errcode_t +internal_function +transit_state_mb (re_match_context_t *mctx, re_dfastate_t *pstate) +{ + const re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx i; + + for (i = 0; i < pstate->nodes.nelem; ++i) + { + re_node_set dest_nodes, *new_nodes; + Idx cur_node_idx = pstate->nodes.elems[i]; + int naccepted; + Idx dest_idx; + unsigned int context; + re_dfastate_t *dest_state; + + if (!dfa->nodes[cur_node_idx].accept_mb) + continue; + + if (dfa->nodes[cur_node_idx].constraint) + { + context = re_string_context_at (&mctx->input, + re_string_cur_idx (&mctx->input), + mctx->eflags); + if (NOT_SATISFY_NEXT_CONSTRAINT (dfa->nodes[cur_node_idx].constraint, + context)) + continue; + } + + /* How many bytes the node can accept? */ + naccepted = check_node_accept_bytes (dfa, cur_node_idx, &mctx->input, + re_string_cur_idx (&mctx->input)); + if (naccepted == 0) + continue; + + /* The node can accepts 'naccepted' bytes. */ + dest_idx = re_string_cur_idx (&mctx->input) + naccepted; + mctx->max_mb_elem_len = ((mctx->max_mb_elem_len < naccepted) ? naccepted + : mctx->max_mb_elem_len); + err = clean_state_log_if_needed (mctx, dest_idx); + if (BE (err != REG_NOERROR, 0)) + return err; +#ifdef DEBUG + assert (dfa->nexts[cur_node_idx] != REG_MISSING); +#endif + new_nodes = dfa->eclosures + dfa->nexts[cur_node_idx]; + + dest_state = mctx->state_log[dest_idx]; + if (dest_state == NULL) + dest_nodes = *new_nodes; + else + { + err = re_node_set_init_union (&dest_nodes, + dest_state->entrance_nodes, new_nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + } + context = re_string_context_at (&mctx->input, dest_idx - 1, + mctx->eflags); + mctx->state_log[dest_idx] + = re_acquire_state_context (&err, dfa, &dest_nodes, context); + if (dest_state != NULL) + re_node_set_free (&dest_nodes); + if (BE (mctx->state_log[dest_idx] == NULL && err != REG_NOERROR, 0)) + return err; + } + return REG_NOERROR; +} +#endif /* RE_ENABLE_I18N */ + +static reg_errcode_t +internal_function +transit_state_bkref (re_match_context_t *mctx, const re_node_set *nodes) +{ + const re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx i; + Idx cur_str_idx = re_string_cur_idx (&mctx->input); + + for (i = 0; i < nodes->nelem; ++i) + { + Idx dest_str_idx, prev_nelem, bkc_idx; + Idx node_idx = nodes->elems[i]; + unsigned int context; + const re_token_t *node = dfa->nodes + node_idx; + re_node_set *new_dest_nodes; + + /* Check whether 'node' is a backreference or not. */ + if (node->type != OP_BACK_REF) + continue; + + if (node->constraint) + { + context = re_string_context_at (&mctx->input, cur_str_idx, + mctx->eflags); + if (NOT_SATISFY_NEXT_CONSTRAINT (node->constraint, context)) + continue; + } + + /* 'node' is a backreference. + Check the substring which the substring matched. */ + bkc_idx = mctx->nbkref_ents; + err = get_subexp (mctx, node_idx, cur_str_idx); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + + /* And add the epsilon closures (which is 'new_dest_nodes') of + the backreference to appropriate state_log. */ +#ifdef DEBUG + assert (dfa->nexts[node_idx] != REG_MISSING); +#endif + for (; bkc_idx < mctx->nbkref_ents; ++bkc_idx) + { + Idx subexp_len; + re_dfastate_t *dest_state; + struct re_backref_cache_entry *bkref_ent; + bkref_ent = mctx->bkref_ents + bkc_idx; + if (bkref_ent->node != node_idx || bkref_ent->str_idx != cur_str_idx) + continue; + subexp_len = bkref_ent->subexp_to - bkref_ent->subexp_from; + new_dest_nodes = (subexp_len == 0 + ? dfa->eclosures + dfa->edests[node_idx].elems[0] + : dfa->eclosures + dfa->nexts[node_idx]); + dest_str_idx = (cur_str_idx + bkref_ent->subexp_to + - bkref_ent->subexp_from); + context = re_string_context_at (&mctx->input, dest_str_idx - 1, + mctx->eflags); + dest_state = mctx->state_log[dest_str_idx]; + prev_nelem = ((mctx->state_log[cur_str_idx] == NULL) ? 0 + : mctx->state_log[cur_str_idx]->nodes.nelem); + /* Add 'new_dest_node' to state_log. */ + if (dest_state == NULL) + { + mctx->state_log[dest_str_idx] + = re_acquire_state_context (&err, dfa, new_dest_nodes, + context); + if (BE (mctx->state_log[dest_str_idx] == NULL + && err != REG_NOERROR, 0)) + goto free_return; + } + else + { + re_node_set dest_nodes; + err = re_node_set_init_union (&dest_nodes, + dest_state->entrance_nodes, + new_dest_nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&dest_nodes); + goto free_return; + } + mctx->state_log[dest_str_idx] + = re_acquire_state_context (&err, dfa, &dest_nodes, context); + re_node_set_free (&dest_nodes); + if (BE (mctx->state_log[dest_str_idx] == NULL + && err != REG_NOERROR, 0)) + goto free_return; + } + /* We need to check recursively if the backreference can epsilon + transit. */ + if (subexp_len == 0 + && mctx->state_log[cur_str_idx]->nodes.nelem > prev_nelem) + { + err = check_subexp_matching_top (mctx, new_dest_nodes, + cur_str_idx); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + err = transit_state_bkref (mctx, new_dest_nodes); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + } + } + err = REG_NOERROR; + free_return: + return err; +} + +/* Enumerate all the candidates which the backreference BKREF_NODE can match + at BKREF_STR_IDX, and register them by match_ctx_add_entry(). + Note that we might collect inappropriate candidates here. + However, the cost of checking them strictly here is too high, then we + delay these checking for prune_impossible_nodes(). */ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +get_subexp (re_match_context_t *mctx, Idx bkref_node, Idx bkref_str_idx) +{ + const re_dfa_t *const dfa = mctx->dfa; + Idx subexp_num, sub_top_idx; + const char *buf = (const char *) re_string_get_buffer (&mctx->input); + /* Return if we have already checked BKREF_NODE at BKREF_STR_IDX. */ + Idx cache_idx = search_cur_bkref_entry (mctx, bkref_str_idx); + if (cache_idx != REG_MISSING) + { + const struct re_backref_cache_entry *entry + = mctx->bkref_ents + cache_idx; + do + if (entry->node == bkref_node) + return REG_NOERROR; /* We already checked it. */ + while (entry++->more); + } + + subexp_num = dfa->nodes[bkref_node].opr.idx; + + /* For each sub expression */ + for (sub_top_idx = 0; sub_top_idx < mctx->nsub_tops; ++sub_top_idx) + { + reg_errcode_t err; + re_sub_match_top_t *sub_top = mctx->sub_tops[sub_top_idx]; + re_sub_match_last_t *sub_last; + Idx sub_last_idx, sl_str, bkref_str_off; + + if (dfa->nodes[sub_top->node].opr.idx != subexp_num) + continue; /* It isn't related. */ + + sl_str = sub_top->str_idx; + bkref_str_off = bkref_str_idx; + /* At first, check the last node of sub expressions we already + evaluated. */ + for (sub_last_idx = 0; sub_last_idx < sub_top->nlasts; ++sub_last_idx) + { + regoff_t sl_str_diff; + sub_last = sub_top->lasts[sub_last_idx]; + sl_str_diff = sub_last->str_idx - sl_str; + /* The matched string by the sub expression match with the substring + at the back reference? */ + if (sl_str_diff > 0) + { + if (BE (bkref_str_off + sl_str_diff > mctx->input.valid_len, 0)) + { + /* Not enough chars for a successful match. */ + if (bkref_str_off + sl_str_diff > mctx->input.len) + break; + + err = clean_state_log_if_needed (mctx, + bkref_str_off + + sl_str_diff); + if (BE (err != REG_NOERROR, 0)) + return err; + buf = (const char *) re_string_get_buffer (&mctx->input); + } + if (memcmp (buf + bkref_str_off, buf + sl_str, sl_str_diff) != 0) + /* We don't need to search this sub expression any more. */ + break; + } + bkref_str_off += sl_str_diff; + sl_str += sl_str_diff; + err = get_subexp_sub (mctx, sub_top, sub_last, bkref_node, + bkref_str_idx); + + /* Reload buf, since the preceding call might have reallocated + the buffer. */ + buf = (const char *) re_string_get_buffer (&mctx->input); + + if (err == REG_NOMATCH) + continue; + if (BE (err != REG_NOERROR, 0)) + return err; + } + + if (sub_last_idx < sub_top->nlasts) + continue; + if (sub_last_idx > 0) + ++sl_str; + /* Then, search for the other last nodes of the sub expression. */ + for (; sl_str <= bkref_str_idx; ++sl_str) + { + Idx cls_node; + regoff_t sl_str_off; + const re_node_set *nodes; + sl_str_off = sl_str - sub_top->str_idx; + /* The matched string by the sub expression match with the substring + at the back reference? */ + if (sl_str_off > 0) + { + if (BE (bkref_str_off >= mctx->input.valid_len, 0)) + { + /* If we are at the end of the input, we cannot match. */ + if (bkref_str_off >= mctx->input.len) + break; + + err = extend_buffers (mctx); + if (BE (err != REG_NOERROR, 0)) + return err; + + buf = (const char *) re_string_get_buffer (&mctx->input); + } + if (buf [bkref_str_off++] != buf[sl_str - 1]) + break; /* We don't need to search this sub expression + any more. */ + } + if (mctx->state_log[sl_str] == NULL) + continue; + /* Does this state have a ')' of the sub expression? */ + nodes = &mctx->state_log[sl_str]->nodes; + cls_node = find_subexp_node (dfa, nodes, subexp_num, + OP_CLOSE_SUBEXP); + if (cls_node == REG_MISSING) + continue; /* No. */ + if (sub_top->path == NULL) + { + sub_top->path = calloc (sizeof (state_array_t), + sl_str - sub_top->str_idx + 1); + if (sub_top->path == NULL) + return REG_ESPACE; + } + /* Can the OP_OPEN_SUBEXP node arrive the OP_CLOSE_SUBEXP node + in the current context? */ + err = check_arrival (mctx, sub_top->path, sub_top->node, + sub_top->str_idx, cls_node, sl_str, + OP_CLOSE_SUBEXP); + if (err == REG_NOMATCH) + continue; + if (BE (err != REG_NOERROR, 0)) + return err; + sub_last = match_ctx_add_sublast (sub_top, cls_node, sl_str); + if (BE (sub_last == NULL, 0)) + return REG_ESPACE; + err = get_subexp_sub (mctx, sub_top, sub_last, bkref_node, + bkref_str_idx); + if (err == REG_NOMATCH) + continue; + } + } + return REG_NOERROR; +} + +/* Helper functions for get_subexp(). */ + +/* Check SUB_LAST can arrive to the back reference BKREF_NODE at BKREF_STR. + If it can arrive, register the sub expression expressed with SUB_TOP + and SUB_LAST. */ + +static reg_errcode_t +internal_function +get_subexp_sub (re_match_context_t *mctx, const re_sub_match_top_t *sub_top, + re_sub_match_last_t *sub_last, Idx bkref_node, Idx bkref_str) +{ + reg_errcode_t err; + Idx to_idx; + /* Can the subexpression arrive the back reference? */ + err = check_arrival (mctx, &sub_last->path, sub_last->node, + sub_last->str_idx, bkref_node, bkref_str, + OP_OPEN_SUBEXP); + if (err != REG_NOERROR) + return err; + err = match_ctx_add_entry (mctx, bkref_node, bkref_str, sub_top->str_idx, + sub_last->str_idx); + if (BE (err != REG_NOERROR, 0)) + return err; + to_idx = bkref_str + sub_last->str_idx - sub_top->str_idx; + return clean_state_log_if_needed (mctx, to_idx); +} + +/* Find the first node which is '(' or ')' and whose index is SUBEXP_IDX. + Search '(' if FL_OPEN, or search ')' otherwise. + TODO: This function isn't efficient... + Because there might be more than one nodes whose types are + OP_OPEN_SUBEXP and whose index is SUBEXP_IDX, we must check all + nodes. + E.g. RE: (a){2} */ + +static Idx +internal_function +find_subexp_node (const re_dfa_t *dfa, const re_node_set *nodes, + Idx subexp_idx, int type) +{ + Idx cls_idx; + for (cls_idx = 0; cls_idx < nodes->nelem; ++cls_idx) + { + Idx cls_node = nodes->elems[cls_idx]; + const re_token_t *node = dfa->nodes + cls_node; + if (node->type == type + && node->opr.idx == subexp_idx) + return cls_node; + } + return REG_MISSING; +} + +/* Check whether the node TOP_NODE at TOP_STR can arrive to the node + LAST_NODE at LAST_STR. We record the path onto PATH since it will be + heavily reused. + Return REG_NOERROR if it can arrive, or REG_NOMATCH otherwise. */ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +check_arrival (re_match_context_t *mctx, state_array_t *path, Idx top_node, + Idx top_str, Idx last_node, Idx last_str, int type) +{ + const re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err = REG_NOERROR; + Idx subexp_num, backup_cur_idx, str_idx, null_cnt; + re_dfastate_t *cur_state = NULL; + re_node_set *cur_nodes, next_nodes; + re_dfastate_t **backup_state_log; + unsigned int context; + + subexp_num = dfa->nodes[top_node].opr.idx; + /* Extend the buffer if we need. */ + if (BE (path->alloc < last_str + mctx->max_mb_elem_len + 1, 0)) + { + re_dfastate_t **new_array; + Idx old_alloc = path->alloc; + Idx incr_alloc = last_str + mctx->max_mb_elem_len + 1; + Idx new_alloc; + if (BE (IDX_MAX - old_alloc < incr_alloc, 0)) + return REG_ESPACE; + new_alloc = old_alloc + incr_alloc; + if (BE (SIZE_MAX / sizeof (re_dfastate_t *) < new_alloc, 0)) + return REG_ESPACE; + new_array = re_realloc (path->array, re_dfastate_t *, new_alloc); + if (BE (new_array == NULL, 0)) + return REG_ESPACE; + path->array = new_array; + path->alloc = new_alloc; + memset (new_array + old_alloc, '\0', + sizeof (re_dfastate_t *) * (path->alloc - old_alloc)); + } + + str_idx = path->next_idx ? path->next_idx : top_str; + + /* Temporary modify MCTX. */ + backup_state_log = mctx->state_log; + backup_cur_idx = mctx->input.cur_idx; + mctx->state_log = path->array; + mctx->input.cur_idx = str_idx; + + /* Setup initial node set. */ + context = re_string_context_at (&mctx->input, str_idx - 1, mctx->eflags); + if (str_idx == top_str) + { + err = re_node_set_init_1 (&next_nodes, top_node); + if (BE (err != REG_NOERROR, 0)) + return err; + err = check_arrival_expand_ecl (dfa, &next_nodes, subexp_num, type); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + else + { + cur_state = mctx->state_log[str_idx]; + if (cur_state && cur_state->has_backref) + { + err = re_node_set_init_copy (&next_nodes, &cur_state->nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + } + else + re_node_set_init_empty (&next_nodes); + } + if (str_idx == top_str || (cur_state && cur_state->has_backref)) + { + if (next_nodes.nelem) + { + err = expand_bkref_cache (mctx, &next_nodes, str_idx, + subexp_num, type); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + cur_state = re_acquire_state_context (&err, dfa, &next_nodes, context); + if (BE (cur_state == NULL && err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + mctx->state_log[str_idx] = cur_state; + } + + for (null_cnt = 0; str_idx < last_str && null_cnt <= mctx->max_mb_elem_len;) + { + re_node_set_empty (&next_nodes); + if (mctx->state_log[str_idx + 1]) + { + err = re_node_set_merge (&next_nodes, + &mctx->state_log[str_idx + 1]->nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + if (cur_state) + { + err = check_arrival_add_next_nodes (mctx, str_idx, + &cur_state->non_eps_nodes, + &next_nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + ++str_idx; + if (next_nodes.nelem) + { + err = check_arrival_expand_ecl (dfa, &next_nodes, subexp_num, type); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + err = expand_bkref_cache (mctx, &next_nodes, str_idx, + subexp_num, type); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + context = re_string_context_at (&mctx->input, str_idx - 1, mctx->eflags); + cur_state = re_acquire_state_context (&err, dfa, &next_nodes, context); + if (BE (cur_state == NULL && err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + mctx->state_log[str_idx] = cur_state; + null_cnt = cur_state == NULL ? null_cnt + 1 : 0; + } + re_node_set_free (&next_nodes); + cur_nodes = (mctx->state_log[last_str] == NULL ? NULL + : &mctx->state_log[last_str]->nodes); + path->next_idx = str_idx; + + /* Fix MCTX. */ + mctx->state_log = backup_state_log; + mctx->input.cur_idx = backup_cur_idx; + + /* Then check the current node set has the node LAST_NODE. */ + if (cur_nodes != NULL && re_node_set_contains (cur_nodes, last_node)) + return REG_NOERROR; + + return REG_NOMATCH; +} + +/* Helper functions for check_arrival. */ + +/* Calculate the destination nodes of CUR_NODES at STR_IDX, and append them + to NEXT_NODES. + TODO: This function is similar to the functions transit_state*(), + however this function has many additional works. + Can't we unify them? */ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +check_arrival_add_next_nodes (re_match_context_t *mctx, Idx str_idx, + re_node_set *cur_nodes, re_node_set *next_nodes) +{ + const re_dfa_t *const dfa = mctx->dfa; + bool ok; + Idx cur_idx; +#ifdef RE_ENABLE_I18N + reg_errcode_t err = REG_NOERROR; +#endif + re_node_set union_set; + re_node_set_init_empty (&union_set); + for (cur_idx = 0; cur_idx < cur_nodes->nelem; ++cur_idx) + { + int naccepted = 0; + Idx cur_node = cur_nodes->elems[cur_idx]; +#ifdef DEBUG + re_token_type_t type = dfa->nodes[cur_node].type; + assert (!IS_EPSILON_NODE (type)); +#endif +#ifdef RE_ENABLE_I18N + /* If the node may accept "multi byte". */ + if (dfa->nodes[cur_node].accept_mb) + { + naccepted = check_node_accept_bytes (dfa, cur_node, &mctx->input, + str_idx); + if (naccepted > 1) + { + re_dfastate_t *dest_state; + Idx next_node = dfa->nexts[cur_node]; + Idx next_idx = str_idx + naccepted; + dest_state = mctx->state_log[next_idx]; + re_node_set_empty (&union_set); + if (dest_state) + { + err = re_node_set_merge (&union_set, &dest_state->nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&union_set); + return err; + } + } + ok = re_node_set_insert (&union_set, next_node); + if (BE (! ok, 0)) + { + re_node_set_free (&union_set); + return REG_ESPACE; + } + mctx->state_log[next_idx] = re_acquire_state (&err, dfa, + &union_set); + if (BE (mctx->state_log[next_idx] == NULL + && err != REG_NOERROR, 0)) + { + re_node_set_free (&union_set); + return err; + } + } + } +#endif /* RE_ENABLE_I18N */ + if (naccepted + || check_node_accept (mctx, dfa->nodes + cur_node, str_idx)) + { + ok = re_node_set_insert (next_nodes, dfa->nexts[cur_node]); + if (BE (! ok, 0)) + { + re_node_set_free (&union_set); + return REG_ESPACE; + } + } + } + re_node_set_free (&union_set); + return REG_NOERROR; +} + +/* For all the nodes in CUR_NODES, add the epsilon closures of them to + CUR_NODES, however exclude the nodes which are: + - inside the sub expression whose number is EX_SUBEXP, if FL_OPEN. + - out of the sub expression whose number is EX_SUBEXP, if !FL_OPEN. +*/ + +static reg_errcode_t +internal_function +check_arrival_expand_ecl (const re_dfa_t *dfa, re_node_set *cur_nodes, + Idx ex_subexp, int type) +{ + reg_errcode_t err; + Idx idx, outside_node; + re_node_set new_nodes; +#ifdef DEBUG + assert (cur_nodes->nelem); +#endif + err = re_node_set_alloc (&new_nodes, cur_nodes->nelem); + if (BE (err != REG_NOERROR, 0)) + return err; + /* Create a new node set NEW_NODES with the nodes which are epsilon + closures of the node in CUR_NODES. */ + + for (idx = 0; idx < cur_nodes->nelem; ++idx) + { + Idx cur_node = cur_nodes->elems[idx]; + const re_node_set *eclosure = dfa->eclosures + cur_node; + outside_node = find_subexp_node (dfa, eclosure, ex_subexp, type); + if (outside_node == REG_MISSING) + { + /* There are no problematic nodes, just merge them. */ + err = re_node_set_merge (&new_nodes, eclosure); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&new_nodes); + return err; + } + } + else + { + /* There are problematic nodes, re-calculate incrementally. */ + err = check_arrival_expand_ecl_sub (dfa, &new_nodes, cur_node, + ex_subexp, type); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&new_nodes); + return err; + } + } + } + re_node_set_free (cur_nodes); + *cur_nodes = new_nodes; + return REG_NOERROR; +} + +/* Helper function for check_arrival_expand_ecl. + Check incrementally the epsilon closure of TARGET, and if it isn't + problematic append it to DST_NODES. */ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +check_arrival_expand_ecl_sub (const re_dfa_t *dfa, re_node_set *dst_nodes, + Idx target, Idx ex_subexp, int type) +{ + Idx cur_node; + for (cur_node = target; !re_node_set_contains (dst_nodes, cur_node);) + { + bool ok; + + if (dfa->nodes[cur_node].type == type + && dfa->nodes[cur_node].opr.idx == ex_subexp) + { + if (type == OP_CLOSE_SUBEXP) + { + ok = re_node_set_insert (dst_nodes, cur_node); + if (BE (! ok, 0)) + return REG_ESPACE; + } + break; + } + ok = re_node_set_insert (dst_nodes, cur_node); + if (BE (! ok, 0)) + return REG_ESPACE; + if (dfa->edests[cur_node].nelem == 0) + break; + if (dfa->edests[cur_node].nelem == 2) + { + reg_errcode_t err; + err = check_arrival_expand_ecl_sub (dfa, dst_nodes, + dfa->edests[cur_node].elems[1], + ex_subexp, type); + if (BE (err != REG_NOERROR, 0)) + return err; + } + cur_node = dfa->edests[cur_node].elems[0]; + } + return REG_NOERROR; +} + + +/* For all the back references in the current state, calculate the + destination of the back references by the appropriate entry + in MCTX->BKREF_ENTS. */ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +expand_bkref_cache (re_match_context_t *mctx, re_node_set *cur_nodes, + Idx cur_str, Idx subexp_num, int type) +{ + const re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx cache_idx_start = search_cur_bkref_entry (mctx, cur_str); + struct re_backref_cache_entry *ent; + + if (cache_idx_start == REG_MISSING) + return REG_NOERROR; + + restart: + ent = mctx->bkref_ents + cache_idx_start; + do + { + Idx to_idx, next_node; + + /* Is this entry ENT is appropriate? */ + if (!re_node_set_contains (cur_nodes, ent->node)) + continue; /* No. */ + + to_idx = cur_str + ent->subexp_to - ent->subexp_from; + /* Calculate the destination of the back reference, and append it + to MCTX->STATE_LOG. */ + if (to_idx == cur_str) + { + /* The backreference did epsilon transit, we must re-check all the + node in the current state. */ + re_node_set new_dests; + reg_errcode_t err2, err3; + next_node = dfa->edests[ent->node].elems[0]; + if (re_node_set_contains (cur_nodes, next_node)) + continue; + err = re_node_set_init_1 (&new_dests, next_node); + err2 = check_arrival_expand_ecl (dfa, &new_dests, subexp_num, type); + err3 = re_node_set_merge (cur_nodes, &new_dests); + re_node_set_free (&new_dests); + if (BE (err != REG_NOERROR || err2 != REG_NOERROR + || err3 != REG_NOERROR, 0)) + { + err = (err != REG_NOERROR ? err + : (err2 != REG_NOERROR ? err2 : err3)); + return err; + } + /* TODO: It is still inefficient... */ + goto restart; + } + else + { + re_node_set union_set; + next_node = dfa->nexts[ent->node]; + if (mctx->state_log[to_idx]) + { + bool ok; + if (re_node_set_contains (&mctx->state_log[to_idx]->nodes, + next_node)) + continue; + err = re_node_set_init_copy (&union_set, + &mctx->state_log[to_idx]->nodes); + ok = re_node_set_insert (&union_set, next_node); + if (BE (err != REG_NOERROR || ! ok, 0)) + { + re_node_set_free (&union_set); + err = err != REG_NOERROR ? err : REG_ESPACE; + return err; + } + } + else + { + err = re_node_set_init_1 (&union_set, next_node); + if (BE (err != REG_NOERROR, 0)) + return err; + } + mctx->state_log[to_idx] = re_acquire_state (&err, dfa, &union_set); + re_node_set_free (&union_set); + if (BE (mctx->state_log[to_idx] == NULL + && err != REG_NOERROR, 0)) + return err; + } + } + while (ent++->more); + return REG_NOERROR; +} + +/* Build transition table for the state. + Return true if successful. */ + +static bool +internal_function +build_trtable (const re_dfa_t *dfa, re_dfastate_t *state) +{ + reg_errcode_t err; + Idx i, j; + int ch; + bool need_word_trtable = false; + bitset_word_t elem, mask; + bool dests_node_malloced = false; + bool dest_states_malloced = false; + Idx ndests; /* Number of the destination states from 'state'. */ + re_dfastate_t **trtable; + re_dfastate_t **dest_states = NULL, **dest_states_word, **dest_states_nl; + re_node_set follows, *dests_node; + bitset_t *dests_ch; + bitset_t acceptable; + + struct dests_alloc + { + re_node_set dests_node[SBC_MAX]; + bitset_t dests_ch[SBC_MAX]; + } *dests_alloc; + + /* We build DFA states which corresponds to the destination nodes + from 'state'. 'dests_node[i]' represents the nodes which i-th + destination state contains, and 'dests_ch[i]' represents the + characters which i-th destination state accepts. */ + if (__libc_use_alloca (sizeof (struct dests_alloc))) + dests_alloc = (struct dests_alloc *) alloca (sizeof (struct dests_alloc)); + else + { + dests_alloc = re_malloc (struct dests_alloc, 1); + if (BE (dests_alloc == NULL, 0)) + return false; + dests_node_malloced = true; + } + dests_node = dests_alloc->dests_node; + dests_ch = dests_alloc->dests_ch; + + /* Initialize transition table. */ + state->word_trtable = state->trtable = NULL; + + /* At first, group all nodes belonging to 'state' into several + destinations. */ + ndests = group_nodes_into_DFAstates (dfa, state, dests_node, dests_ch); + if (BE (! REG_VALID_NONZERO_INDEX (ndests), 0)) + { + if (dests_node_malloced) + free (dests_alloc); + /* Return false in case of an error, true otherwise. */ + if (ndests == 0) + { + state->trtable = (re_dfastate_t **) + calloc (sizeof (re_dfastate_t *), SBC_MAX); + if (BE (state->trtable == NULL, 0)) + return false; + return true; + } + return false; + } + + err = re_node_set_alloc (&follows, ndests + 1); + if (BE (err != REG_NOERROR, 0)) + goto out_free; + + /* Avoid arithmetic overflow in size calculation. */ + if (BE ((((SIZE_MAX - (sizeof (re_node_set) + sizeof (bitset_t)) * SBC_MAX) + / (3 * sizeof (re_dfastate_t *))) + < ndests), + 0)) + goto out_free; + + if (__libc_use_alloca ((sizeof (re_node_set) + sizeof (bitset_t)) * SBC_MAX + + ndests * 3 * sizeof (re_dfastate_t *))) + dest_states = (re_dfastate_t **) + alloca (ndests * 3 * sizeof (re_dfastate_t *)); + else + { + dest_states = (re_dfastate_t **) + malloc (ndests * 3 * sizeof (re_dfastate_t *)); + if (BE (dest_states == NULL, 0)) + { +out_free: + if (dest_states_malloced) + free (dest_states); + re_node_set_free (&follows); + for (i = 0; i < ndests; ++i) + re_node_set_free (dests_node + i); + if (dests_node_malloced) + free (dests_alloc); + return false; + } + dest_states_malloced = true; + } + dest_states_word = dest_states + ndests; + dest_states_nl = dest_states_word + ndests; + bitset_empty (acceptable); + + /* Then build the states for all destinations. */ + for (i = 0; i < ndests; ++i) + { + Idx next_node; + re_node_set_empty (&follows); + /* Merge the follows of this destination states. */ + for (j = 0; j < dests_node[i].nelem; ++j) + { + next_node = dfa->nexts[dests_node[i].elems[j]]; + if (next_node != REG_MISSING) + { + err = re_node_set_merge (&follows, dfa->eclosures + next_node); + if (BE (err != REG_NOERROR, 0)) + goto out_free; + } + } + dest_states[i] = re_acquire_state_context (&err, dfa, &follows, 0); + if (BE (dest_states[i] == NULL && err != REG_NOERROR, 0)) + goto out_free; + /* If the new state has context constraint, + build appropriate states for these contexts. */ + if (dest_states[i]->has_constraint) + { + dest_states_word[i] = re_acquire_state_context (&err, dfa, &follows, + CONTEXT_WORD); + if (BE (dest_states_word[i] == NULL && err != REG_NOERROR, 0)) + goto out_free; + + if (dest_states[i] != dest_states_word[i] && dfa->mb_cur_max > 1) + need_word_trtable = true; + + dest_states_nl[i] = re_acquire_state_context (&err, dfa, &follows, + CONTEXT_NEWLINE); + if (BE (dest_states_nl[i] == NULL && err != REG_NOERROR, 0)) + goto out_free; + } + else + { + dest_states_word[i] = dest_states[i]; + dest_states_nl[i] = dest_states[i]; + } + bitset_merge (acceptable, dests_ch[i]); + } + + if (!BE (need_word_trtable, 0)) + { + /* We don't care about whether the following character is a word + character, or we are in a single-byte character set so we can + discern by looking at the character code: allocate a + 256-entry transition table. */ + trtable = state->trtable = + (re_dfastate_t **) calloc (sizeof (re_dfastate_t *), SBC_MAX); + if (BE (trtable == NULL, 0)) + goto out_free; + + /* For all characters ch...: */ + for (i = 0; i < BITSET_WORDS; ++i) + for (ch = i * BITSET_WORD_BITS, elem = acceptable[i], mask = 1; + elem; + mask <<= 1, elem >>= 1, ++ch) + if (BE (elem & 1, 0)) + { + /* There must be exactly one destination which accepts + character ch. See group_nodes_into_DFAstates. */ + for (j = 0; (dests_ch[j][i] & mask) == 0; ++j) + ; + + /* j-th destination accepts the word character ch. */ + if (dfa->word_char[i] & mask) + trtable[ch] = dest_states_word[j]; + else + trtable[ch] = dest_states[j]; + } + } + else + { + /* We care about whether the following character is a word + character, and we are in a multi-byte character set: discern + by looking at the character code: build two 256-entry + transition tables, one starting at trtable[0] and one + starting at trtable[SBC_MAX]. */ + trtable = state->word_trtable = + (re_dfastate_t **) calloc (sizeof (re_dfastate_t *), 2 * SBC_MAX); + if (BE (trtable == NULL, 0)) + goto out_free; + + /* For all characters ch...: */ + for (i = 0; i < BITSET_WORDS; ++i) + for (ch = i * BITSET_WORD_BITS, elem = acceptable[i], mask = 1; + elem; + mask <<= 1, elem >>= 1, ++ch) + if (BE (elem & 1, 0)) + { + /* There must be exactly one destination which accepts + character ch. See group_nodes_into_DFAstates. */ + for (j = 0; (dests_ch[j][i] & mask) == 0; ++j) + ; + + /* j-th destination accepts the word character ch. */ + trtable[ch] = dest_states[j]; + trtable[ch + SBC_MAX] = dest_states_word[j]; + } + } + + /* new line */ + if (bitset_contain (acceptable, NEWLINE_CHAR)) + { + /* The current state accepts newline character. */ + for (j = 0; j < ndests; ++j) + if (bitset_contain (dests_ch[j], NEWLINE_CHAR)) + { + /* k-th destination accepts newline character. */ + trtable[NEWLINE_CHAR] = dest_states_nl[j]; + if (need_word_trtable) + trtable[NEWLINE_CHAR + SBC_MAX] = dest_states_nl[j]; + /* There must be only one destination which accepts + newline. See group_nodes_into_DFAstates. */ + break; + } + } + + if (dest_states_malloced) + free (dest_states); + + re_node_set_free (&follows); + for (i = 0; i < ndests; ++i) + re_node_set_free (dests_node + i); + + if (dests_node_malloced) + free (dests_alloc); + + return true; +} + +/* Group all nodes belonging to STATE into several destinations. + Then for all destinations, set the nodes belonging to the destination + to DESTS_NODE[i] and set the characters accepted by the destination + to DEST_CH[i]. This function return the number of destinations. */ + +static Idx +internal_function +group_nodes_into_DFAstates (const re_dfa_t *dfa, const re_dfastate_t *state, + re_node_set *dests_node, bitset_t *dests_ch) +{ + reg_errcode_t err; + bool ok; + Idx i, j, k; + Idx ndests; /* Number of the destinations from 'state'. */ + bitset_t accepts; /* Characters a node can accept. */ + const re_node_set *cur_nodes = &state->nodes; + bitset_empty (accepts); + ndests = 0; + + /* For all the nodes belonging to 'state', */ + for (i = 0; i < cur_nodes->nelem; ++i) + { + re_token_t *node = &dfa->nodes[cur_nodes->elems[i]]; + re_token_type_t type = node->type; + unsigned int constraint = node->constraint; + + /* Enumerate all single byte character this node can accept. */ + if (type == CHARACTER) + bitset_set (accepts, node->opr.c); + else if (type == SIMPLE_BRACKET) + { + bitset_merge (accepts, node->opr.sbcset); + } + else if (type == OP_PERIOD) + { +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + bitset_merge (accepts, dfa->sb_char); + else +#endif + bitset_set_all (accepts); + if (!(dfa->syntax & RE_DOT_NEWLINE)) + bitset_clear (accepts, '\n'); + if (dfa->syntax & RE_DOT_NOT_NULL) + bitset_clear (accepts, '\0'); + } +#ifdef RE_ENABLE_I18N + else if (type == OP_UTF8_PERIOD) + { + if (ASCII_CHARS % BITSET_WORD_BITS == 0) + memset (accepts, -1, ASCII_CHARS / CHAR_BIT); + else + bitset_merge (accepts, utf8_sb_map); + if (!(dfa->syntax & RE_DOT_NEWLINE)) + bitset_clear (accepts, '\n'); + if (dfa->syntax & RE_DOT_NOT_NULL) + bitset_clear (accepts, '\0'); + } +#endif + else + continue; + + /* Check the 'accepts' and sift the characters which are not + match it the context. */ + if (constraint) + { + if (constraint & NEXT_NEWLINE_CONSTRAINT) + { + bool accepts_newline = bitset_contain (accepts, NEWLINE_CHAR); + bitset_empty (accepts); + if (accepts_newline) + bitset_set (accepts, NEWLINE_CHAR); + else + continue; + } + if (constraint & NEXT_ENDBUF_CONSTRAINT) + { + bitset_empty (accepts); + continue; + } + + if (constraint & NEXT_WORD_CONSTRAINT) + { + bitset_word_t any_set = 0; + if (type == CHARACTER && !node->word_char) + { + bitset_empty (accepts); + continue; + } +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + for (j = 0; j < BITSET_WORDS; ++j) + any_set |= (accepts[j] &= (dfa->word_char[j] | ~dfa->sb_char[j])); + else +#endif + for (j = 0; j < BITSET_WORDS; ++j) + any_set |= (accepts[j] &= dfa->word_char[j]); + if (!any_set) + continue; + } + if (constraint & NEXT_NOTWORD_CONSTRAINT) + { + bitset_word_t any_set = 0; + if (type == CHARACTER && node->word_char) + { + bitset_empty (accepts); + continue; + } +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + for (j = 0; j < BITSET_WORDS; ++j) + any_set |= (accepts[j] &= ~(dfa->word_char[j] & dfa->sb_char[j])); + else +#endif + for (j = 0; j < BITSET_WORDS; ++j) + any_set |= (accepts[j] &= ~dfa->word_char[j]); + if (!any_set) + continue; + } + } + + /* Then divide 'accepts' into DFA states, or create a new + state. Above, we make sure that accepts is not empty. */ + for (j = 0; j < ndests; ++j) + { + bitset_t intersec; /* Intersection sets, see below. */ + bitset_t remains; + /* Flags, see below. */ + bitset_word_t has_intersec, not_subset, not_consumed; + + /* Optimization, skip if this state doesn't accept the character. */ + if (type == CHARACTER && !bitset_contain (dests_ch[j], node->opr.c)) + continue; + + /* Enumerate the intersection set of this state and 'accepts'. */ + has_intersec = 0; + for (k = 0; k < BITSET_WORDS; ++k) + has_intersec |= intersec[k] = accepts[k] & dests_ch[j][k]; + /* And skip if the intersection set is empty. */ + if (!has_intersec) + continue; + + /* Then check if this state is a subset of 'accepts'. */ + not_subset = not_consumed = 0; + for (k = 0; k < BITSET_WORDS; ++k) + { + not_subset |= remains[k] = ~accepts[k] & dests_ch[j][k]; + not_consumed |= accepts[k] = accepts[k] & ~dests_ch[j][k]; + } + + /* If this state isn't a subset of 'accepts', create a + new group state, which has the 'remains'. */ + if (not_subset) + { + bitset_copy (dests_ch[ndests], remains); + bitset_copy (dests_ch[j], intersec); + err = re_node_set_init_copy (dests_node + ndests, &dests_node[j]); + if (BE (err != REG_NOERROR, 0)) + goto error_return; + ++ndests; + } + + /* Put the position in the current group. */ + ok = re_node_set_insert (&dests_node[j], cur_nodes->elems[i]); + if (BE (! ok, 0)) + goto error_return; + + /* If all characters are consumed, go to next node. */ + if (!not_consumed) + break; + } + /* Some characters remain, create a new group. */ + if (j == ndests) + { + bitset_copy (dests_ch[ndests], accepts); + err = re_node_set_init_1 (dests_node + ndests, cur_nodes->elems[i]); + if (BE (err != REG_NOERROR, 0)) + goto error_return; + ++ndests; + bitset_empty (accepts); + } + } + return ndests; + error_return: + for (j = 0; j < ndests; ++j) + re_node_set_free (dests_node + j); + return REG_MISSING; +} + +#ifdef RE_ENABLE_I18N +/* Check how many bytes the node 'dfa->nodes[node_idx]' accepts. + Return the number of the bytes the node accepts. + STR_IDX is the current index of the input string. + + This function handles the nodes which can accept one character, or + one collating element like '.', '[a-z]', opposite to the other nodes + can only accept one byte. */ + +static int +internal_function +check_node_accept_bytes (const re_dfa_t *dfa, Idx node_idx, + const re_string_t *input, Idx str_idx) +{ + const re_token_t *node = dfa->nodes + node_idx; + int char_len, elem_len; + Idx i; + + if (BE (node->type == OP_UTF8_PERIOD, 0)) + { + unsigned char c = re_string_byte_at (input, str_idx), d; + if (BE (c < 0xc2, 1)) + return 0; + + if (str_idx + 2 > input->len) + return 0; + + d = re_string_byte_at (input, str_idx + 1); + if (c < 0xe0) + return (d < 0x80 || d > 0xbf) ? 0 : 2; + else if (c < 0xf0) + { + char_len = 3; + if (c == 0xe0 && d < 0xa0) + return 0; + } + else if (c < 0xf8) + { + char_len = 4; + if (c == 0xf0 && d < 0x90) + return 0; + } + else if (c < 0xfc) + { + char_len = 5; + if (c == 0xf8 && d < 0x88) + return 0; + } + else if (c < 0xfe) + { + char_len = 6; + if (c == 0xfc && d < 0x84) + return 0; + } + else + return 0; + + if (str_idx + char_len > input->len) + return 0; + + for (i = 1; i < char_len; ++i) + { + d = re_string_byte_at (input, str_idx + i); + if (d < 0x80 || d > 0xbf) + return 0; + } + return char_len; + } + + char_len = re_string_char_size_at (input, str_idx); + if (node->type == OP_PERIOD) + { + if (char_len <= 1) + return 0; + /* FIXME: I don't think this if is needed, as both '\n' + and '\0' are char_len == 1. */ + /* '.' accepts any one character except the following two cases. */ + if ((!(dfa->syntax & RE_DOT_NEWLINE) && + re_string_byte_at (input, str_idx) == '\n') || + ((dfa->syntax & RE_DOT_NOT_NULL) && + re_string_byte_at (input, str_idx) == '\0')) + return 0; + return char_len; + } + + elem_len = re_string_elem_size_at (input, str_idx); + if ((elem_len <= 1 && char_len <= 1) || char_len == 0) + return 0; + + if (node->type == COMPLEX_BRACKET) + { + const re_charset_t *cset = node->opr.mbcset; +# ifdef _LIBC + const unsigned char *pin + = ((const unsigned char *) re_string_get_buffer (input) + str_idx); + Idx j; + uint32_t nrules; +# endif /* _LIBC */ + int match_len = 0; + wchar_t wc = ((cset->nranges || cset->nchar_classes || cset->nmbchars) + ? re_string_wchar_at (input, str_idx) : 0); + + /* match with multibyte character? */ + for (i = 0; i < cset->nmbchars; ++i) + if (wc == cset->mbchars[i]) + { + match_len = char_len; + goto check_node_accept_bytes_match; + } + /* match with character_class? */ + for (i = 0; i < cset->nchar_classes; ++i) + { + wctype_t wt = cset->char_classes[i]; + if (__iswctype (wc, wt)) + { + match_len = char_len; + goto check_node_accept_bytes_match; + } + } + +# ifdef _LIBC + nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + if (nrules != 0) + { + unsigned int in_collseq = 0; + const int32_t *table, *indirect; + const unsigned char *weights, *extra; + const char *collseqwc; + /* This #include defines a local function! */ +# include <locale/weight.h> + + /* match with collating_symbol? */ + if (cset->ncoll_syms) + extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_SYMB_EXTRAMB); + for (i = 0; i < cset->ncoll_syms; ++i) + { + const unsigned char *coll_sym = extra + cset->coll_syms[i]; + /* Compare the length of input collating element and + the length of current collating element. */ + if (*coll_sym != elem_len) + continue; + /* Compare each bytes. */ + for (j = 0; j < *coll_sym; j++) + if (pin[j] != coll_sym[1 + j]) + break; + if (j == *coll_sym) + { + /* Match if every bytes is equal. */ + match_len = j; + goto check_node_accept_bytes_match; + } + } + + if (cset->nranges) + { + if (elem_len <= char_len) + { + collseqwc = _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQWC); + in_collseq = __collseq_table_lookup (collseqwc, wc); + } + else + in_collseq = find_collation_sequence_value (pin, elem_len); + } + /* match with range expression? */ + for (i = 0; i < cset->nranges; ++i) + if (cset->range_starts[i] <= in_collseq + && in_collseq <= cset->range_ends[i]) + { + match_len = elem_len; + goto check_node_accept_bytes_match; + } + + /* match with equivalence_class? */ + if (cset->nequiv_classes) + { + const unsigned char *cp = pin; + table = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB); + weights = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_WEIGHTMB); + extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB); + indirect = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_INDIRECTMB); + int32_t idx = findidx (&cp, elem_len); + if (idx > 0) + for (i = 0; i < cset->nequiv_classes; ++i) + { + int32_t equiv_class_idx = cset->equiv_classes[i]; + size_t weight_len = weights[idx & 0xffffff]; + if (weight_len == weights[equiv_class_idx & 0xffffff] + && (idx >> 24) == (equiv_class_idx >> 24)) + { + Idx cnt = 0; + + idx &= 0xffffff; + equiv_class_idx &= 0xffffff; + + while (cnt <= weight_len + && (weights[equiv_class_idx + 1 + cnt] + == weights[idx + 1 + cnt])) + ++cnt; + if (cnt > weight_len) + { + match_len = elem_len; + goto check_node_accept_bytes_match; + } + } + } + } + } + else +# endif /* _LIBC */ + { + /* match with range expression? */ +#if __GNUC__ >= 2 && ! (__STDC_VERSION__ < 199901L && defined __STRICT_ANSI__) + wchar_t cmp_buf[] = {L'\0', L'\0', wc, L'\0', L'\0', L'\0'}; +#else + wchar_t cmp_buf[] = {L'\0', L'\0', L'\0', L'\0', L'\0', L'\0'}; + cmp_buf[2] = wc; +#endif + for (i = 0; i < cset->nranges; ++i) + { + cmp_buf[0] = cset->range_starts[i]; + cmp_buf[4] = cset->range_ends[i]; + if (wcscoll (cmp_buf, cmp_buf + 2) <= 0 + && wcscoll (cmp_buf + 2, cmp_buf + 4) <= 0) + { + match_len = char_len; + goto check_node_accept_bytes_match; + } + } + } + check_node_accept_bytes_match: + if (!cset->non_match) + return match_len; + else + { + if (match_len > 0) + return 0; + else + return (elem_len > char_len) ? elem_len : char_len; + } + } + return 0; +} + +# ifdef _LIBC +static unsigned int +internal_function +find_collation_sequence_value (const unsigned char *mbs, size_t mbs_len) +{ + uint32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + if (nrules == 0) + { + if (mbs_len == 1) + { + /* No valid character. Match it as a single byte character. */ + const unsigned char *collseq = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQMB); + return collseq[mbs[0]]; + } + return UINT_MAX; + } + else + { + int32_t idx; + const unsigned char *extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_SYMB_EXTRAMB); + int32_t extrasize = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_SYMB_EXTRAMB + 1) - extra; + + for (idx = 0; idx < extrasize;) + { + int mbs_cnt; + bool found = false; + int32_t elem_mbs_len; + /* Skip the name of collating element name. */ + idx = idx + extra[idx] + 1; + elem_mbs_len = extra[idx++]; + if (mbs_len == elem_mbs_len) + { + for (mbs_cnt = 0; mbs_cnt < elem_mbs_len; ++mbs_cnt) + if (extra[idx + mbs_cnt] != mbs[mbs_cnt]) + break; + if (mbs_cnt == elem_mbs_len) + /* Found the entry. */ + found = true; + } + /* Skip the byte sequence of the collating element. */ + idx += elem_mbs_len; + /* Adjust for the alignment. */ + idx = (idx + 3) & ~3; + /* Skip the collation sequence value. */ + idx += sizeof (uint32_t); + /* Skip the wide char sequence of the collating element. */ + idx = idx + sizeof (uint32_t) * (*(int32_t *) (extra + idx) + 1); + /* If we found the entry, return the sequence value. */ + if (found) + return *(uint32_t *) (extra + idx); + /* Skip the collation sequence value. */ + idx += sizeof (uint32_t); + } + return UINT_MAX; + } +} +# endif /* _LIBC */ +#endif /* RE_ENABLE_I18N */ + +/* Check whether the node accepts the byte which is IDX-th + byte of the INPUT. */ + +static bool +internal_function +check_node_accept (const re_match_context_t *mctx, const re_token_t *node, + Idx idx) +{ + unsigned char ch; + ch = re_string_byte_at (&mctx->input, idx); + switch (node->type) + { + case CHARACTER: + if (node->opr.c != ch) + return false; + break; + + case SIMPLE_BRACKET: + if (!bitset_contain (node->opr.sbcset, ch)) + return false; + break; + +#ifdef RE_ENABLE_I18N + case OP_UTF8_PERIOD: + if (ch >= ASCII_CHARS) + return false; + /* FALLTHROUGH */ +#endif + case OP_PERIOD: + if ((ch == '\n' && !(mctx->dfa->syntax & RE_DOT_NEWLINE)) + || (ch == '\0' && (mctx->dfa->syntax & RE_DOT_NOT_NULL))) + return false; + break; + + default: + return false; + } + + if (node->constraint) + { + /* The node has constraints. Check whether the current context + satisfies the constraints. */ + unsigned int context = re_string_context_at (&mctx->input, idx, + mctx->eflags); + if (NOT_SATISFY_NEXT_CONSTRAINT (node->constraint, context)) + return false; + } + + return true; +} + +/* Extend the buffers, if the buffers have run out. */ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +extend_buffers (re_match_context_t *mctx) +{ + reg_errcode_t ret; + re_string_t *pstr = &mctx->input; + + /* Avoid overflow. */ + if (BE (MIN (IDX_MAX, SIZE_MAX / sizeof (re_dfastate_t *)) / 2 + <= pstr->bufs_len, 0)) + return REG_ESPACE; + + /* Double the lengths of the buffers. */ + ret = re_string_realloc_buffers (pstr, MIN (pstr->len, pstr->bufs_len * 2)); + if (BE (ret != REG_NOERROR, 0)) + return ret; + + if (mctx->state_log != NULL) + { + /* And double the length of state_log. */ + /* XXX We have no indication of the size of this buffer. If this + allocation fail we have no indication that the state_log array + does not have the right size. */ + re_dfastate_t **new_array = re_realloc (mctx->state_log, re_dfastate_t *, + pstr->bufs_len + 1); + if (BE (new_array == NULL, 0)) + return REG_ESPACE; + mctx->state_log = new_array; + } + + /* Then reconstruct the buffers. */ + if (pstr->icase) + { +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1) + { + ret = build_wcs_upper_buffer (pstr); + if (BE (ret != REG_NOERROR, 0)) + return ret; + } + else +#endif /* RE_ENABLE_I18N */ + build_upper_buffer (pstr); + } + else + { +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1) + build_wcs_buffer (pstr); + else +#endif /* RE_ENABLE_I18N */ + { + if (pstr->trans != NULL) + re_string_translate_buffer (pstr); + } + } + return REG_NOERROR; +} + + +/* Functions for matching context. */ + +/* Initialize MCTX. */ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +match_ctx_init (re_match_context_t *mctx, int eflags, Idx n) +{ + mctx->eflags = eflags; + mctx->match_last = REG_MISSING; + if (n > 0) + { + /* Avoid overflow. */ + size_t max_object_size = + MAX (sizeof (struct re_backref_cache_entry), + sizeof (re_sub_match_top_t *)); + if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) < n, 0)) + return REG_ESPACE; + + mctx->bkref_ents = re_malloc (struct re_backref_cache_entry, n); + mctx->sub_tops = re_malloc (re_sub_match_top_t *, n); + if (BE (mctx->bkref_ents == NULL || mctx->sub_tops == NULL, 0)) + return REG_ESPACE; + } + /* Already zero-ed by the caller. + else + mctx->bkref_ents = NULL; + mctx->nbkref_ents = 0; + mctx->nsub_tops = 0; */ + mctx->abkref_ents = n; + mctx->max_mb_elem_len = 1; + mctx->asub_tops = n; + return REG_NOERROR; +} + +/* Clean the entries which depend on the current input in MCTX. + This function must be invoked when the matcher changes the start index + of the input, or changes the input string. */ + +static void +internal_function +match_ctx_clean (re_match_context_t *mctx) +{ + Idx st_idx; + for (st_idx = 0; st_idx < mctx->nsub_tops; ++st_idx) + { + Idx sl_idx; + re_sub_match_top_t *top = mctx->sub_tops[st_idx]; + for (sl_idx = 0; sl_idx < top->nlasts; ++sl_idx) + { + re_sub_match_last_t *last = top->lasts[sl_idx]; + re_free (last->path.array); + re_free (last); + } + re_free (top->lasts); + if (top->path) + { + re_free (top->path->array); + re_free (top->path); + } + free (top); + } + + mctx->nsub_tops = 0; + mctx->nbkref_ents = 0; +} + +/* Free all the memory associated with MCTX. */ + +static void +internal_function +match_ctx_free (re_match_context_t *mctx) +{ + /* First, free all the memory associated with MCTX->SUB_TOPS. */ + match_ctx_clean (mctx); + re_free (mctx->sub_tops); + re_free (mctx->bkref_ents); +} + +/* Add a new backreference entry to MCTX. + Note that we assume that caller never call this function with duplicate + entry, and call with STR_IDX which isn't smaller than any existing entry. +*/ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +match_ctx_add_entry (re_match_context_t *mctx, Idx node, Idx str_idx, Idx from, + Idx to) +{ + if (mctx->nbkref_ents >= mctx->abkref_ents) + { + struct re_backref_cache_entry* new_entry; + new_entry = re_realloc (mctx->bkref_ents, struct re_backref_cache_entry, + mctx->abkref_ents * 2); + if (BE (new_entry == NULL, 0)) + { + re_free (mctx->bkref_ents); + return REG_ESPACE; + } + mctx->bkref_ents = new_entry; + memset (mctx->bkref_ents + mctx->nbkref_ents, '\0', + sizeof (struct re_backref_cache_entry) * mctx->abkref_ents); + mctx->abkref_ents *= 2; + } + if (mctx->nbkref_ents > 0 + && mctx->bkref_ents[mctx->nbkref_ents - 1].str_idx == str_idx) + mctx->bkref_ents[mctx->nbkref_ents - 1].more = 1; + + mctx->bkref_ents[mctx->nbkref_ents].node = node; + mctx->bkref_ents[mctx->nbkref_ents].str_idx = str_idx; + mctx->bkref_ents[mctx->nbkref_ents].subexp_from = from; + mctx->bkref_ents[mctx->nbkref_ents].subexp_to = to; + + /* This is a cache that saves negative results of check_dst_limits_calc_pos. + If bit N is clear, means that this entry won't epsilon-transition to + an OP_OPEN_SUBEXP or OP_CLOSE_SUBEXP for the N+1-th subexpression. If + it is set, check_dst_limits_calc_pos_1 will recurse and try to find one + such node. + + A backreference does not epsilon-transition unless it is empty, so set + to all zeros if FROM != TO. */ + mctx->bkref_ents[mctx->nbkref_ents].eps_reachable_subexps_map + = (from == to ? -1 : 0); + + mctx->bkref_ents[mctx->nbkref_ents++].more = 0; + if (mctx->max_mb_elem_len < to - from) + mctx->max_mb_elem_len = to - from; + return REG_NOERROR; +} + +/* Return the first entry with the same str_idx, or REG_MISSING if none is + found. Note that MCTX->BKREF_ENTS is already sorted by MCTX->STR_IDX. */ + +static Idx +internal_function +search_cur_bkref_entry (const re_match_context_t *mctx, Idx str_idx) +{ + Idx left, right, mid, last; + last = right = mctx->nbkref_ents; + for (left = 0; left < right;) + { + mid = (left + right) / 2; + if (mctx->bkref_ents[mid].str_idx < str_idx) + left = mid + 1; + else + right = mid; + } + if (left < last && mctx->bkref_ents[left].str_idx == str_idx) + return left; + else + return REG_MISSING; +} + +/* Register the node NODE, whose type is OP_OPEN_SUBEXP, and which matches + at STR_IDX. */ + +static reg_errcode_t +internal_function __attribute_warn_unused_result__ +match_ctx_add_subtop (re_match_context_t *mctx, Idx node, Idx str_idx) +{ +#ifdef DEBUG + assert (mctx->sub_tops != NULL); + assert (mctx->asub_tops > 0); +#endif + if (BE (mctx->nsub_tops == mctx->asub_tops, 0)) + { + Idx new_asub_tops = mctx->asub_tops * 2; + re_sub_match_top_t **new_array = re_realloc (mctx->sub_tops, + re_sub_match_top_t *, + new_asub_tops); + if (BE (new_array == NULL, 0)) + return REG_ESPACE; + mctx->sub_tops = new_array; + mctx->asub_tops = new_asub_tops; + } + mctx->sub_tops[mctx->nsub_tops] = calloc (1, sizeof (re_sub_match_top_t)); + if (BE (mctx->sub_tops[mctx->nsub_tops] == NULL, 0)) + return REG_ESPACE; + mctx->sub_tops[mctx->nsub_tops]->node = node; + mctx->sub_tops[mctx->nsub_tops++]->str_idx = str_idx; + return REG_NOERROR; +} + +/* Register the node NODE, whose type is OP_CLOSE_SUBEXP, and which matches + at STR_IDX, whose corresponding OP_OPEN_SUBEXP is SUB_TOP. */ + +static re_sub_match_last_t * +internal_function +match_ctx_add_sublast (re_sub_match_top_t *subtop, Idx node, Idx str_idx) +{ + re_sub_match_last_t *new_entry; + if (BE (subtop->nlasts == subtop->alasts, 0)) + { + Idx new_alasts = 2 * subtop->alasts + 1; + re_sub_match_last_t **new_array = re_realloc (subtop->lasts, + re_sub_match_last_t *, + new_alasts); + if (BE (new_array == NULL, 0)) + return NULL; + subtop->lasts = new_array; + subtop->alasts = new_alasts; + } + new_entry = calloc (1, sizeof (re_sub_match_last_t)); + if (BE (new_entry != NULL, 1)) + { + subtop->lasts[subtop->nlasts] = new_entry; + new_entry->node = node; + new_entry->str_idx = str_idx; + ++subtop->nlasts; + } + return new_entry; +} + +static void +internal_function +sift_ctx_init (re_sift_context_t *sctx, re_dfastate_t **sifted_sts, + re_dfastate_t **limited_sts, Idx last_node, Idx last_str_idx) +{ + sctx->sifted_states = sifted_sts; + sctx->limited_states = limited_sts; + sctx->last_node = last_node; + sctx->last_str_idx = last_str_idx; + re_node_set_init_empty (&sctx->limits); +} diff --git a/lib/stdbool.in.h b/lib/stdbool.in.h new file mode 100644 index 0000000..ed1f9aa --- /dev/null +++ b/lib/stdbool.in.h @@ -0,0 +1,121 @@ +/* Copyright (C) 2001-2003, 2006-2012 Free Software Foundation, Inc. + Written by Bruno Haible <haible@clisp.cons.org>, 2001. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#ifndef _GL_STDBOOL_H +#define _GL_STDBOOL_H + +/* ISO C 99 <stdbool.h> for platforms that lack it. */ + +/* Usage suggestions: + + Programs that use <stdbool.h> should be aware of some limitations + and standards compliance issues. + + Standards compliance: + + - <stdbool.h> must be #included before 'bool', 'false', 'true' + can be used. + + - You cannot assume that sizeof (bool) == 1. + + - Programs should not undefine the macros bool, true, and false, + as C99 lists that as an "obsolescent feature". + + Limitations of this substitute, when used in a C89 environment: + + - <stdbool.h> must be #included before the '_Bool' type can be used. + + - You cannot assume that _Bool is a typedef; it might be a macro. + + - Bit-fields of type 'bool' are not supported. Portable code + should use 'unsigned int foo : 1;' rather than 'bool foo : 1;'. + + - In C99, casts and automatic conversions to '_Bool' or 'bool' are + performed in such a way that every nonzero value gets converted + to 'true', and zero gets converted to 'false'. This doesn't work + with this substitute. With this substitute, only the values 0 and 1 + give the expected result when converted to _Bool' or 'bool'. + + - C99 allows the use of (_Bool)0.0 in constant expressions, but + this substitute cannot always provide this property. + + Also, it is suggested that programs use 'bool' rather than '_Bool'; + this isn't required, but 'bool' is more common. */ + + +/* 7.16. Boolean type and values */ + +/* BeOS <sys/socket.h> already #defines false 0, true 1. We use the same + definitions below, but temporarily we have to #undef them. */ +#if defined __BEOS__ && !defined __HAIKU__ +# include <OS.h> /* defines bool but not _Bool */ +# undef false +# undef true +#endif + +/* For the sake of symbolic names in gdb, we define true and false as + enum constants, not only as macros. + It is tempting to write + typedef enum { false = 0, true = 1 } _Bool; + so that gdb prints values of type 'bool' symbolically. But if we do + this, values of type '_Bool' may promote to 'int' or 'unsigned int' + (see ISO C 99 6.7.2.2.(4)); however, '_Bool' must promote to 'int' + (see ISO C 99 6.3.1.1.(2)). So we add a negative value to the + enum; this ensures that '_Bool' promotes to 'int'. */ +#if defined __cplusplus || (defined __BEOS__ && !defined __HAIKU__) + /* A compiler known to have 'bool'. */ + /* If the compiler already has both 'bool' and '_Bool', we can assume they + are the same types. */ +# if !@HAVE__BOOL@ +typedef bool _Bool; +# endif +#else +# if !defined __GNUC__ + /* If @HAVE__BOOL@: + Some HP-UX cc and AIX IBM C compiler versions have compiler bugs when + the built-in _Bool type is used. See + http://gcc.gnu.org/ml/gcc-patches/2003-12/msg02303.html + http://lists.gnu.org/archive/html/bug-coreutils/2005-11/msg00161.html + http://lists.gnu.org/archive/html/bug-coreutils/2005-10/msg00086.html + Similar bugs are likely with other compilers as well; this file + wouldn't be used if <stdbool.h> was working. + So we override the _Bool type. + If !@HAVE__BOOL@: + Need to define _Bool ourselves. As 'signed char' or as an enum type? + Use of a typedef, with SunPRO C, leads to a stupid + "warning: _Bool is a keyword in ISO C99". + Use of an enum type, with IRIX cc, leads to a stupid + "warning(1185): enumerated type mixed with another type". + Even the existence of an enum type, without a typedef, + "Invalid enumerator. (badenum)" with HP-UX cc on Tru64. + The only benefit of the enum, debuggability, is not important + with these compilers. So use 'signed char' and no enum. */ +# define _Bool signed char +# else + /* With this compiler, trust the _Bool type if the compiler has it. */ +# if !@HAVE__BOOL@ +typedef enum { _Bool_must_promote_to_int = -1, false = 0, true = 1 } _Bool; +# endif +# endif +#endif +#define bool _Bool + +/* The other macros must be usable in preprocessor directives. */ +#define false 0 +#define true 1 +#define __bool_true_false_are_defined 1 + +#endif /* _GL_STDBOOL_H */ diff --git a/lib/stddef.in.h b/lib/stddef.in.h new file mode 100644 index 0000000..17fcaea --- /dev/null +++ b/lib/stddef.in.h @@ -0,0 +1,86 @@ +/* A substitute for POSIX 2008 <stddef.h>, for platforms that have issues. + + Copyright (C) 2009-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +/* Written by Eric Blake. */ + +/* + * POSIX 2008 <stddef.h> for platforms that have issues. + * <http://www.opengroup.org/susv3xbd/stddef.h.html> + */ + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +#if defined __need_wchar_t || defined __need_size_t \ + || defined __need_ptrdiff_t || defined __need_NULL \ + || defined __need_wint_t +/* Special invocation convention inside gcc header files. In + particular, gcc provides a version of <stddef.h> that blindly + redefines NULL even when __need_wint_t was defined, even though + wint_t is not normally provided by <stddef.h>. Hence, we must + remember if special invocation has ever been used to obtain wint_t, + in which case we need to clean up NULL yet again. */ + +# if !(defined _@GUARD_PREFIX@_STDDEF_H && defined _GL_STDDEF_WINT_T) +# ifdef __need_wint_t +# undef _@GUARD_PREFIX@_STDDEF_H +# define _GL_STDDEF_WINT_T +# endif +# @INCLUDE_NEXT@ @NEXT_STDDEF_H@ +# endif + +#else +/* Normal invocation convention. */ + +# ifndef _@GUARD_PREFIX@_STDDEF_H + +/* The include_next requires a split double-inclusion guard. */ + +# @INCLUDE_NEXT@ @NEXT_STDDEF_H@ + +# ifndef _@GUARD_PREFIX@_STDDEF_H +# define _@GUARD_PREFIX@_STDDEF_H + +/* On NetBSD 5.0, the definition of NULL lacks proper parentheses. */ +#if @REPLACE_NULL@ +# undef NULL +# ifdef __cplusplus + /* ISO C++ says that the macro NULL must expand to an integer constant + expression, hence '((void *) 0)' is not allowed in C++. */ +# if __GNUG__ >= 3 + /* GNU C++ has a __null macro that behaves like an integer ('int' or + 'long') but has the same size as a pointer. Use that, to avoid + warnings. */ +# define NULL __null +# else +# define NULL 0L +# endif +# else +# define NULL ((void *) 0) +# endif +#endif + +/* Some platforms lack wchar_t. */ +#if !@HAVE_WCHAR_T@ +# define wchar_t int +#endif + +# endif /* _@GUARD_PREFIX@_STDDEF_H */ +# endif /* _@GUARD_PREFIX@_STDDEF_H */ +#endif /* __need_XXX */ diff --git a/lib/stdint.in.h b/lib/stdint.in.h new file mode 100644 index 0000000..8e4c822 --- /dev/null +++ b/lib/stdint.in.h @@ -0,0 +1,619 @@ +/* Copyright (C) 2001-2002, 2004-2012 Free Software Foundation, Inc. + Written by Paul Eggert, Bruno Haible, Sam Steingold, Peter Burwood. + This file is part of gnulib. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +/* + * ISO C 99 <stdint.h> for platforms that lack it. + * <http://www.opengroup.org/susv3xbd/stdint.h.html> + */ + +#ifndef _@GUARD_PREFIX@_STDINT_H + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +/* When including a system file that in turn includes <inttypes.h>, + use the system <inttypes.h>, not our substitute. This avoids + problems with (for example) VMS, whose <sys/bitypes.h> includes + <inttypes.h>. */ +#define _GL_JUST_INCLUDE_SYSTEM_INTTYPES_H + +/* On Android (Bionic libc), <sys/types.h> includes this file before + having defined 'time_t'. Therefore in this case avoid including + other system header files; just include the system's <stdint.h>. + Ideally we should test __BIONIC__ here, but it is only defined after + <sys/cdefs.h> has been included; hence test __ANDROID__ instead. */ +#if defined __ANDROID__ \ + && defined _SYS_TYPES_H_ && !defined _SSIZE_T_DEFINED_ +# @INCLUDE_NEXT@ @NEXT_STDINT_H@ +#else + +/* Get those types that are already defined in other system include + files, so that we can "#define int8_t signed char" below without + worrying about a later system include file containing a "typedef + signed char int8_t;" that will get messed up by our macro. Our + macros should all be consistent with the system versions, except + for the "fast" types and macros, which we recommend against using + in public interfaces due to compiler differences. */ + +#if @HAVE_STDINT_H@ +# if defined __sgi && ! defined __c99 + /* Bypass IRIX's <stdint.h> if in C89 mode, since it merely annoys users + with "This header file is to be used only for c99 mode compilations" + diagnostics. */ +# define __STDINT_H__ +# endif + + /* Some pre-C++11 <stdint.h> implementations need this. */ +# ifdef __cplusplus +# ifndef __STDC_CONSTANT_MACROS +# define __STDC_CONSTANT_MACROS 1 +# endif +# ifndef __STDC_LIMIT_MACROS +# define __STDC_LIMIT_MACROS 1 +# endif +# endif + + /* Other systems may have an incomplete or buggy <stdint.h>. + Include it before <inttypes.h>, since any "#include <stdint.h>" + in <inttypes.h> would reinclude us, skipping our contents because + _@GUARD_PREFIX@_STDINT_H is defined. + The include_next requires a split double-inclusion guard. */ +# @INCLUDE_NEXT@ @NEXT_STDINT_H@ +#endif + +#if ! defined _@GUARD_PREFIX@_STDINT_H && ! defined _GL_JUST_INCLUDE_SYSTEM_STDINT_H +#define _@GUARD_PREFIX@_STDINT_H + +/* <sys/types.h> defines some of the stdint.h types as well, on glibc, + IRIX 6.5, and OpenBSD 3.8 (via <machine/types.h>). + AIX 5.2 <sys/types.h> isn't needed and causes troubles. + MacOS X 10.4.6 <sys/types.h> includes <stdint.h> (which is us), but + relies on the system <stdint.h> definitions, so include + <sys/types.h> after @NEXT_STDINT_H@. */ +#if @HAVE_SYS_TYPES_H@ && ! defined _AIX +# include <sys/types.h> +#endif + +/* Get LONG_MIN, LONG_MAX, ULONG_MAX. */ +#include <limits.h> + +#if @HAVE_INTTYPES_H@ + /* In OpenBSD 3.8, <inttypes.h> includes <machine/types.h>, which defines + int{8,16,32,64}_t, uint{8,16,32,64}_t and __BIT_TYPES_DEFINED__. + <inttypes.h> also defines intptr_t and uintptr_t. */ +# include <inttypes.h> +#elif @HAVE_SYS_INTTYPES_H@ + /* Solaris 7 <sys/inttypes.h> has the types except the *_fast*_t types, and + the macros except for *_FAST*_*, INTPTR_MIN, PTRDIFF_MIN, PTRDIFF_MAX. */ +# include <sys/inttypes.h> +#endif + +#if @HAVE_SYS_BITYPES_H@ && ! defined __BIT_TYPES_DEFINED__ + /* Linux libc4 >= 4.6.7 and libc5 have a <sys/bitypes.h> that defines + int{8,16,32,64}_t and __BIT_TYPES_DEFINED__. In libc5 >= 5.2.2 it is + included by <sys/types.h>. */ +# include <sys/bitypes.h> +#endif + +#undef _GL_JUST_INCLUDE_SYSTEM_INTTYPES_H + +/* Minimum and maximum values for an integer type under the usual assumption. + Return an unspecified value if BITS == 0, adding a check to pacify + picky compilers. */ + +#define _STDINT_MIN(signed, bits, zero) \ + ((signed) ? (- ((zero) + 1) << ((bits) ? (bits) - 1 : 0)) : (zero)) + +#define _STDINT_MAX(signed, bits, zero) \ + ((signed) \ + ? ~ _STDINT_MIN (signed, bits, zero) \ + : /* The expression for the unsigned case. The subtraction of (signed) \ + is a nop in the unsigned case and avoids "signed integer overflow" \ + warnings in the signed case. */ \ + ((((zero) + 1) << ((bits) ? (bits) - 1 - (signed) : 0)) - 1) * 2 + 1) + +#if !GNULIB_defined_stdint_types + +/* 7.18.1.1. Exact-width integer types */ + +/* Here we assume a standard architecture where the hardware integer + types have 8, 16, 32, optionally 64 bits. */ + +#undef int8_t +#undef uint8_t +typedef signed char gl_int8_t; +typedef unsigned char gl_uint8_t; +#define int8_t gl_int8_t +#define uint8_t gl_uint8_t + +#undef int16_t +#undef uint16_t +typedef short int gl_int16_t; +typedef unsigned short int gl_uint16_t; +#define int16_t gl_int16_t +#define uint16_t gl_uint16_t + +#undef int32_t +#undef uint32_t +typedef int gl_int32_t; +typedef unsigned int gl_uint32_t; +#define int32_t gl_int32_t +#define uint32_t gl_uint32_t + +/* If the system defines INT64_MAX, assume int64_t works. That way, + if the underlying platform defines int64_t to be a 64-bit long long + int, the code below won't mistakenly define it to be a 64-bit long + int, which would mess up C++ name mangling. We must use #ifdef + rather than #if, to avoid an error with HP-UX 10.20 cc. */ + +#ifdef INT64_MAX +# define GL_INT64_T +#else +/* Do not undefine int64_t if gnulib is not being used with 64-bit + types, since otherwise it breaks platforms like Tandem/NSK. */ +# if LONG_MAX >> 31 >> 31 == 1 +# undef int64_t +typedef long int gl_int64_t; +# define int64_t gl_int64_t +# define GL_INT64_T +# elif defined _MSC_VER +# undef int64_t +typedef __int64 gl_int64_t; +# define int64_t gl_int64_t +# define GL_INT64_T +# elif @HAVE_LONG_LONG_INT@ +# undef int64_t +typedef long long int gl_int64_t; +# define int64_t gl_int64_t +# define GL_INT64_T +# endif +#endif + +#ifdef UINT64_MAX +# define GL_UINT64_T +#else +# if ULONG_MAX >> 31 >> 31 >> 1 == 1 +# undef uint64_t +typedef unsigned long int gl_uint64_t; +# define uint64_t gl_uint64_t +# define GL_UINT64_T +# elif defined _MSC_VER +# undef uint64_t +typedef unsigned __int64 gl_uint64_t; +# define uint64_t gl_uint64_t +# define GL_UINT64_T +# elif @HAVE_UNSIGNED_LONG_LONG_INT@ +# undef uint64_t +typedef unsigned long long int gl_uint64_t; +# define uint64_t gl_uint64_t +# define GL_UINT64_T +# endif +#endif + +/* Avoid collision with Solaris 2.5.1 <pthread.h> etc. */ +#define _UINT8_T +#define _UINT32_T +#define _UINT64_T + + +/* 7.18.1.2. Minimum-width integer types */ + +/* Here we assume a standard architecture where the hardware integer + types have 8, 16, 32, optionally 64 bits. Therefore the leastN_t types + are the same as the corresponding N_t types. */ + +#undef int_least8_t +#undef uint_least8_t +#undef int_least16_t +#undef uint_least16_t +#undef int_least32_t +#undef uint_least32_t +#undef int_least64_t +#undef uint_least64_t +#define int_least8_t int8_t +#define uint_least8_t uint8_t +#define int_least16_t int16_t +#define uint_least16_t uint16_t +#define int_least32_t int32_t +#define uint_least32_t uint32_t +#ifdef GL_INT64_T +# define int_least64_t int64_t +#endif +#ifdef GL_UINT64_T +# define uint_least64_t uint64_t +#endif + +/* 7.18.1.3. Fastest minimum-width integer types */ + +/* Note: Other <stdint.h> substitutes may define these types differently. + It is not recommended to use these types in public header files. */ + +/* Here we assume a standard architecture where the hardware integer + types have 8, 16, 32, optionally 64 bits. Therefore the fastN_t types + are taken from the same list of types. Assume that 'long int' + is fast enough for all narrower integers. */ + +#undef int_fast8_t +#undef uint_fast8_t +#undef int_fast16_t +#undef uint_fast16_t +#undef int_fast32_t +#undef uint_fast32_t +#undef int_fast64_t +#undef uint_fast64_t +typedef long int gl_int_fast8_t; +typedef unsigned long int gl_uint_fast8_t; +typedef long int gl_int_fast16_t; +typedef unsigned long int gl_uint_fast16_t; +typedef long int gl_int_fast32_t; +typedef unsigned long int gl_uint_fast32_t; +#define int_fast8_t gl_int_fast8_t +#define uint_fast8_t gl_uint_fast8_t +#define int_fast16_t gl_int_fast16_t +#define uint_fast16_t gl_uint_fast16_t +#define int_fast32_t gl_int_fast32_t +#define uint_fast32_t gl_uint_fast32_t +#ifdef GL_INT64_T +# define int_fast64_t int64_t +#endif +#ifdef GL_UINT64_T +# define uint_fast64_t uint64_t +#endif + +/* 7.18.1.4. Integer types capable of holding object pointers */ + +#undef intptr_t +#undef uintptr_t +typedef long int gl_intptr_t; +typedef unsigned long int gl_uintptr_t; +#define intptr_t gl_intptr_t +#define uintptr_t gl_uintptr_t + +/* 7.18.1.5. Greatest-width integer types */ + +/* Note: These types are compiler dependent. It may be unwise to use them in + public header files. */ + +/* If the system defines INTMAX_MAX, assume that intmax_t works, and + similarly for UINTMAX_MAX and uintmax_t. This avoids problems with + assuming one type where another is used by the system. */ + +#ifndef INTMAX_MAX +# undef INTMAX_C +# undef intmax_t +# if @HAVE_LONG_LONG_INT@ && LONG_MAX >> 30 == 1 +typedef long long int gl_intmax_t; +# define intmax_t gl_intmax_t +# elif defined GL_INT64_T +# define intmax_t int64_t +# else +typedef long int gl_intmax_t; +# define intmax_t gl_intmax_t +# endif +#endif + +#ifndef UINTMAX_MAX +# undef UINTMAX_C +# undef uintmax_t +# if @HAVE_UNSIGNED_LONG_LONG_INT@ && ULONG_MAX >> 31 == 1 +typedef unsigned long long int gl_uintmax_t; +# define uintmax_t gl_uintmax_t +# elif defined GL_UINT64_T +# define uintmax_t uint64_t +# else +typedef unsigned long int gl_uintmax_t; +# define uintmax_t gl_uintmax_t +# endif +#endif + +/* Verify that intmax_t and uintmax_t have the same size. Too much code + breaks if this is not the case. If this check fails, the reason is likely + to be found in the autoconf macros. */ +typedef int _verify_intmax_size[sizeof (intmax_t) == sizeof (uintmax_t) + ? 1 : -1]; + +#define GNULIB_defined_stdint_types 1 +#endif /* !GNULIB_defined_stdint_types */ + +/* 7.18.2. Limits of specified-width integer types */ + +/* 7.18.2.1. Limits of exact-width integer types */ + +/* Here we assume a standard architecture where the hardware integer + types have 8, 16, 32, optionally 64 bits. */ + +#undef INT8_MIN +#undef INT8_MAX +#undef UINT8_MAX +#define INT8_MIN (~ INT8_MAX) +#define INT8_MAX 127 +#define UINT8_MAX 255 + +#undef INT16_MIN +#undef INT16_MAX +#undef UINT16_MAX +#define INT16_MIN (~ INT16_MAX) +#define INT16_MAX 32767 +#define UINT16_MAX 65535 + +#undef INT32_MIN +#undef INT32_MAX +#undef UINT32_MAX +#define INT32_MIN (~ INT32_MAX) +#define INT32_MAX 2147483647 +#define UINT32_MAX 4294967295U + +#if defined GL_INT64_T && ! defined INT64_MAX +/* Prefer (- INTMAX_C (1) << 63) over (~ INT64_MAX) because SunPRO C 5.0 + evaluates the latter incorrectly in preprocessor expressions. */ +# define INT64_MIN (- INTMAX_C (1) << 63) +# define INT64_MAX INTMAX_C (9223372036854775807) +#endif + +#if defined GL_UINT64_T && ! defined UINT64_MAX +# define UINT64_MAX UINTMAX_C (18446744073709551615) +#endif + +/* 7.18.2.2. Limits of minimum-width integer types */ + +/* Here we assume a standard architecture where the hardware integer + types have 8, 16, 32, optionally 64 bits. Therefore the leastN_t types + are the same as the corresponding N_t types. */ + +#undef INT_LEAST8_MIN +#undef INT_LEAST8_MAX +#undef UINT_LEAST8_MAX +#define INT_LEAST8_MIN INT8_MIN +#define INT_LEAST8_MAX INT8_MAX +#define UINT_LEAST8_MAX UINT8_MAX + +#undef INT_LEAST16_MIN +#undef INT_LEAST16_MAX +#undef UINT_LEAST16_MAX +#define INT_LEAST16_MIN INT16_MIN +#define INT_LEAST16_MAX INT16_MAX +#define UINT_LEAST16_MAX UINT16_MAX + +#undef INT_LEAST32_MIN +#undef INT_LEAST32_MAX +#undef UINT_LEAST32_MAX +#define INT_LEAST32_MIN INT32_MIN +#define INT_LEAST32_MAX INT32_MAX +#define UINT_LEAST32_MAX UINT32_MAX + +#undef INT_LEAST64_MIN +#undef INT_LEAST64_MAX +#ifdef GL_INT64_T +# define INT_LEAST64_MIN INT64_MIN +# define INT_LEAST64_MAX INT64_MAX +#endif + +#undef UINT_LEAST64_MAX +#ifdef GL_UINT64_T +# define UINT_LEAST64_MAX UINT64_MAX +#endif + +/* 7.18.2.3. Limits of fastest minimum-width integer types */ + +/* Here we assume a standard architecture where the hardware integer + types have 8, 16, 32, optionally 64 bits. Therefore the fastN_t types + are taken from the same list of types. */ + +#undef INT_FAST8_MIN +#undef INT_FAST8_MAX +#undef UINT_FAST8_MAX +#define INT_FAST8_MIN LONG_MIN +#define INT_FAST8_MAX LONG_MAX +#define UINT_FAST8_MAX ULONG_MAX + +#undef INT_FAST16_MIN +#undef INT_FAST16_MAX +#undef UINT_FAST16_MAX +#define INT_FAST16_MIN LONG_MIN +#define INT_FAST16_MAX LONG_MAX +#define UINT_FAST16_MAX ULONG_MAX + +#undef INT_FAST32_MIN +#undef INT_FAST32_MAX +#undef UINT_FAST32_MAX +#define INT_FAST32_MIN LONG_MIN +#define INT_FAST32_MAX LONG_MAX +#define UINT_FAST32_MAX ULONG_MAX + +#undef INT_FAST64_MIN +#undef INT_FAST64_MAX +#ifdef GL_INT64_T +# define INT_FAST64_MIN INT64_MIN +# define INT_FAST64_MAX INT64_MAX +#endif + +#undef UINT_FAST64_MAX +#ifdef GL_UINT64_T +# define UINT_FAST64_MAX UINT64_MAX +#endif + +/* 7.18.2.4. Limits of integer types capable of holding object pointers */ + +#undef INTPTR_MIN +#undef INTPTR_MAX +#undef UINTPTR_MAX +#define INTPTR_MIN LONG_MIN +#define INTPTR_MAX LONG_MAX +#define UINTPTR_MAX ULONG_MAX + +/* 7.18.2.5. Limits of greatest-width integer types */ + +#ifndef INTMAX_MAX +# undef INTMAX_MIN +# ifdef INT64_MAX +# define INTMAX_MIN INT64_MIN +# define INTMAX_MAX INT64_MAX +# else +# define INTMAX_MIN INT32_MIN +# define INTMAX_MAX INT32_MAX +# endif +#endif + +#ifndef UINTMAX_MAX +# ifdef UINT64_MAX +# define UINTMAX_MAX UINT64_MAX +# else +# define UINTMAX_MAX UINT32_MAX +# endif +#endif + +/* 7.18.3. Limits of other integer types */ + +/* ptrdiff_t limits */ +#undef PTRDIFF_MIN +#undef PTRDIFF_MAX +#if @APPLE_UNIVERSAL_BUILD@ +# ifdef _LP64 +# define PTRDIFF_MIN _STDINT_MIN (1, 64, 0l) +# define PTRDIFF_MAX _STDINT_MAX (1, 64, 0l) +# else +# define PTRDIFF_MIN _STDINT_MIN (1, 32, 0) +# define PTRDIFF_MAX _STDINT_MAX (1, 32, 0) +# endif +#else +# define PTRDIFF_MIN \ + _STDINT_MIN (1, @BITSIZEOF_PTRDIFF_T@, 0@PTRDIFF_T_SUFFIX@) +# define PTRDIFF_MAX \ + _STDINT_MAX (1, @BITSIZEOF_PTRDIFF_T@, 0@PTRDIFF_T_SUFFIX@) +#endif + +/* sig_atomic_t limits */ +#undef SIG_ATOMIC_MIN +#undef SIG_ATOMIC_MAX +#define SIG_ATOMIC_MIN \ + _STDINT_MIN (@HAVE_SIGNED_SIG_ATOMIC_T@, @BITSIZEOF_SIG_ATOMIC_T@, \ + 0@SIG_ATOMIC_T_SUFFIX@) +#define SIG_ATOMIC_MAX \ + _STDINT_MAX (@HAVE_SIGNED_SIG_ATOMIC_T@, @BITSIZEOF_SIG_ATOMIC_T@, \ + 0@SIG_ATOMIC_T_SUFFIX@) + + +/* size_t limit */ +#undef SIZE_MAX +#if @APPLE_UNIVERSAL_BUILD@ +# ifdef _LP64 +# define SIZE_MAX _STDINT_MAX (0, 64, 0ul) +# else +# define SIZE_MAX _STDINT_MAX (0, 32, 0ul) +# endif +#else +# define SIZE_MAX _STDINT_MAX (0, @BITSIZEOF_SIZE_T@, 0@SIZE_T_SUFFIX@) +#endif + +/* wchar_t limits */ +/* Get WCHAR_MIN, WCHAR_MAX. + This include is not on the top, above, because on OSF/1 4.0 we have a + sequence of nested includes + <wchar.h> -> <stdio.h> -> <getopt.h> -> <stdlib.h>, and the latter includes + <stdint.h> and assumes its types are already defined. */ +#if @HAVE_WCHAR_H@ && ! (defined WCHAR_MIN && defined WCHAR_MAX) + /* 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> +# define _GL_JUST_INCLUDE_SYSTEM_WCHAR_H +# include <wchar.h> +# undef _GL_JUST_INCLUDE_SYSTEM_WCHAR_H +#endif +#undef WCHAR_MIN +#undef WCHAR_MAX +#define WCHAR_MIN \ + _STDINT_MIN (@HAVE_SIGNED_WCHAR_T@, @BITSIZEOF_WCHAR_T@, 0@WCHAR_T_SUFFIX@) +#define WCHAR_MAX \ + _STDINT_MAX (@HAVE_SIGNED_WCHAR_T@, @BITSIZEOF_WCHAR_T@, 0@WCHAR_T_SUFFIX@) + +/* wint_t limits */ +#undef WINT_MIN +#undef WINT_MAX +#define WINT_MIN \ + _STDINT_MIN (@HAVE_SIGNED_WINT_T@, @BITSIZEOF_WINT_T@, 0@WINT_T_SUFFIX@) +#define WINT_MAX \ + _STDINT_MAX (@HAVE_SIGNED_WINT_T@, @BITSIZEOF_WINT_T@, 0@WINT_T_SUFFIX@) + +/* 7.18.4. Macros for integer constants */ + +/* 7.18.4.1. Macros for minimum-width integer constants */ +/* According to ISO C 99 Technical Corrigendum 1 */ + +/* Here we assume a standard architecture where the hardware integer + types have 8, 16, 32, optionally 64 bits, and int is 32 bits. */ + +#undef INT8_C +#undef UINT8_C +#define INT8_C(x) x +#define UINT8_C(x) x + +#undef INT16_C +#undef UINT16_C +#define INT16_C(x) x +#define UINT16_C(x) x + +#undef INT32_C +#undef UINT32_C +#define INT32_C(x) x +#define UINT32_C(x) x ## U + +#undef INT64_C +#undef UINT64_C +#if LONG_MAX >> 31 >> 31 == 1 +# define INT64_C(x) x##L +#elif defined _MSC_VER +# define INT64_C(x) x##i64 +#elif @HAVE_LONG_LONG_INT@ +# define INT64_C(x) x##LL +#endif +#if ULONG_MAX >> 31 >> 31 >> 1 == 1 +# define UINT64_C(x) x##UL +#elif defined _MSC_VER +# define UINT64_C(x) x##ui64 +#elif @HAVE_UNSIGNED_LONG_LONG_INT@ +# define UINT64_C(x) x##ULL +#endif + +/* 7.18.4.2. Macros for greatest-width integer constants */ + +#ifndef INTMAX_C +# if @HAVE_LONG_LONG_INT@ && LONG_MAX >> 30 == 1 +# define INTMAX_C(x) x##LL +# elif defined GL_INT64_T +# define INTMAX_C(x) INT64_C(x) +# else +# define INTMAX_C(x) x##L +# endif +#endif + +#ifndef UINTMAX_C +# if @HAVE_UNSIGNED_LONG_LONG_INT@ && ULONG_MAX >> 31 == 1 +# define UINTMAX_C(x) x##ULL +# elif defined GL_UINT64_T +# define UINTMAX_C(x) UINT64_C(x) +# else +# define UINTMAX_C(x) x##UL +# endif +#endif + +#endif /* _@GUARD_PREFIX@_STDINT_H */ +#endif /* !(defined __ANDROID__ && ...) */ +#endif /* !defined _@GUARD_PREFIX@_STDINT_H && !defined _GL_JUST_INCLUDE_SYSTEM_STDINT_H */ diff --git a/lib/stdlib.in.h b/lib/stdlib.in.h new file mode 100644 index 0000000..e92a614 --- /dev/null +++ b/lib/stdlib.in.h @@ -0,0 +1,918 @@ +/* A GNU-like <stdlib.h>. + + Copyright (C) 1995, 2001-2004, 2006-2012 Free Software Foundation, Inc. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +#if defined __need_malloc_and_calloc +/* Special invocation convention inside glibc header files. */ + +#@INCLUDE_NEXT@ @NEXT_STDLIB_H@ + +#else +/* Normal invocation convention. */ + +#ifndef _@GUARD_PREFIX@_STDLIB_H + +/* The include_next requires a split double-inclusion guard. */ +#@INCLUDE_NEXT@ @NEXT_STDLIB_H@ + +#ifndef _@GUARD_PREFIX@_STDLIB_H +#define _@GUARD_PREFIX@_STDLIB_H + +/* NetBSD 5.0 mis-defines NULL. */ +#include <stddef.h> + +/* MirBSD 10 defines WEXITSTATUS in <sys/wait.h>, not in <stdlib.h>. */ +#if @GNULIB_SYSTEM_POSIX@ && !defined WEXITSTATUS +# include <sys/wait.h> +#endif + +/* Solaris declares getloadavg() in <sys/loadavg.h>. */ +#if (@GNULIB_GETLOADAVG@ || defined GNULIB_POSIXCHECK) && @HAVE_SYS_LOADAVG_H@ +# include <sys/loadavg.h> +#endif + +/* Native Windows platforms declare mktemp() in <io.h>. */ +#if 0 && ((defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__) +# include <io.h> +#endif + +#if @GNULIB_RANDOM_R@ + +/* OSF/1 5.1 declares 'struct random_data' in <random.h>, which is included + from <stdlib.h> if _REENTRANT is defined. Include it whenever we need + 'struct random_data'. */ +# if @HAVE_RANDOM_H@ +# include <random.h> +# endif + +# if !@HAVE_STRUCT_RANDOM_DATA@ || @REPLACE_RANDOM_R@ || !@HAVE_RANDOM_R@ +# include <stdint.h> +# endif + +# if !@HAVE_STRUCT_RANDOM_DATA@ +/* Define 'struct random_data'. + But allow multiple gnulib generated <stdlib.h> replacements to coexist. */ +# if !GNULIB_defined_struct_random_data +struct random_data +{ + int32_t *fptr; /* Front pointer. */ + int32_t *rptr; /* Rear pointer. */ + int32_t *state; /* Array of state values. */ + int rand_type; /* Type of random number generator. */ + int rand_deg; /* Degree of random number generator. */ + int rand_sep; /* Distance between front and rear. */ + int32_t *end_ptr; /* Pointer behind state table. */ +}; +# define GNULIB_defined_struct_random_data 1 +# endif +# endif +#endif + +#if (@GNULIB_MKSTEMP@ || @GNULIB_MKSTEMPS@ || @GNULIB_GETSUBOPT@ || defined GNULIB_POSIXCHECK) && ! defined __GLIBC__ && !((defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__) +/* On MacOS X 10.3, only <unistd.h> declares mkstemp. */ +/* On MacOS X 10.5, only <unistd.h> declares mkstemps. */ +/* On Cygwin 1.7.1, only <unistd.h> declares getsubopt. */ +/* But avoid namespace pollution on glibc systems and native Windows. */ +# include <unistd.h> +#endif + +/* The definition of _Noreturn is copied here. */ + +/* The definitions of _GL_FUNCDECL_RPL etc. are copied here. */ + +/* The definition of _GL_ARG_NONNULL is copied here. */ + +/* The definition of _GL_WARN_ON_USE is copied here. */ + + +/* Some systems do not define EXIT_*, despite otherwise supporting C89. */ +#ifndef EXIT_SUCCESS +# define EXIT_SUCCESS 0 +#endif +/* Tandem/NSK and other platforms that define EXIT_FAILURE as -1 interfere + with proper operation of xargs. */ +#ifndef EXIT_FAILURE +# define EXIT_FAILURE 1 +#elif EXIT_FAILURE != 1 +# undef EXIT_FAILURE +# define EXIT_FAILURE 1 +#endif + + +#if @GNULIB__EXIT@ +/* Terminate the current process with the given return code, without running + the 'atexit' handlers. */ +# if !@HAVE__EXIT@ +_GL_FUNCDECL_SYS (_Exit, _Noreturn void, (int status)); +# endif +_GL_CXXALIAS_SYS (_Exit, void, (int status)); +_GL_CXXALIASWARN (_Exit); +#elif defined GNULIB_POSIXCHECK +# undef _Exit +# if HAVE_RAW_DECL__EXIT +_GL_WARN_ON_USE (_Exit, "_Exit is unportable - " + "use gnulib module _Exit for portability"); +# endif +#endif + + +#if @GNULIB_ATOLL@ +/* Parse a signed decimal integer. + Returns the value of the integer. Errors are not detected. */ +# if !@HAVE_ATOLL@ +_GL_FUNCDECL_SYS (atoll, long long, (const char *string) _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (atoll, long long, (const char *string)); +_GL_CXXALIASWARN (atoll); +#elif defined GNULIB_POSIXCHECK +# undef atoll +# if HAVE_RAW_DECL_ATOLL +_GL_WARN_ON_USE (atoll, "atoll is unportable - " + "use gnulib module atoll for portability"); +# endif +#endif + +#if @GNULIB_CALLOC_POSIX@ +# if @REPLACE_CALLOC@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef calloc +# define calloc rpl_calloc +# endif +_GL_FUNCDECL_RPL (calloc, void *, (size_t nmemb, size_t size)); +_GL_CXXALIAS_RPL (calloc, void *, (size_t nmemb, size_t size)); +# else +_GL_CXXALIAS_SYS (calloc, void *, (size_t nmemb, size_t size)); +# endif +_GL_CXXALIASWARN (calloc); +#elif defined GNULIB_POSIXCHECK +# undef calloc +/* Assume calloc is always declared. */ +_GL_WARN_ON_USE (calloc, "calloc is not POSIX compliant everywhere - " + "use gnulib module calloc-posix for portability"); +#endif + +#if @GNULIB_CANONICALIZE_FILE_NAME@ +# if @REPLACE_CANONICALIZE_FILE_NAME@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define canonicalize_file_name rpl_canonicalize_file_name +# endif +_GL_FUNCDECL_RPL (canonicalize_file_name, char *, (const char *name) + _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (canonicalize_file_name, char *, (const char *name)); +# else +# if !@HAVE_CANONICALIZE_FILE_NAME@ +_GL_FUNCDECL_SYS (canonicalize_file_name, char *, (const char *name) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (canonicalize_file_name, char *, (const char *name)); +# endif +_GL_CXXALIASWARN (canonicalize_file_name); +#elif defined GNULIB_POSIXCHECK +# undef canonicalize_file_name +# if HAVE_RAW_DECL_CANONICALIZE_FILE_NAME +_GL_WARN_ON_USE (canonicalize_file_name, + "canonicalize_file_name is unportable - " + "use gnulib module canonicalize-lgpl for portability"); +# endif +#endif + +#if @GNULIB_GETLOADAVG@ +/* Store max(NELEM,3) load average numbers in LOADAVG[]. + The three numbers are the load average of the last 1 minute, the last 5 + minutes, and the last 15 minutes, respectively. + LOADAVG is an array of NELEM numbers. */ +# if !@HAVE_DECL_GETLOADAVG@ +_GL_FUNCDECL_SYS (getloadavg, int, (double loadavg[], int nelem) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (getloadavg, int, (double loadavg[], int nelem)); +_GL_CXXALIASWARN (getloadavg); +#elif defined GNULIB_POSIXCHECK +# undef getloadavg +# if HAVE_RAW_DECL_GETLOADAVG +_GL_WARN_ON_USE (getloadavg, "getloadavg is not portable - " + "use gnulib module getloadavg for portability"); +# endif +#endif + +#if @GNULIB_GETSUBOPT@ +/* Assuming *OPTIONP is a comma separated list of elements of the form + "token" or "token=value", getsubopt parses the first of these elements. + If the first element refers to a "token" that is member of the given + NULL-terminated array of tokens: + - It replaces the comma with a NUL byte, updates *OPTIONP to point past + the first option and the comma, sets *VALUEP to the value of the + element (or NULL if it doesn't contain an "=" sign), + - It returns the index of the "token" in the given array of tokens. + Otherwise it returns -1, and *OPTIONP and *VALUEP are undefined. + For more details see the POSIX:2001 specification. + http://www.opengroup.org/susv3xsh/getsubopt.html */ +# if !@HAVE_GETSUBOPT@ +_GL_FUNCDECL_SYS (getsubopt, int, + (char **optionp, char *const *tokens, char **valuep) + _GL_ARG_NONNULL ((1, 2, 3))); +# endif +_GL_CXXALIAS_SYS (getsubopt, int, + (char **optionp, char *const *tokens, char **valuep)); +_GL_CXXALIASWARN (getsubopt); +#elif defined GNULIB_POSIXCHECK +# undef getsubopt +# if HAVE_RAW_DECL_GETSUBOPT +_GL_WARN_ON_USE (getsubopt, "getsubopt is unportable - " + "use gnulib module getsubopt for portability"); +# endif +#endif + +#if @GNULIB_GRANTPT@ +/* Change the ownership and access permission of the slave side of the + pseudo-terminal whose master side is specified by FD. */ +# if !@HAVE_GRANTPT@ +_GL_FUNCDECL_SYS (grantpt, int, (int fd)); +# endif +_GL_CXXALIAS_SYS (grantpt, int, (int fd)); +_GL_CXXALIASWARN (grantpt); +#elif defined GNULIB_POSIXCHECK +# undef grantpt +# if HAVE_RAW_DECL_GRANTPT +_GL_WARN_ON_USE (grantpt, "grantpt is not portable - " + "use gnulib module grantpt for portability"); +# endif +#endif + +/* If _GL_USE_STDLIB_ALLOC is nonzero, the including module does not + rely on GNU or POSIX semantics for malloc and realloc (for example, + by never specifying a zero size), so it does not need malloc or + realloc to be redefined. */ +#if @GNULIB_MALLOC_POSIX@ +# if @REPLACE_MALLOC@ +# if !((defined __cplusplus && defined GNULIB_NAMESPACE) \ + || _GL_USE_STDLIB_ALLOC) +# undef malloc +# define malloc rpl_malloc +# endif +_GL_FUNCDECL_RPL (malloc, void *, (size_t size)); +_GL_CXXALIAS_RPL (malloc, void *, (size_t size)); +# else +_GL_CXXALIAS_SYS (malloc, void *, (size_t size)); +# endif +_GL_CXXALIASWARN (malloc); +#elif defined GNULIB_POSIXCHECK && !_GL_USE_STDLIB_ALLOC +# undef malloc +/* Assume malloc is always declared. */ +_GL_WARN_ON_USE (malloc, "malloc is not POSIX compliant everywhere - " + "use gnulib module malloc-posix for portability"); +#endif + +/* Convert a multibyte character to a wide character. */ +#if @GNULIB_MBTOWC@ +# if @REPLACE_MBTOWC@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef mbtowc +# define mbtowc rpl_mbtowc +# endif +_GL_FUNCDECL_RPL (mbtowc, int, (wchar_t *pwc, const char *s, size_t n)); +_GL_CXXALIAS_RPL (mbtowc, int, (wchar_t *pwc, const char *s, size_t n)); +# else +_GL_CXXALIAS_SYS (mbtowc, int, (wchar_t *pwc, const char *s, size_t n)); +# endif +_GL_CXXALIASWARN (mbtowc); +#endif + +#if @GNULIB_MKDTEMP@ +/* Create a unique temporary directory from TEMPLATE. + The last six characters of TEMPLATE must be "XXXXXX"; + they are replaced with a string that makes the directory name unique. + Returns TEMPLATE, or a null pointer if it cannot get a unique name. + The directory is created mode 700. */ +# if !@HAVE_MKDTEMP@ +_GL_FUNCDECL_SYS (mkdtemp, char *, (char * /*template*/) _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (mkdtemp, char *, (char * /*template*/)); +_GL_CXXALIASWARN (mkdtemp); +#elif defined GNULIB_POSIXCHECK +# undef mkdtemp +# if HAVE_RAW_DECL_MKDTEMP +_GL_WARN_ON_USE (mkdtemp, "mkdtemp is unportable - " + "use gnulib module mkdtemp for portability"); +# endif +#endif + +#if @GNULIB_MKOSTEMP@ +/* Create a unique temporary file from TEMPLATE. + The last six characters of TEMPLATE must be "XXXXXX"; + they are replaced with a string that makes the file name unique. + The flags are a bitmask, possibly including O_CLOEXEC (defined in <fcntl.h>) + and O_TEXT, O_BINARY (defined in "binary-io.h"). + The file is then created, with the specified flags, ensuring it didn't exist + before. + The file is created read-write (mask at least 0600 & ~umask), but it may be + world-readable and world-writable (mask 0666 & ~umask), depending on the + implementation. + Returns the open file descriptor if successful, otherwise -1 and errno + set. */ +# if !@HAVE_MKOSTEMP@ +_GL_FUNCDECL_SYS (mkostemp, int, (char * /*template*/, int /*flags*/) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (mkostemp, int, (char * /*template*/, int /*flags*/)); +_GL_CXXALIASWARN (mkostemp); +#elif defined GNULIB_POSIXCHECK +# undef mkostemp +# if HAVE_RAW_DECL_MKOSTEMP +_GL_WARN_ON_USE (mkostemp, "mkostemp is unportable - " + "use gnulib module mkostemp for portability"); +# endif +#endif + +#if @GNULIB_MKOSTEMPS@ +/* Create a unique temporary file from TEMPLATE. + The last six characters of TEMPLATE before a suffix of length + SUFFIXLEN must be "XXXXXX"; + they are replaced with a string that makes the file name unique. + The flags are a bitmask, possibly including O_CLOEXEC (defined in <fcntl.h>) + and O_TEXT, O_BINARY (defined in "binary-io.h"). + The file is then created, with the specified flags, ensuring it didn't exist + before. + The file is created read-write (mask at least 0600 & ~umask), but it may be + world-readable and world-writable (mask 0666 & ~umask), depending on the + implementation. + Returns the open file descriptor if successful, otherwise -1 and errno + set. */ +# if !@HAVE_MKOSTEMPS@ +_GL_FUNCDECL_SYS (mkostemps, int, + (char * /*template*/, int /*suffixlen*/, int /*flags*/) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (mkostemps, int, + (char * /*template*/, int /*suffixlen*/, int /*flags*/)); +_GL_CXXALIASWARN (mkostemps); +#elif defined GNULIB_POSIXCHECK +# undef mkostemps +# if HAVE_RAW_DECL_MKOSTEMPS +_GL_WARN_ON_USE (mkostemps, "mkostemps is unportable - " + "use gnulib module mkostemps for portability"); +# endif +#endif + +#if @GNULIB_MKSTEMP@ +/* Create a unique temporary file from TEMPLATE. + The last six characters of TEMPLATE must be "XXXXXX"; + they are replaced with a string that makes the file name unique. + The file is then created, ensuring it didn't exist before. + The file is created read-write (mask at least 0600 & ~umask), but it may be + world-readable and world-writable (mask 0666 & ~umask), depending on the + implementation. + Returns the open file descriptor if successful, otherwise -1 and errno + set. */ +# if @REPLACE_MKSTEMP@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define mkstemp rpl_mkstemp +# endif +_GL_FUNCDECL_RPL (mkstemp, int, (char * /*template*/) _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (mkstemp, int, (char * /*template*/)); +# else +# if ! @HAVE_MKSTEMP@ +_GL_FUNCDECL_SYS (mkstemp, int, (char * /*template*/) _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (mkstemp, int, (char * /*template*/)); +# endif +_GL_CXXALIASWARN (mkstemp); +#elif defined GNULIB_POSIXCHECK +# undef mkstemp +# if HAVE_RAW_DECL_MKSTEMP +_GL_WARN_ON_USE (mkstemp, "mkstemp is unportable - " + "use gnulib module mkstemp for portability"); +# endif +#endif + +#if @GNULIB_MKSTEMPS@ +/* Create a unique temporary file from TEMPLATE. + The last six characters of TEMPLATE prior to a suffix of length + SUFFIXLEN must be "XXXXXX"; + they are replaced with a string that makes the file name unique. + The file is then created, ensuring it didn't exist before. + The file is created read-write (mask at least 0600 & ~umask), but it may be + world-readable and world-writable (mask 0666 & ~umask), depending on the + implementation. + Returns the open file descriptor if successful, otherwise -1 and errno + set. */ +# if !@HAVE_MKSTEMPS@ +_GL_FUNCDECL_SYS (mkstemps, int, (char * /*template*/, int /*suffixlen*/) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (mkstemps, int, (char * /*template*/, int /*suffixlen*/)); +_GL_CXXALIASWARN (mkstemps); +#elif defined GNULIB_POSIXCHECK +# undef mkstemps +# if HAVE_RAW_DECL_MKSTEMPS +_GL_WARN_ON_USE (mkstemps, "mkstemps is unportable - " + "use gnulib module mkstemps for portability"); +# endif +#endif + +#if @GNULIB_POSIX_OPENPT@ +/* Return an FD open to the master side of a pseudo-terminal. Flags should + include O_RDWR, and may also include O_NOCTTY. */ +# if !@HAVE_POSIX_OPENPT@ +_GL_FUNCDECL_SYS (posix_openpt, int, (int flags)); +# endif +_GL_CXXALIAS_SYS (posix_openpt, int, (int flags)); +_GL_CXXALIASWARN (posix_openpt); +#elif defined GNULIB_POSIXCHECK +# undef posix_openpt +# if HAVE_RAW_DECL_POSIX_OPENPT +_GL_WARN_ON_USE (posix_openpt, "posix_openpt is not portable - " + "use gnulib module posix_openpt for portability"); +# endif +#endif + +#if @GNULIB_PTSNAME@ +/* Return the pathname of the pseudo-terminal slave associated with + the master FD is open on, or NULL on errors. */ +# if !@HAVE_PTSNAME@ +_GL_FUNCDECL_SYS (ptsname, char *, (int fd)); +# endif +_GL_CXXALIAS_SYS (ptsname, char *, (int fd)); +_GL_CXXALIASWARN (ptsname); +#elif defined GNULIB_POSIXCHECK +# undef ptsname +# if HAVE_RAW_DECL_PTSNAME +_GL_WARN_ON_USE (ptsname, "ptsname is not portable - " + "use gnulib module ptsname for portability"); +# endif +#endif + +#if @GNULIB_PTSNAME_R@ +/* Set the pathname of the pseudo-terminal slave associated with + the master FD is open on and return 0, or set errno and return + non-zero on errors. */ +# if @REPLACE_PTSNAME_R@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef ptsname_r +# define ptsname_r rpl_ptsname_r +# endif +_GL_FUNCDECL_RPL (ptsname_r, int, (int fd, char *buf, size_t len)); +_GL_CXXALIAS_RPL (ptsname_r, int, (int fd, char *buf, size_t len)); +# else +# if !@HAVE_PTSNAME_R@ +_GL_FUNCDECL_SYS (ptsname_r, int, (int fd, char *buf, size_t len)); +# endif +_GL_CXXALIAS_SYS (ptsname_r, int, (int fd, char *buf, size_t len)); +# endif +_GL_CXXALIASWARN (ptsname_r); +#elif defined GNULIB_POSIXCHECK +# undef ptsname_r +# if HAVE_RAW_DECL_PTSNAME_R +_GL_WARN_ON_USE (ptsname_r, "ptsname_r is not portable - " + "use gnulib module ptsname_r for portability"); +# endif +#endif + +#if @GNULIB_PUTENV@ +# if @REPLACE_PUTENV@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef putenv +# define putenv rpl_putenv +# endif +_GL_FUNCDECL_RPL (putenv, int, (char *string) _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (putenv, int, (char *string)); +# else +_GL_CXXALIAS_SYS (putenv, int, (char *string)); +# endif +_GL_CXXALIASWARN (putenv); +#endif + + +#if @GNULIB_RANDOM_R@ +# if !@HAVE_RANDOM_R@ +# ifndef RAND_MAX +# define RAND_MAX 2147483647 +# endif +# endif +#endif + + +#if @GNULIB_RANDOM@ +# if !@HAVE_RANDOM@ +_GL_FUNCDECL_SYS (random, long, (void)); +# endif +_GL_CXXALIAS_SYS (random, long, (void)); +_GL_CXXALIASWARN (random); +#elif defined GNULIB_POSIXCHECK +# undef random +# if HAVE_RAW_DECL_RANDOM +_GL_WARN_ON_USE (random, "random is unportable - " + "use gnulib module random for portability"); +# endif +#endif + +#if @GNULIB_RANDOM@ +# if !@HAVE_RANDOM@ +_GL_FUNCDECL_SYS (srandom, void, (unsigned int seed)); +# endif +_GL_CXXALIAS_SYS (srandom, void, (unsigned int seed)); +_GL_CXXALIASWARN (srandom); +#elif defined GNULIB_POSIXCHECK +# undef srandom +# if HAVE_RAW_DECL_SRANDOM +_GL_WARN_ON_USE (srandom, "srandom is unportable - " + "use gnulib module random for portability"); +# endif +#endif + +#if @GNULIB_RANDOM@ +# if !@HAVE_RANDOM@ +_GL_FUNCDECL_SYS (initstate, char *, + (unsigned int seed, char *buf, size_t buf_size) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (initstate, char *, + (unsigned int seed, char *buf, size_t buf_size)); +_GL_CXXALIASWARN (initstate); +#elif defined GNULIB_POSIXCHECK +# undef initstate +# if HAVE_RAW_DECL_INITSTATE_R +_GL_WARN_ON_USE (initstate, "initstate is unportable - " + "use gnulib module random for portability"); +# endif +#endif + +#if @GNULIB_RANDOM@ +# if !@HAVE_RANDOM@ +_GL_FUNCDECL_SYS (setstate, char *, (char *arg_state) _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (setstate, char *, (char *arg_state)); +_GL_CXXALIASWARN (setstate); +#elif defined GNULIB_POSIXCHECK +# undef setstate +# if HAVE_RAW_DECL_SETSTATE_R +_GL_WARN_ON_USE (setstate, "setstate is unportable - " + "use gnulib module random for portability"); +# endif +#endif + + +#if @GNULIB_RANDOM_R@ +# if @REPLACE_RANDOM_R@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef random_r +# define random_r rpl_random_r +# endif +_GL_FUNCDECL_RPL (random_r, int, (struct random_data *buf, int32_t *result) + _GL_ARG_NONNULL ((1, 2))); +_GL_CXXALIAS_RPL (random_r, int, (struct random_data *buf, int32_t *result)); +# else +# if !@HAVE_RANDOM_R@ +_GL_FUNCDECL_SYS (random_r, int, (struct random_data *buf, int32_t *result) + _GL_ARG_NONNULL ((1, 2))); +# endif +_GL_CXXALIAS_SYS (random_r, int, (struct random_data *buf, int32_t *result)); +# endif +_GL_CXXALIASWARN (random_r); +#elif defined GNULIB_POSIXCHECK +# undef random_r +# if HAVE_RAW_DECL_RANDOM_R +_GL_WARN_ON_USE (random_r, "random_r is unportable - " + "use gnulib module random_r for portability"); +# endif +#endif + +#if @GNULIB_RANDOM_R@ +# if @REPLACE_RANDOM_R@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef srandom_r +# define srandom_r rpl_srandom_r +# endif +_GL_FUNCDECL_RPL (srandom_r, int, + (unsigned int seed, struct random_data *rand_state) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (srandom_r, int, + (unsigned int seed, struct random_data *rand_state)); +# else +# if !@HAVE_RANDOM_R@ +_GL_FUNCDECL_SYS (srandom_r, int, + (unsigned int seed, struct random_data *rand_state) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (srandom_r, int, + (unsigned int seed, struct random_data *rand_state)); +# endif +_GL_CXXALIASWARN (srandom_r); +#elif defined GNULIB_POSIXCHECK +# undef srandom_r +# if HAVE_RAW_DECL_SRANDOM_R +_GL_WARN_ON_USE (srandom_r, "srandom_r is unportable - " + "use gnulib module random_r for portability"); +# endif +#endif + +#if @GNULIB_RANDOM_R@ +# if @REPLACE_RANDOM_R@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef initstate_r +# define initstate_r rpl_initstate_r +# endif +_GL_FUNCDECL_RPL (initstate_r, int, + (unsigned int seed, char *buf, size_t buf_size, + struct random_data *rand_state) + _GL_ARG_NONNULL ((2, 4))); +_GL_CXXALIAS_RPL (initstate_r, int, + (unsigned int seed, char *buf, size_t buf_size, + struct random_data *rand_state)); +# else +# if !@HAVE_RANDOM_R@ +_GL_FUNCDECL_SYS (initstate_r, int, + (unsigned int seed, char *buf, size_t buf_size, + struct random_data *rand_state) + _GL_ARG_NONNULL ((2, 4))); +# endif +_GL_CXXALIAS_SYS (initstate_r, int, + (unsigned int seed, char *buf, size_t buf_size, + struct random_data *rand_state)); +# endif +_GL_CXXALIASWARN (initstate_r); +#elif defined GNULIB_POSIXCHECK +# undef initstate_r +# if HAVE_RAW_DECL_INITSTATE_R +_GL_WARN_ON_USE (initstate_r, "initstate_r is unportable - " + "use gnulib module random_r for portability"); +# endif +#endif + +#if @GNULIB_RANDOM_R@ +# if @REPLACE_RANDOM_R@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef setstate_r +# define setstate_r rpl_setstate_r +# endif +_GL_FUNCDECL_RPL (setstate_r, int, + (char *arg_state, struct random_data *rand_state) + _GL_ARG_NONNULL ((1, 2))); +_GL_CXXALIAS_RPL (setstate_r, int, + (char *arg_state, struct random_data *rand_state)); +# else +# if !@HAVE_RANDOM_R@ +_GL_FUNCDECL_SYS (setstate_r, int, + (char *arg_state, struct random_data *rand_state) + _GL_ARG_NONNULL ((1, 2))); +# endif +_GL_CXXALIAS_SYS (setstate_r, int, + (char *arg_state, struct random_data *rand_state)); +# endif +_GL_CXXALIASWARN (setstate_r); +#elif defined GNULIB_POSIXCHECK +# undef setstate_r +# if HAVE_RAW_DECL_SETSTATE_R +_GL_WARN_ON_USE (setstate_r, "setstate_r is unportable - " + "use gnulib module random_r for portability"); +# endif +#endif + + +#if @GNULIB_REALLOC_POSIX@ +# if @REPLACE_REALLOC@ +# if !((defined __cplusplus && defined GNULIB_NAMESPACE) \ + || _GL_USE_STDLIB_ALLOC) +# undef realloc +# define realloc rpl_realloc +# endif +_GL_FUNCDECL_RPL (realloc, void *, (void *ptr, size_t size)); +_GL_CXXALIAS_RPL (realloc, void *, (void *ptr, size_t size)); +# else +_GL_CXXALIAS_SYS (realloc, void *, (void *ptr, size_t size)); +# endif +_GL_CXXALIASWARN (realloc); +#elif defined GNULIB_POSIXCHECK && !_GL_USE_STDLIB_ALLOC +# undef realloc +/* Assume realloc is always declared. */ +_GL_WARN_ON_USE (realloc, "realloc is not POSIX compliant everywhere - " + "use gnulib module realloc-posix for portability"); +#endif + +#if @GNULIB_REALPATH@ +# if @REPLACE_REALPATH@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define realpath rpl_realpath +# endif +_GL_FUNCDECL_RPL (realpath, char *, (const char *name, char *resolved) + _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (realpath, char *, (const char *name, char *resolved)); +# else +# if !@HAVE_REALPATH@ +_GL_FUNCDECL_SYS (realpath, char *, (const char *name, char *resolved) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (realpath, char *, (const char *name, char *resolved)); +# endif +_GL_CXXALIASWARN (realpath); +#elif defined GNULIB_POSIXCHECK +# undef realpath +# if HAVE_RAW_DECL_REALPATH +_GL_WARN_ON_USE (realpath, "realpath is unportable - use gnulib module " + "canonicalize or canonicalize-lgpl for portability"); +# endif +#endif + +#if @GNULIB_RPMATCH@ +/* Test a user response to a question. + Return 1 if it is affirmative, 0 if it is negative, or -1 if not clear. */ +# if !@HAVE_RPMATCH@ +_GL_FUNCDECL_SYS (rpmatch, int, (const char *response) _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (rpmatch, int, (const char *response)); +_GL_CXXALIASWARN (rpmatch); +#elif defined GNULIB_POSIXCHECK +# undef rpmatch +# if HAVE_RAW_DECL_RPMATCH +_GL_WARN_ON_USE (rpmatch, "rpmatch is unportable - " + "use gnulib module rpmatch for portability"); +# endif +#endif + +#if @GNULIB_SETENV@ +/* Set NAME to VALUE in the environment. + If REPLACE is nonzero, overwrite an existing value. */ +# if @REPLACE_SETENV@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef setenv +# define setenv rpl_setenv +# endif +_GL_FUNCDECL_RPL (setenv, int, + (const char *name, const char *value, int replace) + _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (setenv, int, + (const char *name, const char *value, int replace)); +# else +# if !@HAVE_DECL_SETENV@ +_GL_FUNCDECL_SYS (setenv, int, + (const char *name, const char *value, int replace) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (setenv, int, + (const char *name, const char *value, int replace)); +# endif +# if !(@REPLACE_SETENV@ && !@HAVE_DECL_SETENV@) +_GL_CXXALIASWARN (setenv); +# endif +#elif defined GNULIB_POSIXCHECK +# undef setenv +# if HAVE_RAW_DECL_SETENV +_GL_WARN_ON_USE (setenv, "setenv is unportable - " + "use gnulib module setenv for portability"); +# endif +#endif + +#if @GNULIB_STRTOD@ + /* Parse a double from STRING, updating ENDP if appropriate. */ +# if @REPLACE_STRTOD@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define strtod rpl_strtod +# endif +_GL_FUNCDECL_RPL (strtod, double, (const char *str, char **endp) + _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (strtod, double, (const char *str, char **endp)); +# else +# if !@HAVE_STRTOD@ +_GL_FUNCDECL_SYS (strtod, double, (const char *str, char **endp) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (strtod, double, (const char *str, char **endp)); +# endif +_GL_CXXALIASWARN (strtod); +#elif defined GNULIB_POSIXCHECK +# undef strtod +# if HAVE_RAW_DECL_STRTOD +_GL_WARN_ON_USE (strtod, "strtod is unportable - " + "use gnulib module strtod for portability"); +# endif +#endif + +#if @GNULIB_STRTOLL@ +/* Parse a signed integer whose textual representation starts at STRING. + The integer is expected to be in base BASE (2 <= BASE <= 36); if BASE == 0, + it may be decimal or octal (with prefix "0") or hexadecimal (with prefix + "0x"). + If ENDPTR is not NULL, the address of the first byte after the integer is + stored in *ENDPTR. + Upon overflow, the return value is LLONG_MAX or LLONG_MIN, and errno is set + to ERANGE. */ +# if !@HAVE_STRTOLL@ +_GL_FUNCDECL_SYS (strtoll, long long, + (const char *string, char **endptr, int base) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (strtoll, long long, + (const char *string, char **endptr, int base)); +_GL_CXXALIASWARN (strtoll); +#elif defined GNULIB_POSIXCHECK +# undef strtoll +# if HAVE_RAW_DECL_STRTOLL +_GL_WARN_ON_USE (strtoll, "strtoll is unportable - " + "use gnulib module strtoll for portability"); +# endif +#endif + +#if @GNULIB_STRTOULL@ +/* Parse an unsigned integer whose textual representation starts at STRING. + The integer is expected to be in base BASE (2 <= BASE <= 36); if BASE == 0, + it may be decimal or octal (with prefix "0") or hexadecimal (with prefix + "0x"). + If ENDPTR is not NULL, the address of the first byte after the integer is + stored in *ENDPTR. + Upon overflow, the return value is ULLONG_MAX, and errno is set to + ERANGE. */ +# if !@HAVE_STRTOULL@ +_GL_FUNCDECL_SYS (strtoull, unsigned long long, + (const char *string, char **endptr, int base) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (strtoull, unsigned long long, + (const char *string, char **endptr, int base)); +_GL_CXXALIASWARN (strtoull); +#elif defined GNULIB_POSIXCHECK +# undef strtoull +# if HAVE_RAW_DECL_STRTOULL +_GL_WARN_ON_USE (strtoull, "strtoull is unportable - " + "use gnulib module strtoull for portability"); +# endif +#endif + +#if @GNULIB_UNLOCKPT@ +/* Unlock the slave side of the pseudo-terminal whose master side is specified + by FD, so that it can be opened. */ +# if !@HAVE_UNLOCKPT@ +_GL_FUNCDECL_SYS (unlockpt, int, (int fd)); +# endif +_GL_CXXALIAS_SYS (unlockpt, int, (int fd)); +_GL_CXXALIASWARN (unlockpt); +#elif defined GNULIB_POSIXCHECK +# undef unlockpt +# if HAVE_RAW_DECL_UNLOCKPT +_GL_WARN_ON_USE (unlockpt, "unlockpt is not portable - " + "use gnulib module unlockpt for portability"); +# endif +#endif + +#if @GNULIB_UNSETENV@ +/* Remove the variable NAME from the environment. */ +# if @REPLACE_UNSETENV@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef unsetenv +# define unsetenv rpl_unsetenv +# endif +_GL_FUNCDECL_RPL (unsetenv, int, (const char *name) _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (unsetenv, int, (const char *name)); +# else +# if !@HAVE_DECL_UNSETENV@ +_GL_FUNCDECL_SYS (unsetenv, int, (const char *name) _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (unsetenv, int, (const char *name)); +# endif +# if !(@REPLACE_UNSETENV@ && !@HAVE_DECL_UNSETENV@) +_GL_CXXALIASWARN (unsetenv); +# endif +#elif defined GNULIB_POSIXCHECK +# undef unsetenv +# if HAVE_RAW_DECL_UNSETENV +_GL_WARN_ON_USE (unsetenv, "unsetenv is unportable - " + "use gnulib module unsetenv for portability"); +# endif +#endif + +/* Convert a wide character to a multibyte character. */ +#if @GNULIB_WCTOMB@ +# if @REPLACE_WCTOMB@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef wctomb +# define wctomb rpl_wctomb +# endif +_GL_FUNCDECL_RPL (wctomb, int, (char *s, wchar_t wc)); +_GL_CXXALIAS_RPL (wctomb, int, (char *s, wchar_t wc)); +# else +_GL_CXXALIAS_SYS (wctomb, int, (char *s, wchar_t wc)); +# endif +_GL_CXXALIASWARN (wctomb); +#endif + + +#endif /* _@GUARD_PREFIX@_STDLIB_H */ +#endif /* _@GUARD_PREFIX@_STDLIB_H */ +#endif diff --git a/lib/strcasecmp.c b/lib/strcasecmp.c new file mode 100644 index 0000000..77ca288 --- /dev/null +++ b/lib/strcasecmp.c @@ -0,0 +1,62 @@ +/* Case-insensitive string comparison function. + Copyright (C) 1998-1999, 2005-2007, 2009-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +/* Specification. */ +#include <string.h> + +#include <ctype.h> +#include <limits.h> + +#define TOLOWER(Ch) (isupper (Ch) ? tolower (Ch) : (Ch)) + +/* Compare strings S1 and S2, ignoring case, returning less than, equal to or + greater than zero if S1 is lexicographically less than, equal to or greater + than S2. + Note: This function does not work with multibyte strings! */ + +int +strcasecmp (const char *s1, const char *s2) +{ + const unsigned char *p1 = (const unsigned char *) s1; + const unsigned char *p2 = (const unsigned char *) s2; + unsigned char c1, c2; + + if (p1 == p2) + return 0; + + do + { + c1 = TOLOWER (*p1); + c2 = TOLOWER (*p2); + + if (c1 == '\0') + break; + + ++p1; + ++p2; + } + while (c1 == c2); + + if (UCHAR_MAX <= INT_MAX) + return c1 - c2; + else + /* On machines where 'char' and 'int' are types of the same size, the + difference of two 'unsigned char' values - including the sign bit - + doesn't fit in an 'int'. */ + return (c1 > c2 ? 1 : c1 < c2 ? -1 : 0); +} diff --git a/lib/streq.h b/lib/streq.h new file mode 100644 index 0000000..b312f7f --- /dev/null +++ b/lib/streq.h @@ -0,0 +1,176 @@ +/* Optimized string comparison. + Copyright (C) 2001-2002, 2007, 2009-2012 Free Software Foundation, Inc. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +/* Written by Bruno Haible <bruno@clisp.org>. */ + +#ifndef _GL_STREQ_H +#define _GL_STREQ_H + +#include <string.h> + +/* STREQ_OPT allows to optimize string comparison with a small literal string. + STREQ_OPT (s, "EUC-KR", 'E', 'U', 'C', '-', 'K', 'R', 0, 0, 0) + is semantically equivalent to + strcmp (s, "EUC-KR") == 0 + just faster. */ + +/* Help GCC to generate good code for string comparisons with + immediate strings. */ +#if defined (__GNUC__) && defined (__OPTIMIZE__) + +static inline int +streq9 (const char *s1, const char *s2) +{ + return strcmp (s1 + 9, s2 + 9) == 0; +} + +static inline int +streq8 (const char *s1, const char *s2, char s28) +{ + if (s1[8] == s28) + { + if (s28 == 0) + return 1; + else + return streq9 (s1, s2); + } + else + return 0; +} + +static inline int +streq7 (const char *s1, const char *s2, char s27, char s28) +{ + if (s1[7] == s27) + { + if (s27 == 0) + return 1; + else + return streq8 (s1, s2, s28); + } + else + return 0; +} + +static inline int +streq6 (const char *s1, const char *s2, char s26, char s27, char s28) +{ + if (s1[6] == s26) + { + if (s26 == 0) + return 1; + else + return streq7 (s1, s2, s27, s28); + } + else + return 0; +} + +static inline int +streq5 (const char *s1, const char *s2, char s25, char s26, char s27, char s28) +{ + if (s1[5] == s25) + { + if (s25 == 0) + return 1; + else + return streq6 (s1, s2, s26, s27, s28); + } + else + return 0; +} + +static inline int +streq4 (const char *s1, const char *s2, char s24, char s25, char s26, char s27, char s28) +{ + if (s1[4] == s24) + { + if (s24 == 0) + return 1; + else + return streq5 (s1, s2, s25, s26, s27, s28); + } + else + return 0; +} + +static inline int +streq3 (const char *s1, const char *s2, char s23, char s24, char s25, char s26, char s27, char s28) +{ + if (s1[3] == s23) + { + if (s23 == 0) + return 1; + else + return streq4 (s1, s2, s24, s25, s26, s27, s28); + } + else + return 0; +} + +static inline int +streq2 (const char *s1, const char *s2, char s22, char s23, char s24, char s25, char s26, char s27, char s28) +{ + if (s1[2] == s22) + { + if (s22 == 0) + return 1; + else + return streq3 (s1, s2, s23, s24, s25, s26, s27, s28); + } + else + return 0; +} + +static inline int +streq1 (const char *s1, const char *s2, char s21, char s22, char s23, char s24, char s25, char s26, char s27, char s28) +{ + if (s1[1] == s21) + { + if (s21 == 0) + return 1; + else + return streq2 (s1, s2, s22, s23, s24, s25, s26, s27, s28); + } + else + return 0; +} + +static inline int +streq0 (const char *s1, const char *s2, char s20, char s21, char s22, char s23, char s24, char s25, char s26, char s27, char s28) +{ + if (s1[0] == s20) + { + if (s20 == 0) + return 1; + else + return streq1 (s1, s2, s21, s22, s23, s24, s25, s26, s27, s28); + } + else + return 0; +} + +#define STREQ_OPT(s1,s2,s20,s21,s22,s23,s24,s25,s26,s27,s28) \ + streq0 (s1, s2, s20, s21, s22, s23, s24, s25, s26, s27, s28) + +#else + +#define STREQ_OPT(s1,s2,s20,s21,s22,s23,s24,s25,s26,s27,s28) \ + (strcmp (s1, s2) == 0) + +#endif + +#endif /* _GL_STREQ_H */ diff --git a/lib/strings.in.h b/lib/strings.in.h new file mode 100644 index 0000000..7ed88e9 --- /dev/null +++ b/lib/strings.in.h @@ -0,0 +1,122 @@ +/* A substitute <strings.h>. + + Copyright (C) 2007-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#ifndef _@GUARD_PREFIX@_STRINGS_H + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +/* Minix 3.1.8 has a bug: <sys/types.h> must be included before <strings.h>. + But avoid namespace pollution on glibc systems. */ +#if defined __minix && !defined __GLIBC__ +# include <sys/types.h> +#endif + +/* The include_next requires a split double-inclusion guard. */ +#if @HAVE_STRINGS_H@ +# @INCLUDE_NEXT@ @NEXT_STRINGS_H@ +#endif + +#ifndef _@GUARD_PREFIX@_STRINGS_H +#define _@GUARD_PREFIX@_STRINGS_H + +#if ! @HAVE_DECL_STRNCASECMP@ +/* Get size_t. */ +# include <stddef.h> +#endif + + +/* The definitions of _GL_FUNCDECL_RPL etc. are copied here. */ + +/* The definition of _GL_ARG_NONNULL is copied here. */ + +/* The definition of _GL_WARN_ON_USE is copied here. */ + +#ifdef __cplusplus +extern "C" { +#endif + + + /* Find the index of the least-significant set bit. */ +#if @GNULIB_FFS@ +# if !@HAVE_FFS@ +_GL_FUNCDECL_SYS (ffs, int, (int i)); +# endif +_GL_CXXALIAS_SYS (ffs, int, (int i)); +_GL_CXXALIASWARN (ffs); +#elif defined GNULIB_POSIXCHECK +# undef ffs +# if HAVE_RAW_DECL_FFS +_GL_WARN_ON_USE (ffs, "ffs is not portable - use the ffs module"); +# endif +#endif + +/* Compare strings S1 and S2, ignoring case, returning less than, equal to or + greater than zero if S1 is lexicographically less than, equal to or greater + than S2. + Note: This function does not work in multibyte locales. */ +#if ! @HAVE_STRCASECMP@ +extern int strcasecmp (char const *s1, char const *s2) + _GL_ARG_NONNULL ((1, 2)); +#endif +#if defined GNULIB_POSIXCHECK +/* strcasecmp() does not work with multibyte strings: + POSIX says that it operates on "strings", and "string" in POSIX is defined + as a sequence of bytes, not of characters. */ +# undef strcasecmp +# if HAVE_RAW_DECL_STRCASECMP +_GL_WARN_ON_USE (strcasecmp, "strcasecmp cannot work correctly on character " + "strings in multibyte locales - " + "use mbscasecmp if you care about " + "internationalization, or use c_strcasecmp , " + "gnulib module c-strcase) if you want a locale " + "independent function"); +# endif +#endif + +/* Compare no more than N bytes of strings S1 and S2, ignoring case, + returning less than, equal to or greater than zero if S1 is + lexicographically less than, equal to or greater than S2. + Note: This function cannot work correctly in multibyte locales. */ +#if ! @HAVE_DECL_STRNCASECMP@ +extern int strncasecmp (char const *s1, char const *s2, size_t n) + _GL_ARG_NONNULL ((1, 2)); +#endif +#if defined GNULIB_POSIXCHECK +/* strncasecmp() does not work with multibyte strings: + POSIX says that it operates on "strings", and "string" in POSIX is defined + as a sequence of bytes, not of characters. */ +# undef strncasecmp +# if HAVE_RAW_DECL_STRNCASECMP +_GL_WARN_ON_USE (strncasecmp, "strncasecmp cannot work correctly on character " + "strings in multibyte locales - " + "use mbsncasecmp or mbspcasecmp if you care about " + "internationalization, or use c_strncasecmp , " + "gnulib module c-strcase) if you want a locale " + "independent function"); +# endif +#endif + + +#ifdef __cplusplus +} +#endif + +#endif /* _@GUARD_PREFIX@_STRING_H */ +#endif /* _@GUARD_PREFIX@_STRING_H */ diff --git a/lib/strncasecmp.c b/lib/strncasecmp.c new file mode 100644 index 0000000..bf413b4 --- /dev/null +++ b/lib/strncasecmp.c @@ -0,0 +1,62 @@ +/* strncasecmp.c -- case insensitive string comparator + Copyright (C) 1998-1999, 2005-2007, 2009-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +/* Specification. */ +#include <string.h> + +#include <ctype.h> +#include <limits.h> + +#define TOLOWER(Ch) (isupper (Ch) ? tolower (Ch) : (Ch)) + +/* Compare no more than N bytes of strings S1 and S2, ignoring case, + returning less than, equal to or greater than zero if S1 is + lexicographically less than, equal to or greater than S2. + Note: This function cannot work correctly in multibyte locales. */ + +int +strncasecmp (const char *s1, const char *s2, size_t n) +{ + register const unsigned char *p1 = (const unsigned char *) s1; + register const unsigned char *p2 = (const unsigned char *) s2; + unsigned char c1, c2; + + if (p1 == p2 || n == 0) + return 0; + + do + { + c1 = TOLOWER (*p1); + c2 = TOLOWER (*p2); + + if (--n == 0 || c1 == '\0') + break; + + ++p1; + ++p2; + } + while (c1 == c2); + + if (UCHAR_MAX <= INT_MAX) + return c1 - c2; + else + /* On machines where 'char' and 'int' are types of the same size, the + difference of two 'unsigned char' values - including the sign bit - + doesn't fit in an 'int'. */ + return (c1 > c2 ? 1 : c1 < c2 ? -1 : 0); +} diff --git a/lib/sys_file.in.h b/lib/sys_file.in.h new file mode 100644 index 0000000..0dc3d5f --- /dev/null +++ b/lib/sys_file.in.h @@ -0,0 +1,63 @@ +/* Provide a more complete sys/file.h. + + Copyright (C) 2007-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +/* Written by Richard W.M. Jones. */ + +#ifndef _@GUARD_PREFIX@_SYS_FILE_H + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +/* The include_next requires a split double-inclusion guard. */ +#if @HAVE_SYS_FILE_H@ +# @INCLUDE_NEXT@ @NEXT_SYS_FILE_H@ +#endif + +#ifndef _@GUARD_PREFIX@_SYS_FILE_H +#define _@GUARD_PREFIX@_SYS_FILE_H + +#ifndef LOCK_SH +/* Operations for the 'flock' call (same as Linux kernel constants). */ +# define LOCK_SH 1 /* Shared lock. */ +# define LOCK_EX 2 /* Exclusive lock. */ +# define LOCK_UN 8 /* Unlock. */ + +/* Can be OR'd in to one of the above. */ +# define LOCK_NB 4 /* Don't block when locking. */ +#endif + +/* The definition of _GL_WARN_ON_USE is copied here. */ + +#if @GNULIB_FLOCK@ +/* Apply or remove advisory locks on an open file. + Return 0 if successful, otherwise -1 and errno set. */ +# if !@HAVE_FLOCK@ +extern int flock (int fd, int operation); +# endif +#elif defined GNULIB_POSIXCHECK +# undef flock +# if HAVE_RAW_DECL_FLOCK +_GL_WARN_ON_USE (flock, "flock is unportable - " + "use gnulib module flock for portability"); +# endif +#endif + + +#endif /* _@GUARD_PREFIX@_SYS_FILE_H */ +#endif /* _@GUARD_PREFIX@_SYS_FILE_H */ diff --git a/lib/sys_types.in.h b/lib/sys_types.in.h new file mode 100644 index 0000000..574fe51 --- /dev/null +++ b/lib/sys_types.in.h @@ -0,0 +1,39 @@ +/* Provide a more complete sys/types.h. + + Copyright (C) 2011-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +#ifndef _@GUARD_PREFIX@_SYS_TYPES_H + +/* The include_next requires a split double-inclusion guard. */ +#@INCLUDE_NEXT@ @NEXT_SYS_TYPES_H@ + +#ifndef _@GUARD_PREFIX@_SYS_TYPES_H +#define _@GUARD_PREFIX@_SYS_TYPES_H + +/* MSVC 9 defines size_t in <stddef.h>, not in <sys/types.h>. */ +/* But avoid namespace pollution on glibc systems. */ +#if ((defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__) \ + && ! defined __GLIBC__ +# include <stddef.h> +#endif + +#endif /* _@GUARD_PREFIX@_SYS_TYPES_H */ +#endif /* _@GUARD_PREFIX@_SYS_TYPES_H */ diff --git a/lib/unistd.in.h b/lib/unistd.in.h new file mode 100644 index 0000000..1b26c09 --- /dev/null +++ b/lib/unistd.in.h @@ -0,0 +1,1525 @@ +/* Substitute for and wrapper around <unistd.h>. + Copyright (C) 2003-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +/* Special invocation convention: + - On mingw, several headers, including <winsock2.h>, include <unistd.h>, + but we need to ensure that both the system <unistd.h> and <winsock2.h> + are completely included before we replace gethostname. */ +#if @GNULIB_GETHOSTNAME@ && @UNISTD_H_HAVE_WINSOCK2_H@ \ + && !defined _GL_WINSOCK2_H_WITNESS && defined _WINSOCK2_H +/* <unistd.h> is being indirectly included for the first time from + <winsock2.h>; avoid declaring any overrides. */ +# if @HAVE_UNISTD_H@ +# @INCLUDE_NEXT@ @NEXT_UNISTD_H@ +# else +# error unexpected; report this to bug-gnulib@gnu.org +# endif +# define _GL_WINSOCK2_H_WITNESS + +/* Normal invocation. */ +#elif !defined _@GUARD_PREFIX@_UNISTD_H + +/* The include_next requires a split double-inclusion guard. */ +#if @HAVE_UNISTD_H@ +# @INCLUDE_NEXT@ @NEXT_UNISTD_H@ +#endif + +/* Get all possible declarations of gethostname(). */ +#if @GNULIB_GETHOSTNAME@ && @UNISTD_H_HAVE_WINSOCK2_H@ \ + && !defined _GL_INCLUDING_WINSOCK2_H +# define _GL_INCLUDING_WINSOCK2_H +# include <winsock2.h> +# undef _GL_INCLUDING_WINSOCK2_H +#endif + +#if !defined _@GUARD_PREFIX@_UNISTD_H && !defined _GL_INCLUDING_WINSOCK2_H +#define _@GUARD_PREFIX@_UNISTD_H + +/* NetBSD 5.0 mis-defines NULL. Also get size_t. */ +#include <stddef.h> + +/* mingw doesn't define the SEEK_* or *_FILENO macros in <unistd.h>. */ +/* Cygwin 1.7.1 declares symlinkat in <stdio.h>, not in <unistd.h>. */ +/* But avoid namespace pollution on glibc systems. */ +#if (!(defined SEEK_CUR && defined SEEK_END && defined SEEK_SET) \ + || ((@GNULIB_SYMLINKAT@ || defined GNULIB_POSIXCHECK) \ + && defined __CYGWIN__)) \ + && ! defined __GLIBC__ +# include <stdio.h> +#endif + +/* Cygwin 1.7.1 declares unlinkat in <fcntl.h>, not in <unistd.h>. */ +/* But avoid namespace pollution on glibc systems. */ +#if (@GNULIB_UNLINKAT@ || defined GNULIB_POSIXCHECK) && defined __CYGWIN__ \ + && ! defined __GLIBC__ +# include <fcntl.h> +#endif + +/* mingw fails to declare _exit in <unistd.h>. */ +/* mingw, MSVC, BeOS, Haiku declare environ in <stdlib.h>, not in + <unistd.h>. */ +/* Solaris declares getcwd not only in <unistd.h> but also in <stdlib.h>. */ +/* But avoid namespace pollution on glibc systems. */ +#ifndef __GLIBC__ +# include <stdlib.h> +#endif + +/* Native Windows platforms declare chdir, getcwd, rmdir in + <io.h> and/or <direct.h>, not in <unistd.h>. + They also declare access(), chmod(), close(), dup(), dup2(), isatty(), + lseek(), read(), unlink(), write() in <io.h>. */ +#if ((@GNULIB_CHDIR@ || @GNULIB_GETCWD@ || @GNULIB_RMDIR@ \ + || defined GNULIB_POSIXCHECK) \ + && ((defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__)) +# include <io.h> /* mingw32, mingw64 */ +# include <direct.h> /* mingw64, MSVC 9 */ +#elif (@GNULIB_CLOSE@ || @GNULIB_DUP@ || @GNULIB_DUP2@ || @GNULIB_ISATTY@ \ + || @GNULIB_LSEEK@ || @GNULIB_READ@ || @GNULIB_UNLINK@ || @GNULIB_WRITE@ \ + || defined GNULIB_POSIXCHECK) \ + && ((defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__) +# include <io.h> +#endif + +/* AIX and OSF/1 5.1 declare getdomainname in <netdb.h>, not in <unistd.h>. + NonStop Kernel declares gethostname in <netdb.h>, not in <unistd.h>. */ +/* But avoid namespace pollution on glibc systems. */ +#if ((@GNULIB_GETDOMAINNAME@ && (defined _AIX || defined __osf__)) \ + || (@GNULIB_GETHOSTNAME@ && defined __TANDEM)) \ + && !defined __GLIBC__ +# include <netdb.h> +#endif + +/* MSVC defines off_t in <sys/types.h>. */ +#if !@HAVE_UNISTD_H@ +/* Get off_t. */ +# include <sys/types.h> +#endif + +#if (@GNULIB_READ@ || @GNULIB_WRITE@ \ + || @GNULIB_READLINK@ || @GNULIB_READLINKAT@ \ + || @GNULIB_PREAD@ || @GNULIB_PWRITE@ || defined GNULIB_POSIXCHECK) +/* Get ssize_t. */ +# include <sys/types.h> +#endif + +/* Get getopt(), optarg, optind, opterr, optopt. + But avoid namespace pollution on glibc systems. */ +#if @GNULIB_UNISTD_H_GETOPT@ && !defined __GLIBC__ && !defined _GL_SYSTEM_GETOPT +# include <getopt.h> +#endif + +/* The definitions of _GL_FUNCDECL_RPL etc. are copied here. */ + +/* The definition of _GL_ARG_NONNULL is copied here. */ + +/* The definition of _GL_WARN_ON_USE is copied here. */ + + +/* Hide some function declarations from <winsock2.h>. */ + +#if @GNULIB_GETHOSTNAME@ && @UNISTD_H_HAVE_WINSOCK2_H@ +# if !defined _@GUARD_PREFIX@_SYS_SOCKET_H +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef socket +# define socket socket_used_without_including_sys_socket_h +# undef connect +# define connect connect_used_without_including_sys_socket_h +# undef accept +# define accept accept_used_without_including_sys_socket_h +# undef bind +# define bind bind_used_without_including_sys_socket_h +# undef getpeername +# define getpeername getpeername_used_without_including_sys_socket_h +# undef getsockname +# define getsockname getsockname_used_without_including_sys_socket_h +# undef getsockopt +# define getsockopt getsockopt_used_without_including_sys_socket_h +# undef listen +# define listen listen_used_without_including_sys_socket_h +# undef recv +# define recv recv_used_without_including_sys_socket_h +# undef send +# define send send_used_without_including_sys_socket_h +# undef recvfrom +# define recvfrom recvfrom_used_without_including_sys_socket_h +# undef sendto +# define sendto sendto_used_without_including_sys_socket_h +# undef setsockopt +# define setsockopt setsockopt_used_without_including_sys_socket_h +# undef shutdown +# define shutdown shutdown_used_without_including_sys_socket_h +# else + _GL_WARN_ON_USE (socket, + "socket() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (connect, + "connect() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (accept, + "accept() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (bind, + "bind() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (getpeername, + "getpeername() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (getsockname, + "getsockname() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (getsockopt, + "getsockopt() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (listen, + "listen() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (recv, + "recv() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (send, + "send() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (recvfrom, + "recvfrom() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (sendto, + "sendto() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (setsockopt, + "setsockopt() used without including <sys/socket.h>"); + _GL_WARN_ON_USE (shutdown, + "shutdown() used without including <sys/socket.h>"); +# endif +# endif +# if !defined _@GUARD_PREFIX@_SYS_SELECT_H +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef select +# define select select_used_without_including_sys_select_h +# else + _GL_WARN_ON_USE (select, + "select() used without including <sys/select.h>"); +# endif +# endif +#endif + + +/* OS/2 EMX lacks these macros. */ +#ifndef STDIN_FILENO +# define STDIN_FILENO 0 +#endif +#ifndef STDOUT_FILENO +# define STDOUT_FILENO 1 +#endif +#ifndef STDERR_FILENO +# define STDERR_FILENO 2 +#endif + +/* Ensure *_OK macros exist. */ +#ifndef F_OK +# define F_OK 0 +# define X_OK 1 +# define W_OK 2 +# define R_OK 4 +#endif + + +/* Declare overridden functions. */ + + +#if defined GNULIB_POSIXCHECK +/* The access() function is a security risk. */ +_GL_WARN_ON_USE (access, "the access function is a security risk - " + "use the gnulib module faccessat instead"); +#endif + + +#if @GNULIB_CHDIR@ +_GL_CXXALIAS_SYS (chdir, int, (const char *file) _GL_ARG_NONNULL ((1))); +_GL_CXXALIASWARN (chdir); +#elif defined GNULIB_POSIXCHECK +# undef chdir +# if HAVE_RAW_DECL_CHDIR +_GL_WARN_ON_USE (chown, "chdir is not always in <unistd.h> - " + "use gnulib module chdir for portability"); +# endif +#endif + + +#if @GNULIB_CHOWN@ +/* Change the owner of FILE to UID (if UID is not -1) and the group of FILE + to GID (if GID is not -1). Follow symbolic links. + Return 0 if successful, otherwise -1 and errno set. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/chown.html. */ +# if @REPLACE_CHOWN@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef chown +# define chown rpl_chown +# endif +_GL_FUNCDECL_RPL (chown, int, (const char *file, uid_t uid, gid_t gid) + _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (chown, int, (const char *file, uid_t uid, gid_t gid)); +# else +# if !@HAVE_CHOWN@ +_GL_FUNCDECL_SYS (chown, int, (const char *file, uid_t uid, gid_t gid) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (chown, int, (const char *file, uid_t uid, gid_t gid)); +# endif +_GL_CXXALIASWARN (chown); +#elif defined GNULIB_POSIXCHECK +# undef chown +# if HAVE_RAW_DECL_CHOWN +_GL_WARN_ON_USE (chown, "chown fails to follow symlinks on some systems and " + "doesn't treat a uid or gid of -1 on some systems - " + "use gnulib module chown for portability"); +# endif +#endif + + +#if @GNULIB_CLOSE@ +# if @REPLACE_CLOSE@ +/* Automatically included by modules that need a replacement for close. */ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef close +# define close rpl_close +# endif +_GL_FUNCDECL_RPL (close, int, (int fd)); +_GL_CXXALIAS_RPL (close, int, (int fd)); +# else +_GL_CXXALIAS_SYS (close, int, (int fd)); +# endif +_GL_CXXALIASWARN (close); +#elif @UNISTD_H_HAVE_WINSOCK2_H_AND_USE_SOCKETS@ +# undef close +# define close close_used_without_requesting_gnulib_module_close +#elif defined GNULIB_POSIXCHECK +# undef close +/* Assume close is always declared. */ +_GL_WARN_ON_USE (close, "close does not portably work on sockets - " + "use gnulib module close for portability"); +#endif + + +#if @GNULIB_DUP@ +# if @REPLACE_DUP@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define dup rpl_dup +# endif +_GL_FUNCDECL_RPL (dup, int, (int oldfd)); +_GL_CXXALIAS_RPL (dup, int, (int oldfd)); +# else +_GL_CXXALIAS_SYS (dup, int, (int oldfd)); +# endif +_GL_CXXALIASWARN (dup); +#elif defined GNULIB_POSIXCHECK +# undef dup +# if HAVE_RAW_DECL_DUP +_GL_WARN_ON_USE (dup, "dup is unportable - " + "use gnulib module dup for portability"); +# endif +#endif + + +#if @GNULIB_DUP2@ +/* Copy the file descriptor OLDFD into file descriptor NEWFD. Do nothing if + NEWFD = OLDFD, otherwise close NEWFD first if it is open. + Return newfd if successful, otherwise -1 and errno set. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/dup2.html>. */ +# if @REPLACE_DUP2@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define dup2 rpl_dup2 +# endif +_GL_FUNCDECL_RPL (dup2, int, (int oldfd, int newfd)); +_GL_CXXALIAS_RPL (dup2, int, (int oldfd, int newfd)); +# else +# if !@HAVE_DUP2@ +_GL_FUNCDECL_SYS (dup2, int, (int oldfd, int newfd)); +# endif +_GL_CXXALIAS_SYS (dup2, int, (int oldfd, int newfd)); +# endif +_GL_CXXALIASWARN (dup2); +#elif defined GNULIB_POSIXCHECK +# undef dup2 +# if HAVE_RAW_DECL_DUP2 +_GL_WARN_ON_USE (dup2, "dup2 is unportable - " + "use gnulib module dup2 for portability"); +# endif +#endif + + +#if @GNULIB_DUP3@ +/* Copy the file descriptor OLDFD into file descriptor NEWFD, with the + specified flags. + The flags are a bitmask, possibly including O_CLOEXEC (defined in <fcntl.h>) + and O_TEXT, O_BINARY (defined in "binary-io.h"). + Close NEWFD first if it is open. + Return newfd if successful, otherwise -1 and errno set. + See the Linux man page at + <http://www.kernel.org/doc/man-pages/online/pages/man2/dup3.2.html>. */ +# if @HAVE_DUP3@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define dup3 rpl_dup3 +# endif +_GL_FUNCDECL_RPL (dup3, int, (int oldfd, int newfd, int flags)); +_GL_CXXALIAS_RPL (dup3, int, (int oldfd, int newfd, int flags)); +# else +_GL_FUNCDECL_SYS (dup3, int, (int oldfd, int newfd, int flags)); +_GL_CXXALIAS_SYS (dup3, int, (int oldfd, int newfd, int flags)); +# endif +_GL_CXXALIASWARN (dup3); +#elif defined GNULIB_POSIXCHECK +# undef dup3 +# if HAVE_RAW_DECL_DUP3 +_GL_WARN_ON_USE (dup3, "dup3 is unportable - " + "use gnulib module dup3 for portability"); +# endif +#endif + + +#if @GNULIB_ENVIRON@ +# if !@HAVE_DECL_ENVIRON@ +/* Set of environment variables and values. An array of strings of the form + "VARIABLE=VALUE", terminated with a NULL. */ +# if defined __APPLE__ && defined __MACH__ +# include <crt_externs.h> +# define environ (*_NSGetEnviron ()) +# else +# ifdef __cplusplus +extern "C" { +# endif +extern char **environ; +# ifdef __cplusplus +} +# endif +# endif +# endif +#elif defined GNULIB_POSIXCHECK +# if HAVE_RAW_DECL_ENVIRON +static inline char *** +rpl_environ (void) +{ + return &environ; +} +_GL_WARN_ON_USE (rpl_environ, "environ is unportable - " + "use gnulib module environ for portability"); +# undef environ +# define environ (*rpl_environ ()) +# endif +#endif + + +#if @GNULIB_EUIDACCESS@ +/* Like access(), except that it uses the effective user id and group id of + the current process. */ +# if !@HAVE_EUIDACCESS@ +_GL_FUNCDECL_SYS (euidaccess, int, (const char *filename, int mode) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (euidaccess, int, (const char *filename, int mode)); +_GL_CXXALIASWARN (euidaccess); +# if defined GNULIB_POSIXCHECK +/* Like access(), this function is a security risk. */ +_GL_WARN_ON_USE (euidaccess, "the euidaccess function is a security risk - " + "use the gnulib module faccessat instead"); +# endif +#elif defined GNULIB_POSIXCHECK +# undef euidaccess +# if HAVE_RAW_DECL_EUIDACCESS +_GL_WARN_ON_USE (euidaccess, "euidaccess is unportable - " + "use gnulib module euidaccess for portability"); +# endif +#endif + + +#if @GNULIB_FACCESSAT@ +# if !@HAVE_FACCESSAT@ +_GL_FUNCDECL_SYS (faccessat, int, + (int fd, char const *file, int mode, int flag) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (faccessat, int, + (int fd, char const *file, int mode, int flag)); +_GL_CXXALIASWARN (faccessat); +#elif defined GNULIB_POSIXCHECK +# undef faccessat +# if HAVE_RAW_DECL_FACCESSAT +_GL_WARN_ON_USE (faccessat, "faccessat is not portable - " + "use gnulib module faccessat for portability"); +# endif +#endif + + +#if @GNULIB_FCHDIR@ +/* Change the process' current working directory to the directory on which + the given file descriptor is open. + Return 0 if successful, otherwise -1 and errno set. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/fchdir.html>. */ +# if ! @HAVE_FCHDIR@ +_GL_FUNCDECL_SYS (fchdir, int, (int /*fd*/)); + +/* Gnulib internal hooks needed to maintain the fchdir metadata. */ +_GL_EXTERN_C int _gl_register_fd (int fd, const char *filename) + _GL_ARG_NONNULL ((2)); +_GL_EXTERN_C void _gl_unregister_fd (int fd); +_GL_EXTERN_C int _gl_register_dup (int oldfd, int newfd); +_GL_EXTERN_C const char *_gl_directory_name (int fd); + +# else +# if !@HAVE_DECL_FCHDIR@ +_GL_FUNCDECL_SYS (fchdir, int, (int /*fd*/)); +# endif +# endif +_GL_CXXALIAS_SYS (fchdir, int, (int /*fd*/)); +_GL_CXXALIASWARN (fchdir); +#elif defined GNULIB_POSIXCHECK +# undef fchdir +# if HAVE_RAW_DECL_FCHDIR +_GL_WARN_ON_USE (fchdir, "fchdir is unportable - " + "use gnulib module fchdir for portability"); +# endif +#endif + + +#if @GNULIB_FCHOWNAT@ +# if @REPLACE_FCHOWNAT@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef fchownat +# define fchownat rpl_fchownat +# endif +_GL_FUNCDECL_RPL (fchownat, int, (int fd, char const *file, + uid_t owner, gid_t group, int flag) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (fchownat, int, (int fd, char const *file, + uid_t owner, gid_t group, int flag)); +# else +# if !@HAVE_FCHOWNAT@ +_GL_FUNCDECL_SYS (fchownat, int, (int fd, char const *file, + uid_t owner, gid_t group, int flag) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (fchownat, int, (int fd, char const *file, + uid_t owner, gid_t group, int flag)); +# endif +_GL_CXXALIASWARN (fchownat); +#elif defined GNULIB_POSIXCHECK +# undef fchownat +# if HAVE_RAW_DECL_FCHOWNAT +_GL_WARN_ON_USE (fchownat, "fchownat is not portable - " + "use gnulib module openat for portability"); +# endif +#endif + + +#if @GNULIB_FDATASYNC@ +/* Synchronize changes to a file. + Return 0 if successful, otherwise -1 and errno set. + See POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/fdatasync.html>. */ +# if !@HAVE_FDATASYNC@ || !@HAVE_DECL_FDATASYNC@ +_GL_FUNCDECL_SYS (fdatasync, int, (int fd)); +# endif +_GL_CXXALIAS_SYS (fdatasync, int, (int fd)); +_GL_CXXALIASWARN (fdatasync); +#elif defined GNULIB_POSIXCHECK +# undef fdatasync +# if HAVE_RAW_DECL_FDATASYNC +_GL_WARN_ON_USE (fdatasync, "fdatasync is unportable - " + "use gnulib module fdatasync for portability"); +# endif +#endif + + +#if @GNULIB_FSYNC@ +/* Synchronize changes, including metadata, to a file. + Return 0 if successful, otherwise -1 and errno set. + See POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/fsync.html>. */ +# if !@HAVE_FSYNC@ +_GL_FUNCDECL_SYS (fsync, int, (int fd)); +# endif +_GL_CXXALIAS_SYS (fsync, int, (int fd)); +_GL_CXXALIASWARN (fsync); +#elif defined GNULIB_POSIXCHECK +# undef fsync +# if HAVE_RAW_DECL_FSYNC +_GL_WARN_ON_USE (fsync, "fsync is unportable - " + "use gnulib module fsync for portability"); +# endif +#endif + + +#if @GNULIB_FTRUNCATE@ +/* Change the size of the file to which FD is opened to become equal to LENGTH. + Return 0 if successful, otherwise -1 and errno set. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/ftruncate.html>. */ +# if !@HAVE_FTRUNCATE@ +_GL_FUNCDECL_SYS (ftruncate, int, (int fd, off_t length)); +# endif +_GL_CXXALIAS_SYS (ftruncate, int, (int fd, off_t length)); +_GL_CXXALIASWARN (ftruncate); +#elif defined GNULIB_POSIXCHECK +# undef ftruncate +# if HAVE_RAW_DECL_FTRUNCATE +_GL_WARN_ON_USE (ftruncate, "ftruncate is unportable - " + "use gnulib module ftruncate for portability"); +# endif +#endif + + +#if @GNULIB_GETCWD@ +/* Get the name of the current working directory, and put it in SIZE bytes + of BUF. + Return BUF if successful, or NULL if the directory couldn't be determined + or SIZE was too small. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/getcwd.html>. + Additionally, the gnulib module 'getcwd' guarantees the following GNU + extension: If BUF is NULL, an array is allocated with 'malloc'; the array + is SIZE bytes long, unless SIZE == 0, in which case it is as big as + necessary. */ +# if @REPLACE_GETCWD@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define getcwd rpl_getcwd +# endif +_GL_FUNCDECL_RPL (getcwd, char *, (char *buf, size_t size)); +_GL_CXXALIAS_RPL (getcwd, char *, (char *buf, size_t size)); +# else +/* Need to cast, because on mingw, the second parameter is + int size. */ +_GL_CXXALIAS_SYS_CAST (getcwd, char *, (char *buf, size_t size)); +# endif +_GL_CXXALIASWARN (getcwd); +#elif defined GNULIB_POSIXCHECK +# undef getcwd +# if HAVE_RAW_DECL_GETCWD +_GL_WARN_ON_USE (getcwd, "getcwd is unportable - " + "use gnulib module getcwd for portability"); +# endif +#endif + + +#if @GNULIB_GETDOMAINNAME@ +/* Return the NIS domain name of the machine. + WARNING! The NIS domain name is unrelated to the fully qualified host name + of the machine. It is also unrelated to email addresses. + WARNING! The NIS domain name is usually the empty string or "(none)" when + not using NIS. + + Put up to LEN bytes of the NIS domain name into NAME. + Null terminate it if the name is shorter than LEN. + If the NIS domain name is longer than LEN, set errno = EINVAL and return -1. + Return 0 if successful, otherwise set errno and return -1. */ +# if @REPLACE_GETDOMAINNAME@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef getdomainname +# define getdomainname rpl_getdomainname +# endif +_GL_FUNCDECL_RPL (getdomainname, int, (char *name, size_t len) + _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (getdomainname, int, (char *name, size_t len)); +# else +# if !@HAVE_DECL_GETDOMAINNAME@ +_GL_FUNCDECL_SYS (getdomainname, int, (char *name, size_t len) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (getdomainname, int, (char *name, size_t len)); +# endif +_GL_CXXALIASWARN (getdomainname); +#elif defined GNULIB_POSIXCHECK +# undef getdomainname +# if HAVE_RAW_DECL_GETDOMAINNAME +_GL_WARN_ON_USE (getdomainname, "getdomainname is unportable - " + "use gnulib module getdomainname for portability"); +# endif +#endif + + +#if @GNULIB_GETDTABLESIZE@ +/* Return the maximum number of file descriptors in the current process. + In POSIX, this is same as sysconf (_SC_OPEN_MAX). */ +# if !@HAVE_GETDTABLESIZE@ +_GL_FUNCDECL_SYS (getdtablesize, int, (void)); +# endif +_GL_CXXALIAS_SYS (getdtablesize, int, (void)); +_GL_CXXALIASWARN (getdtablesize); +#elif defined GNULIB_POSIXCHECK +# undef getdtablesize +# if HAVE_RAW_DECL_GETDTABLESIZE +_GL_WARN_ON_USE (getdtablesize, "getdtablesize is unportable - " + "use gnulib module getdtablesize for portability"); +# endif +#endif + + +#if @GNULIB_GETGROUPS@ +/* Return the supplemental groups that the current process belongs to. + It is unspecified whether the effective group id is in the list. + If N is 0, return the group count; otherwise, N describes how many + entries are available in GROUPS. Return -1 and set errno if N is + not 0 and not large enough. Fails with ENOSYS on some systems. */ +# if @REPLACE_GETGROUPS@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef getgroups +# define getgroups rpl_getgroups +# endif +_GL_FUNCDECL_RPL (getgroups, int, (int n, gid_t *groups)); +_GL_CXXALIAS_RPL (getgroups, int, (int n, gid_t *groups)); +# else +# if !@HAVE_GETGROUPS@ +_GL_FUNCDECL_SYS (getgroups, int, (int n, gid_t *groups)); +# endif +_GL_CXXALIAS_SYS (getgroups, int, (int n, gid_t *groups)); +# endif +_GL_CXXALIASWARN (getgroups); +#elif defined GNULIB_POSIXCHECK +# undef getgroups +# if HAVE_RAW_DECL_GETGROUPS +_GL_WARN_ON_USE (getgroups, "getgroups is unportable - " + "use gnulib module getgroups for portability"); +# endif +#endif + + +#if @GNULIB_GETHOSTNAME@ +/* Return the standard host name of the machine. + WARNING! The host name may or may not be fully qualified. + + Put up to LEN bytes of the host name into NAME. + Null terminate it if the name is shorter than LEN. + If the host name is longer than LEN, set errno = EINVAL and return -1. + Return 0 if successful, otherwise set errno and return -1. */ +# if @UNISTD_H_HAVE_WINSOCK2_H@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef gethostname +# define gethostname rpl_gethostname +# endif +_GL_FUNCDECL_RPL (gethostname, int, (char *name, size_t len) + _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (gethostname, int, (char *name, size_t len)); +# else +# if !@HAVE_GETHOSTNAME@ +_GL_FUNCDECL_SYS (gethostname, int, (char *name, size_t len) + _GL_ARG_NONNULL ((1))); +# endif +/* Need to cast, because on Solaris 10 and OSF/1 5.1 systems, the second + parameter is + int len. */ +_GL_CXXALIAS_SYS_CAST (gethostname, int, (char *name, size_t len)); +# endif +_GL_CXXALIASWARN (gethostname); +#elif @UNISTD_H_HAVE_WINSOCK2_H@ +# undef gethostname +# define gethostname gethostname_used_without_requesting_gnulib_module_gethostname +#elif defined GNULIB_POSIXCHECK +# undef gethostname +# if HAVE_RAW_DECL_GETHOSTNAME +_GL_WARN_ON_USE (gethostname, "gethostname is unportable - " + "use gnulib module gethostname for portability"); +# endif +#endif + + +#if @GNULIB_GETLOGIN@ +/* Returns the user's login name, or NULL if it cannot be found. Upon error, + returns NULL with errno set. + + See <http://www.opengroup.org/susv3xsh/getlogin.html>. + + Most programs don't need to use this function, because the information is + available through environment variables: + ${LOGNAME-$USER} on Unix platforms, + $USERNAME on native Windows platforms. + */ +# if !@HAVE_GETLOGIN@ +_GL_FUNCDECL_SYS (getlogin, char *, (void)); +# endif +_GL_CXXALIAS_SYS (getlogin, char *, (void)); +_GL_CXXALIASWARN (getlogin); +#elif defined GNULIB_POSIXCHECK +# undef getlogin +# if HAVE_RAW_DECL_GETLOGIN +_GL_WARN_ON_USE (getlogin, "getlogin is unportable - " + "use gnulib module getlogin for portability"); +# endif +#endif + + +#if @GNULIB_GETLOGIN_R@ +/* Copies the user's login name to NAME. + The array pointed to by NAME has room for SIZE bytes. + + Returns 0 if successful. Upon error, an error number is returned, or -1 in + the case that the login name cannot be found but no specific error is + provided (this case is hopefully rare but is left open by the POSIX spec). + + See <http://www.opengroup.org/susv3xsh/getlogin.html>. + + Most programs don't need to use this function, because the information is + available through environment variables: + ${LOGNAME-$USER} on Unix platforms, + $USERNAME on native Windows platforms. + */ +# if @REPLACE_GETLOGIN_R@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define getlogin_r rpl_getlogin_r +# endif +_GL_FUNCDECL_RPL (getlogin_r, int, (char *name, size_t size) + _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (getlogin_r, int, (char *name, size_t size)); +# else +# if !@HAVE_DECL_GETLOGIN_R@ +_GL_FUNCDECL_SYS (getlogin_r, int, (char *name, size_t size) + _GL_ARG_NONNULL ((1))); +# endif +/* Need to cast, because on Solaris 10 systems, the second argument is + int size. */ +_GL_CXXALIAS_SYS_CAST (getlogin_r, int, (char *name, size_t size)); +# endif +_GL_CXXALIASWARN (getlogin_r); +#elif defined GNULIB_POSIXCHECK +# undef getlogin_r +# if HAVE_RAW_DECL_GETLOGIN_R +_GL_WARN_ON_USE (getlogin_r, "getlogin_r is unportable - " + "use gnulib module getlogin_r for portability"); +# endif +#endif + + +#if @GNULIB_GETPAGESIZE@ +# if @REPLACE_GETPAGESIZE@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define getpagesize rpl_getpagesize +# endif +_GL_FUNCDECL_RPL (getpagesize, int, (void)); +_GL_CXXALIAS_RPL (getpagesize, int, (void)); +# else +# if !@HAVE_GETPAGESIZE@ +# if !defined getpagesize +/* This is for POSIX systems. */ +# if !defined _gl_getpagesize && defined _SC_PAGESIZE +# if ! (defined __VMS && __VMS_VER < 70000000) +# define _gl_getpagesize() sysconf (_SC_PAGESIZE) +# endif +# endif +/* This is for older VMS. */ +# if !defined _gl_getpagesize && defined __VMS +# ifdef __ALPHA +# define _gl_getpagesize() 8192 +# else +# define _gl_getpagesize() 512 +# endif +# endif +/* This is for BeOS. */ +# if !defined _gl_getpagesize && @HAVE_OS_H@ +# include <OS.h> +# if defined B_PAGE_SIZE +# define _gl_getpagesize() B_PAGE_SIZE +# endif +# endif +/* This is for AmigaOS4.0. */ +# if !defined _gl_getpagesize && defined __amigaos4__ +# define _gl_getpagesize() 2048 +# endif +/* This is for older Unix systems. */ +# if !defined _gl_getpagesize && @HAVE_SYS_PARAM_H@ +# include <sys/param.h> +# ifdef EXEC_PAGESIZE +# define _gl_getpagesize() EXEC_PAGESIZE +# else +# ifdef NBPG +# ifndef CLSIZE +# define CLSIZE 1 +# endif +# define _gl_getpagesize() (NBPG * CLSIZE) +# else +# ifdef NBPC +# define _gl_getpagesize() NBPC +# endif +# endif +# endif +# endif +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define getpagesize() _gl_getpagesize () +# else +# if !GNULIB_defined_getpagesize_function +static inline int +getpagesize () +{ + return _gl_getpagesize (); +} +# define GNULIB_defined_getpagesize_function 1 +# endif +# endif +# endif +# endif +/* Need to cast, because on Cygwin 1.5.x systems, the return type is size_t. */ +_GL_CXXALIAS_SYS_CAST (getpagesize, int, (void)); +# endif +# if @HAVE_DECL_GETPAGESIZE@ +_GL_CXXALIASWARN (getpagesize); +# endif +#elif defined GNULIB_POSIXCHECK +# undef getpagesize +# if HAVE_RAW_DECL_GETPAGESIZE +_GL_WARN_ON_USE (getpagesize, "getpagesize is unportable - " + "use gnulib module getpagesize for portability"); +# endif +#endif + + +#if @GNULIB_GETUSERSHELL@ +/* Return the next valid login shell on the system, or NULL when the end of + the list has been reached. */ +# if !@HAVE_DECL_GETUSERSHELL@ +_GL_FUNCDECL_SYS (getusershell, char *, (void)); +# endif +_GL_CXXALIAS_SYS (getusershell, char *, (void)); +_GL_CXXALIASWARN (getusershell); +#elif defined GNULIB_POSIXCHECK +# undef getusershell +# if HAVE_RAW_DECL_GETUSERSHELL +_GL_WARN_ON_USE (getusershell, "getusershell is unportable - " + "use gnulib module getusershell for portability"); +# endif +#endif + +#if @GNULIB_GETUSERSHELL@ +/* Rewind to pointer that is advanced at each getusershell() call. */ +# if !@HAVE_DECL_GETUSERSHELL@ +_GL_FUNCDECL_SYS (setusershell, void, (void)); +# endif +_GL_CXXALIAS_SYS (setusershell, void, (void)); +_GL_CXXALIASWARN (setusershell); +#elif defined GNULIB_POSIXCHECK +# undef setusershell +# if HAVE_RAW_DECL_SETUSERSHELL +_GL_WARN_ON_USE (setusershell, "setusershell is unportable - " + "use gnulib module getusershell for portability"); +# endif +#endif + +#if @GNULIB_GETUSERSHELL@ +/* Free the pointer that is advanced at each getusershell() call and + associated resources. */ +# if !@HAVE_DECL_GETUSERSHELL@ +_GL_FUNCDECL_SYS (endusershell, void, (void)); +# endif +_GL_CXXALIAS_SYS (endusershell, void, (void)); +_GL_CXXALIASWARN (endusershell); +#elif defined GNULIB_POSIXCHECK +# undef endusershell +# if HAVE_RAW_DECL_ENDUSERSHELL +_GL_WARN_ON_USE (endusershell, "endusershell is unportable - " + "use gnulib module getusershell for portability"); +# endif +#endif + + +#if @GNULIB_GROUP_MEMBER@ +/* Determine whether group id is in calling user's group list. */ +# if !@HAVE_GROUP_MEMBER@ +_GL_FUNCDECL_SYS (group_member, int, (gid_t gid)); +# endif +_GL_CXXALIAS_SYS (group_member, int, (gid_t gid)); +_GL_CXXALIASWARN (group_member); +#elif defined GNULIB_POSIXCHECK +# undef group_member +# if HAVE_RAW_DECL_GROUP_MEMBER +_GL_WARN_ON_USE (group_member, "group_member is unportable - " + "use gnulib module group-member for portability"); +# endif +#endif + + +#if @GNULIB_ISATTY@ +# if @REPLACE_ISATTY@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef isatty +# define isatty rpl_isatty +# endif +_GL_FUNCDECL_RPL (isatty, int, (int fd)); +_GL_CXXALIAS_RPL (isatty, int, (int fd)); +# else +_GL_CXXALIAS_SYS (isatty, int, (int fd)); +# endif +_GL_CXXALIASWARN (isatty); +#elif defined GNULIB_POSIXCHECK +# undef isatty +# if HAVE_RAW_DECL_ISATTY +_GL_WARN_ON_USE (isatty, "isatty has portability problems on native Windows - " + "use gnulib module isatty for portability"); +# endif +#endif + + +#if @GNULIB_LCHOWN@ +/* Change the owner of FILE to UID (if UID is not -1) and the group of FILE + to GID (if GID is not -1). Do not follow symbolic links. + Return 0 if successful, otherwise -1 and errno set. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/lchown.html>. */ +# if @REPLACE_LCHOWN@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef lchown +# define lchown rpl_lchown +# endif +_GL_FUNCDECL_RPL (lchown, int, (char const *file, uid_t owner, gid_t group) + _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (lchown, int, (char const *file, uid_t owner, gid_t group)); +# else +# if !@HAVE_LCHOWN@ +_GL_FUNCDECL_SYS (lchown, int, (char const *file, uid_t owner, gid_t group) + _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (lchown, int, (char const *file, uid_t owner, gid_t group)); +# endif +_GL_CXXALIASWARN (lchown); +#elif defined GNULIB_POSIXCHECK +# undef lchown +# if HAVE_RAW_DECL_LCHOWN +_GL_WARN_ON_USE (lchown, "lchown is unportable to pre-POSIX.1-2001 systems - " + "use gnulib module lchown for portability"); +# endif +#endif + + +#if @GNULIB_LINK@ +/* Create a new hard link for an existing file. + Return 0 if successful, otherwise -1 and errno set. + See POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/link.html>. */ +# if @REPLACE_LINK@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define link rpl_link +# endif +_GL_FUNCDECL_RPL (link, int, (const char *path1, const char *path2) + _GL_ARG_NONNULL ((1, 2))); +_GL_CXXALIAS_RPL (link, int, (const char *path1, const char *path2)); +# else +# if !@HAVE_LINK@ +_GL_FUNCDECL_SYS (link, int, (const char *path1, const char *path2) + _GL_ARG_NONNULL ((1, 2))); +# endif +_GL_CXXALIAS_SYS (link, int, (const char *path1, const char *path2)); +# endif +_GL_CXXALIASWARN (link); +#elif defined GNULIB_POSIXCHECK +# undef link +# if HAVE_RAW_DECL_LINK +_GL_WARN_ON_USE (link, "link is unportable - " + "use gnulib module link for portability"); +# endif +#endif + + +#if @GNULIB_LINKAT@ +/* Create a new hard link for an existing file, relative to two + directories. FLAG controls whether symlinks are followed. + Return 0 if successful, otherwise -1 and errno set. */ +# if @REPLACE_LINKAT@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef linkat +# define linkat rpl_linkat +# endif +_GL_FUNCDECL_RPL (linkat, int, + (int fd1, const char *path1, int fd2, const char *path2, + int flag) + _GL_ARG_NONNULL ((2, 4))); +_GL_CXXALIAS_RPL (linkat, int, + (int fd1, const char *path1, int fd2, const char *path2, + int flag)); +# else +# if !@HAVE_LINKAT@ +_GL_FUNCDECL_SYS (linkat, int, + (int fd1, const char *path1, int fd2, const char *path2, + int flag) + _GL_ARG_NONNULL ((2, 4))); +# endif +_GL_CXXALIAS_SYS (linkat, int, + (int fd1, const char *path1, int fd2, const char *path2, + int flag)); +# endif +_GL_CXXALIASWARN (linkat); +#elif defined GNULIB_POSIXCHECK +# undef linkat +# if HAVE_RAW_DECL_LINKAT +_GL_WARN_ON_USE (linkat, "linkat is unportable - " + "use gnulib module linkat for portability"); +# endif +#endif + + +#if @GNULIB_LSEEK@ +/* Set the offset of FD relative to SEEK_SET, SEEK_CUR, or SEEK_END. + Return the new offset if successful, otherwise -1 and errno set. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/lseek.html>. */ +# if @REPLACE_LSEEK@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define lseek rpl_lseek +# endif +_GL_FUNCDECL_RPL (lseek, off_t, (int fd, off_t offset, int whence)); +_GL_CXXALIAS_RPL (lseek, off_t, (int fd, off_t offset, int whence)); +# else +_GL_CXXALIAS_SYS (lseek, off_t, (int fd, off_t offset, int whence)); +# endif +_GL_CXXALIASWARN (lseek); +#elif defined GNULIB_POSIXCHECK +# undef lseek +# if HAVE_RAW_DECL_LSEEK +_GL_WARN_ON_USE (lseek, "lseek does not fail with ESPIPE on pipes on some " + "systems - use gnulib module lseek for portability"); +# endif +#endif + + +#if @GNULIB_PIPE@ +/* Create a pipe, defaulting to O_BINARY mode. + Store the read-end as fd[0] and the write-end as fd[1]. + Return 0 upon success, or -1 with errno set upon failure. */ +# if !@HAVE_PIPE@ +_GL_FUNCDECL_SYS (pipe, int, (int fd[2]) _GL_ARG_NONNULL ((1))); +# endif +_GL_CXXALIAS_SYS (pipe, int, (int fd[2])); +_GL_CXXALIASWARN (pipe); +#elif defined GNULIB_POSIXCHECK +# undef pipe +# if HAVE_RAW_DECL_PIPE +_GL_WARN_ON_USE (pipe, "pipe is unportable - " + "use gnulib module pipe-posix for portability"); +# endif +#endif + + +#if @GNULIB_PIPE2@ +/* Create a pipe, applying the given flags when opening the read-end of the + pipe and the write-end of the pipe. + The flags are a bitmask, possibly including O_CLOEXEC (defined in <fcntl.h>) + and O_TEXT, O_BINARY (defined in "binary-io.h"). + Store the read-end as fd[0] and the write-end as fd[1]. + Return 0 upon success, or -1 with errno set upon failure. + See also the Linux man page at + <http://www.kernel.org/doc/man-pages/online/pages/man2/pipe2.2.html>. */ +# if @HAVE_PIPE2@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define pipe2 rpl_pipe2 +# endif +_GL_FUNCDECL_RPL (pipe2, int, (int fd[2], int flags) _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (pipe2, int, (int fd[2], int flags)); +# else +_GL_FUNCDECL_SYS (pipe2, int, (int fd[2], int flags) _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_SYS (pipe2, int, (int fd[2], int flags)); +# endif +_GL_CXXALIASWARN (pipe2); +#elif defined GNULIB_POSIXCHECK +# undef pipe2 +# if HAVE_RAW_DECL_PIPE2 +_GL_WARN_ON_USE (pipe2, "pipe2 is unportable - " + "use gnulib module pipe2 for portability"); +# endif +#endif + + +#if @GNULIB_PREAD@ +/* Read at most BUFSIZE bytes from FD into BUF, starting at OFFSET. + Return the number of bytes placed into BUF if successful, otherwise + set errno and return -1. 0 indicates EOF. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/pread.html>. */ +# if @REPLACE_PREAD@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef pread +# define pread rpl_pread +# endif +_GL_FUNCDECL_RPL (pread, ssize_t, + (int fd, void *buf, size_t bufsize, off_t offset) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (pread, ssize_t, + (int fd, void *buf, size_t bufsize, off_t offset)); +# else +# if !@HAVE_PREAD@ +_GL_FUNCDECL_SYS (pread, ssize_t, + (int fd, void *buf, size_t bufsize, off_t offset) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (pread, ssize_t, + (int fd, void *buf, size_t bufsize, off_t offset)); +# endif +_GL_CXXALIASWARN (pread); +#elif defined GNULIB_POSIXCHECK +# undef pread +# if HAVE_RAW_DECL_PREAD +_GL_WARN_ON_USE (pread, "pread is unportable - " + "use gnulib module pread for portability"); +# endif +#endif + + +#if @GNULIB_PWRITE@ +/* Write at most BUFSIZE bytes from BUF into FD, starting at OFFSET. + Return the number of bytes written if successful, otherwise + set errno and return -1. 0 indicates nothing written. See the + POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/pwrite.html>. */ +# if @REPLACE_PWRITE@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef pwrite +# define pwrite rpl_pwrite +# endif +_GL_FUNCDECL_RPL (pwrite, ssize_t, + (int fd, const void *buf, size_t bufsize, off_t offset) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (pwrite, ssize_t, + (int fd, const void *buf, size_t bufsize, off_t offset)); +# else +# if !@HAVE_PWRITE@ +_GL_FUNCDECL_SYS (pwrite, ssize_t, + (int fd, const void *buf, size_t bufsize, off_t offset) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (pwrite, ssize_t, + (int fd, const void *buf, size_t bufsize, off_t offset)); +# endif +_GL_CXXALIASWARN (pwrite); +#elif defined GNULIB_POSIXCHECK +# undef pwrite +# if HAVE_RAW_DECL_PWRITE +_GL_WARN_ON_USE (pwrite, "pwrite is unportable - " + "use gnulib module pwrite for portability"); +# endif +#endif + + +#if @GNULIB_READ@ +/* Read up to COUNT bytes from file descriptor FD into the buffer starting + at BUF. See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/read.html>. */ +# if @REPLACE_READ@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef read +# define read rpl_read +# endif +_GL_FUNCDECL_RPL (read, ssize_t, (int fd, void *buf, size_t count) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (read, ssize_t, (int fd, void *buf, size_t count)); +# else +/* Need to cast, because on mingw, the third parameter is + unsigned int count + and the return type is 'int'. */ +_GL_CXXALIAS_SYS_CAST (read, ssize_t, (int fd, void *buf, size_t count)); +# endif +_GL_CXXALIASWARN (read); +#endif + + +#if @GNULIB_READLINK@ +/* Read the contents of the symbolic link FILE and place the first BUFSIZE + bytes of it into BUF. Return the number of bytes placed into BUF if + successful, otherwise -1 and errno set. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/readlink.html>. */ +# if @REPLACE_READLINK@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define readlink rpl_readlink +# endif +_GL_FUNCDECL_RPL (readlink, ssize_t, + (const char *file, char *buf, size_t bufsize) + _GL_ARG_NONNULL ((1, 2))); +_GL_CXXALIAS_RPL (readlink, ssize_t, + (const char *file, char *buf, size_t bufsize)); +# else +# if !@HAVE_READLINK@ +_GL_FUNCDECL_SYS (readlink, ssize_t, + (const char *file, char *buf, size_t bufsize) + _GL_ARG_NONNULL ((1, 2))); +# endif +_GL_CXXALIAS_SYS (readlink, ssize_t, + (const char *file, char *buf, size_t bufsize)); +# endif +_GL_CXXALIASWARN (readlink); +#elif defined GNULIB_POSIXCHECK +# undef readlink +# if HAVE_RAW_DECL_READLINK +_GL_WARN_ON_USE (readlink, "readlink is unportable - " + "use gnulib module readlink for portability"); +# endif +#endif + + +#if @GNULIB_READLINKAT@ +# if !@HAVE_READLINKAT@ +_GL_FUNCDECL_SYS (readlinkat, ssize_t, + (int fd, char const *file, char *buf, size_t len) + _GL_ARG_NONNULL ((2, 3))); +# endif +_GL_CXXALIAS_SYS (readlinkat, ssize_t, + (int fd, char const *file, char *buf, size_t len)); +_GL_CXXALIASWARN (readlinkat); +#elif defined GNULIB_POSIXCHECK +# undef readlinkat +# if HAVE_RAW_DECL_READLINKAT +_GL_WARN_ON_USE (readlinkat, "readlinkat is not portable - " + "use gnulib module readlinkat for portability"); +# endif +#endif + + +#if @GNULIB_RMDIR@ +/* Remove the directory DIR. */ +# if @REPLACE_RMDIR@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define rmdir rpl_rmdir +# endif +_GL_FUNCDECL_RPL (rmdir, int, (char const *name) _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (rmdir, int, (char const *name)); +# else +_GL_CXXALIAS_SYS (rmdir, int, (char const *name)); +# endif +_GL_CXXALIASWARN (rmdir); +#elif defined GNULIB_POSIXCHECK +# undef rmdir +# if HAVE_RAW_DECL_RMDIR +_GL_WARN_ON_USE (rmdir, "rmdir is unportable - " + "use gnulib module rmdir for portability"); +# endif +#endif + + +#if @GNULIB_SETHOSTNAME@ +/* Set the host name of the machine. + The host name may or may not be fully qualified. + + Put LEN bytes of NAME into the host name. + Return 0 if successful, otherwise, set errno and return -1. + + Platforms with no ability to set the hostname return -1 and set + errno = ENOSYS. */ +# if !@HAVE_SETHOSTNAME@ || !@HAVE_DECL_SETHOSTNAME@ +_GL_FUNCDECL_SYS (sethostname, int, (const char *name, size_t len) + _GL_ARG_NONNULL ((1))); +# endif +/* Need to cast, because on Solaris 11 2011-10, MacOS X 10.5, IRIX 6.5 + and FreeBSD 6.4 the second parameter is int. On Solaris 11 + 2011-10, the first parameter is not const. */ +_GL_CXXALIAS_SYS_CAST (sethostname, int, (const char *name, size_t len)); +_GL_CXXALIASWARN (sethostname); +#elif defined GNULIB_POSIXCHECK +# undef sethostname +# if HAVE_RAW_DECL_SETHOSTNAME +_GL_WARN_ON_USE (sethostname, "sethostname is unportable - " + "use gnulib module sethostname for portability"); +# endif +#endif + + +#if @GNULIB_SLEEP@ +/* Pause the execution of the current thread for N seconds. + Returns the number of seconds left to sleep. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/sleep.html>. */ +# if @REPLACE_SLEEP@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef sleep +# define sleep rpl_sleep +# endif +_GL_FUNCDECL_RPL (sleep, unsigned int, (unsigned int n)); +_GL_CXXALIAS_RPL (sleep, unsigned int, (unsigned int n)); +# else +# if !@HAVE_SLEEP@ +_GL_FUNCDECL_SYS (sleep, unsigned int, (unsigned int n)); +# endif +_GL_CXXALIAS_SYS (sleep, unsigned int, (unsigned int n)); +# endif +_GL_CXXALIASWARN (sleep); +#elif defined GNULIB_POSIXCHECK +# undef sleep +# if HAVE_RAW_DECL_SLEEP +_GL_WARN_ON_USE (sleep, "sleep is unportable - " + "use gnulib module sleep for portability"); +# endif +#endif + + +#if @GNULIB_SYMLINK@ +# if @REPLACE_SYMLINK@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef symlink +# define symlink rpl_symlink +# endif +_GL_FUNCDECL_RPL (symlink, int, (char const *contents, char const *file) + _GL_ARG_NONNULL ((1, 2))); +_GL_CXXALIAS_RPL (symlink, int, (char const *contents, char const *file)); +# else +# if !@HAVE_SYMLINK@ +_GL_FUNCDECL_SYS (symlink, int, (char const *contents, char const *file) + _GL_ARG_NONNULL ((1, 2))); +# endif +_GL_CXXALIAS_SYS (symlink, int, (char const *contents, char const *file)); +# endif +_GL_CXXALIASWARN (symlink); +#elif defined GNULIB_POSIXCHECK +# undef symlink +# if HAVE_RAW_DECL_SYMLINK +_GL_WARN_ON_USE (symlink, "symlink is not portable - " + "use gnulib module symlink for portability"); +# endif +#endif + + +#if @GNULIB_SYMLINKAT@ +# if !@HAVE_SYMLINKAT@ +_GL_FUNCDECL_SYS (symlinkat, int, + (char const *contents, int fd, char const *file) + _GL_ARG_NONNULL ((1, 3))); +# endif +_GL_CXXALIAS_SYS (symlinkat, int, + (char const *contents, int fd, char const *file)); +_GL_CXXALIASWARN (symlinkat); +#elif defined GNULIB_POSIXCHECK +# undef symlinkat +# if HAVE_RAW_DECL_SYMLINKAT +_GL_WARN_ON_USE (symlinkat, "symlinkat is not portable - " + "use gnulib module symlinkat for portability"); +# endif +#endif + + +#if @GNULIB_TTYNAME_R@ +/* Store at most BUFLEN characters of the pathname of the terminal FD is + open on in BUF. Return 0 on success, otherwise an error number. */ +# if @REPLACE_TTYNAME_R@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef ttyname_r +# define ttyname_r rpl_ttyname_r +# endif +_GL_FUNCDECL_RPL (ttyname_r, int, + (int fd, char *buf, size_t buflen) _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (ttyname_r, int, + (int fd, char *buf, size_t buflen)); +# else +# if !@HAVE_DECL_TTYNAME_R@ +_GL_FUNCDECL_SYS (ttyname_r, int, + (int fd, char *buf, size_t buflen) _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (ttyname_r, int, + (int fd, char *buf, size_t buflen)); +# endif +_GL_CXXALIASWARN (ttyname_r); +#elif defined GNULIB_POSIXCHECK +# undef ttyname_r +# if HAVE_RAW_DECL_TTYNAME_R +_GL_WARN_ON_USE (ttyname_r, "ttyname_r is not portable - " + "use gnulib module ttyname_r for portability"); +# endif +#endif + + +#if @GNULIB_UNLINK@ +# if @REPLACE_UNLINK@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef unlink +# define unlink rpl_unlink +# endif +_GL_FUNCDECL_RPL (unlink, int, (char const *file) _GL_ARG_NONNULL ((1))); +_GL_CXXALIAS_RPL (unlink, int, (char const *file)); +# else +_GL_CXXALIAS_SYS (unlink, int, (char const *file)); +# endif +_GL_CXXALIASWARN (unlink); +#elif defined GNULIB_POSIXCHECK +# undef unlink +# if HAVE_RAW_DECL_UNLINK +_GL_WARN_ON_USE (unlink, "unlink is not portable - " + "use gnulib module unlink for portability"); +# endif +#endif + + +#if @GNULIB_UNLINKAT@ +# if @REPLACE_UNLINKAT@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef unlinkat +# define unlinkat rpl_unlinkat +# endif +_GL_FUNCDECL_RPL (unlinkat, int, (int fd, char const *file, int flag) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (unlinkat, int, (int fd, char const *file, int flag)); +# else +# if !@HAVE_UNLINKAT@ +_GL_FUNCDECL_SYS (unlinkat, int, (int fd, char const *file, int flag) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (unlinkat, int, (int fd, char const *file, int flag)); +# endif +_GL_CXXALIASWARN (unlinkat); +#elif defined GNULIB_POSIXCHECK +# undef unlinkat +# if HAVE_RAW_DECL_UNLINKAT +_GL_WARN_ON_USE (unlinkat, "unlinkat is not portable - " + "use gnulib module openat for portability"); +# endif +#endif + + +#if @GNULIB_USLEEP@ +/* Pause the execution of the current thread for N microseconds. + Returns 0 on completion, or -1 on range error. + See the POSIX:2001 specification + <http://www.opengroup.org/susv3xsh/usleep.html>. */ +# if @REPLACE_USLEEP@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef usleep +# define usleep rpl_usleep +# endif +_GL_FUNCDECL_RPL (usleep, int, (useconds_t n)); +_GL_CXXALIAS_RPL (usleep, int, (useconds_t n)); +# else +# if !@HAVE_USLEEP@ +_GL_FUNCDECL_SYS (usleep, int, (useconds_t n)); +# endif +_GL_CXXALIAS_SYS (usleep, int, (useconds_t n)); +# endif +_GL_CXXALIASWARN (usleep); +#elif defined GNULIB_POSIXCHECK +# undef usleep +# if HAVE_RAW_DECL_USLEEP +_GL_WARN_ON_USE (usleep, "usleep is unportable - " + "use gnulib module usleep for portability"); +# endif +#endif + + +#if @GNULIB_WRITE@ +/* Write up to COUNT bytes starting at BUF to file descriptor FD. + See the POSIX:2008 specification + <http://pubs.opengroup.org/onlinepubs/9699919799/functions/write.html>. */ +# if @REPLACE_WRITE@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef write +# define write rpl_write +# endif +_GL_FUNCDECL_RPL (write, ssize_t, (int fd, const void *buf, size_t count) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (write, ssize_t, (int fd, const void *buf, size_t count)); +# else +/* Need to cast, because on mingw, the third parameter is + unsigned int count + and the return type is 'int'. */ +_GL_CXXALIAS_SYS_CAST (write, ssize_t, (int fd, const void *buf, size_t count)); +# endif +_GL_CXXALIASWARN (write); +#endif + + +#endif /* _@GUARD_PREFIX@_UNISTD_H */ +#endif /* _@GUARD_PREFIX@_UNISTD_H */ diff --git a/lib/verify.h b/lib/verify.h new file mode 100644 index 0000000..cef14ad --- /dev/null +++ b/lib/verify.h @@ -0,0 +1,241 @@ +/* Compile-time assert-like macros. + + Copyright (C) 2005-2006, 2009-2012 Free Software Foundation, Inc. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +/* Written by Paul Eggert, Bruno Haible, and Jim Meyering. */ + +#ifndef _GL_VERIFY_H +# define _GL_VERIFY_H + + +/* Define _GL_HAVE__STATIC_ASSERT to 1 if _Static_assert works as per C11. + This is supported by GCC 4.6.0 and later, in C mode, and its use + here generates easier-to-read diagnostics when verify (R) fails. + + Define _GL_HAVE_STATIC_ASSERT to 1 if static_assert works as per C++11. + This will likely be supported by future GCC versions, in C++ mode. + + Use this only with GCC. If we were willing to slow 'configure' + down we could also use it with other compilers, but since this + affects only the quality of diagnostics, why bother? */ +# if (4 < __GNUC__ || (__GNUC__ == 4 && 6 <= __GNUC_MINOR__)) && !defined __cplusplus +# define _GL_HAVE__STATIC_ASSERT 1 +# endif +/* The condition (99 < __GNUC__) is temporary, until we know about the + first G++ release that supports static_assert. */ +# if (99 < __GNUC__) && defined __cplusplus +# define _GL_HAVE_STATIC_ASSERT 1 +# endif + +/* Each of these macros verifies that its argument R is nonzero. To + be portable, R should be an integer constant expression. Unlike + assert (R), there is no run-time overhead. + + If _Static_assert works, verify (R) uses it directly. Similarly, + _GL_VERIFY_TRUE works by packaging a _Static_assert inside a struct + that is an operand of sizeof. + + The code below uses several ideas for C++ compilers, and for C + compilers that do not support _Static_assert: + + * The first step is ((R) ? 1 : -1). Given an expression R, of + integral or boolean or floating-point type, this yields an + expression of integral type, whose value is later verified to be + constant and nonnegative. + + * Next this expression W is wrapped in a type + struct _gl_verify_type { + unsigned int _gl_verify_error_if_negative: W; + }. + If W is negative, this yields a compile-time error. No compiler can + deal with a bit-field of negative size. + + One might think that an array size check would have the same + effect, that is, that the type struct { unsigned int dummy[W]; } + would work as well. However, inside a function, some compilers + (such as C++ compilers and GNU C) allow local parameters and + variables inside array size expressions. With these compilers, + an array size check would not properly diagnose this misuse of + the verify macro: + + void function (int n) { verify (n < 0); } + + * For the verify macro, the struct _gl_verify_type will need to + somehow be embedded into a declaration. To be portable, this + declaration must declare an object, a constant, a function, or a + typedef name. If the declared entity uses the type directly, + such as in + + struct dummy {...}; + typedef struct {...} dummy; + extern struct {...} *dummy; + extern void dummy (struct {...} *); + extern struct {...} *dummy (void); + + two uses of the verify macro would yield colliding declarations + if the entity names are not disambiguated. A workaround is to + attach the current line number to the entity name: + + #define _GL_CONCAT0(x, y) x##y + #define _GL_CONCAT(x, y) _GL_CONCAT0 (x, y) + extern struct {...} * _GL_CONCAT (dummy, __LINE__); + + But this has the problem that two invocations of verify from + within the same macro would collide, since the __LINE__ value + would be the same for both invocations. (The GCC __COUNTER__ + macro solves this problem, but is not portable.) + + A solution is to use the sizeof operator. It yields a number, + getting rid of the identity of the type. Declarations like + + extern int dummy [sizeof (struct {...})]; + extern void dummy (int [sizeof (struct {...})]); + extern int (*dummy (void)) [sizeof (struct {...})]; + + can be repeated. + + * Should the implementation use a named struct or an unnamed struct? + Which of the following alternatives can be used? + + extern int dummy [sizeof (struct {...})]; + extern int dummy [sizeof (struct _gl_verify_type {...})]; + extern void dummy (int [sizeof (struct {...})]); + extern void dummy (int [sizeof (struct _gl_verify_type {...})]); + extern int (*dummy (void)) [sizeof (struct {...})]; + extern int (*dummy (void)) [sizeof (struct _gl_verify_type {...})]; + + In the second and sixth case, the struct type is exported to the + outer scope; two such declarations therefore collide. GCC warns + about the first, third, and fourth cases. So the only remaining + possibility is the fifth case: + + extern int (*dummy (void)) [sizeof (struct {...})]; + + * GCC warns about duplicate declarations of the dummy function if + -Wredundant_decls is used. GCC 4.3 and later have a builtin + __COUNTER__ macro that can let us generate unique identifiers for + each dummy function, to suppress this warning. + + * This implementation exploits the fact that older versions of GCC, + which do not support _Static_assert, also do not warn about the + last declaration mentioned above. + + * In C++, any struct definition inside sizeof is invalid. + Use a template type to work around the problem. */ + +/* Concatenate two preprocessor tokens. */ +# define _GL_CONCAT(x, y) _GL_CONCAT0 (x, y) +# define _GL_CONCAT0(x, y) x##y + +/* _GL_COUNTER is an integer, preferably one that changes each time we + use it. Use __COUNTER__ if it works, falling back on __LINE__ + otherwise. __LINE__ isn't perfect, but it's better than a + constant. */ +# if defined __COUNTER__ && __COUNTER__ != __COUNTER__ +# define _GL_COUNTER __COUNTER__ +# else +# define _GL_COUNTER __LINE__ +# endif + +/* Generate a symbol with the given prefix, making it unique if + possible. */ +# define _GL_GENSYM(prefix) _GL_CONCAT (prefix, _GL_COUNTER) + +/* Verify requirement R at compile-time, as an integer constant expression + that returns 1. If R is false, fail at compile-time, preferably + with a diagnostic that includes the string-literal DIAGNOSTIC. */ + +# define _GL_VERIFY_TRUE(R, DIAGNOSTIC) \ + (!!sizeof (_GL_VERIFY_TYPE (R, DIAGNOSTIC))) + +# ifdef __cplusplus +# if !GNULIB_defined_struct__gl_verify_type +template <int w> + struct _gl_verify_type { + unsigned int _gl_verify_error_if_negative: w; + }; +# define GNULIB_defined_struct__gl_verify_type 1 +# endif +# define _GL_VERIFY_TYPE(R, DIAGNOSTIC) \ + _gl_verify_type<(R) ? 1 : -1> +# elif defined _GL_HAVE__STATIC_ASSERT +# define _GL_VERIFY_TYPE(R, DIAGNOSTIC) \ + struct { \ + _Static_assert (R, DIAGNOSTIC); \ + int _gl_dummy; \ + } +# else +# define _GL_VERIFY_TYPE(R, DIAGNOSTIC) \ + struct { unsigned int _gl_verify_error_if_negative: (R) ? 1 : -1; } +# endif + +/* Verify requirement R at compile-time, as a declaration without a + trailing ';'. If R is false, fail at compile-time, preferably + with a diagnostic that includes the string-literal DIAGNOSTIC. + + Unfortunately, unlike C11, this implementation must appear as an + ordinary declaration, and cannot appear inside struct { ... }. */ + +# ifdef _GL_HAVE__STATIC_ASSERT +# define _GL_VERIFY _Static_assert +# else +# define _GL_VERIFY(R, DIAGNOSTIC) \ + extern int (*_GL_GENSYM (_gl_verify_function) (void)) \ + [_GL_VERIFY_TRUE (R, DIAGNOSTIC)] +# endif + +/* _GL_STATIC_ASSERT_H is defined if this code is copied into assert.h. */ +# ifdef _GL_STATIC_ASSERT_H +# if !defined _GL_HAVE__STATIC_ASSERT && !defined _Static_assert +# define _Static_assert(R, DIAGNOSTIC) _GL_VERIFY (R, DIAGNOSTIC) +# endif +# if !defined _GL_HAVE_STATIC_ASSERT && !defined static_assert +# define static_assert _Static_assert /* C11 requires this #define. */ +# endif +# endif + +/* @assert.h omit start@ */ + +/* Each of these macros verifies that its argument R is nonzero. To + be portable, R should be an integer constant expression. Unlike + assert (R), there is no run-time overhead. + + There are two macros, since no single macro can be used in all + contexts in C. verify_true (R) is for scalar contexts, including + integer constant expression contexts. verify (R) is for declaration + contexts, e.g., the top level. */ + +/* Verify requirement R at compile-time, as an integer constant expression. + Return 1. This is equivalent to verify_expr (R, 1). + + verify_true is obsolescent; please use verify_expr instead. */ + +# define verify_true(R) _GL_VERIFY_TRUE (R, "verify_true (" #R ")") + +/* Verify requirement R at compile-time. Return the value of the + expression E. */ + +# define verify_expr(R, E) \ + (_GL_VERIFY_TRUE (R, "verify_expr (" #R ", " #E ")") ? (E) : (E)) + +/* Verify requirement R at compile-time, as a declaration without a + trailing ';'. */ + +# define verify(R) _GL_VERIFY (R, "verify (" #R ")") + +/* @assert.h omit end@ */ + +#endif diff --git a/lib/wchar.in.h b/lib/wchar.in.h new file mode 100644 index 0000000..4e92a77 --- /dev/null +++ b/lib/wchar.in.h @@ -0,0 +1,1005 @@ +/* A substitute for ISO C99 <wchar.h>, for platforms that have issues. + + Copyright (C) 2007-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +/* Written by Eric Blake. */ + +/* + * ISO C 99 <wchar.h> for platforms that have issues. + * <http://www.opengroup.org/susv3xbd/wchar.h.html> + * + * For now, this just ensures proper prerequisite inclusion order and + * the declaration of wcwidth(). + */ + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +#if defined __need_mbstate_t || defined __need_wint_t || (defined __hpux && ((defined _INTTYPES_INCLUDED && !defined strtoimax) || defined _GL_JUST_INCLUDE_SYSTEM_WCHAR_H)) || defined _GL_ALREADY_INCLUDING_WCHAR_H +/* Special invocation convention: + - Inside glibc and uClibc header files. + - On HP-UX 11.00 we have a sequence of nested includes + <wchar.h> -> <stdlib.h> -> <stdint.h>, and the latter includes <wchar.h>, + once indirectly <stdint.h> -> <sys/types.h> -> <inttypes.h> -> <wchar.h> + and once directly. In both situations 'wint_t' is not yet defined, + therefore we cannot provide the function overrides; instead include only + the system's <wchar.h>. + - On IRIX 6.5, similarly, we have an include <wchar.h> -> <wctype.h>, and + the latter includes <wchar.h>. But here, we have no way to detect whether + <wctype.h> is completely included or is still being included. */ + +#@INCLUDE_NEXT@ @NEXT_WCHAR_H@ + +#else +/* Normal invocation convention. */ + +#ifndef _@GUARD_PREFIX@_WCHAR_H + +#define _GL_ALREADY_INCLUDING_WCHAR_H + +#if @HAVE_FEATURES_H@ +# include <features.h> /* for __GLIBC__ */ +#endif + +/* 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>. + In some builds of uClibc, <wchar.h> is nonexistent and wchar_t is defined + by <stddef.h>. + But avoid namespace pollution on glibc systems. */ +#if !(defined __GLIBC__ && !defined __UCLIBC__) +# include <stddef.h> +#endif +#ifndef __GLIBC__ +# include <stdio.h> +# include <time.h> +#endif + +/* Include the original <wchar.h> if it exists. + Some builds of uClibc lack it. */ +/* The include_next requires a split double-inclusion guard. */ +#if @HAVE_WCHAR_H@ +# @INCLUDE_NEXT@ @NEXT_WCHAR_H@ +#endif + +#undef _GL_ALREADY_INCLUDING_WCHAR_H + +#ifndef _@GUARD_PREFIX@_WCHAR_H +#define _@GUARD_PREFIX@_WCHAR_H + +/* The definitions of _GL_FUNCDECL_RPL etc. are copied here. */ + +/* The definition of _GL_ARG_NONNULL is copied here. */ + +/* The definition of _GL_WARN_ON_USE is copied here. */ + + +/* Define wint_t and WEOF. (Also done in wctype.in.h.) */ +#if !@HAVE_WINT_T@ && !defined wint_t +# define wint_t int +# ifndef WEOF +# define WEOF -1 +# endif +#else +/* MSVC defines wint_t as 'unsigned short' in <crtdefs.h>. + This is too small: ISO C 99 section 7.24.1.(2) says that wint_t must be + "unchanged by default argument promotions". Override it. */ +# if defined _MSC_VER +# if !GNULIB_defined_wint_t +# include <crtdefs.h> +typedef unsigned int rpl_wint_t; +# undef wint_t +# define wint_t rpl_wint_t +# define GNULIB_defined_wint_t 1 +# endif +# endif +# ifndef WEOF +# define WEOF ((wint_t) -1) +# endif +#endif + + +/* Override mbstate_t if it is too small. + On IRIX 6.5, sizeof (mbstate_t) == 1, which is not sufficient for + implementing mbrtowc for encodings like UTF-8. */ +#if !(@HAVE_MBSINIT@ && @HAVE_MBRTOWC@) || @REPLACE_MBSTATE_T@ +# if !GNULIB_defined_mbstate_t +typedef int rpl_mbstate_t; +# undef mbstate_t +# define mbstate_t rpl_mbstate_t +# define GNULIB_defined_mbstate_t 1 +# endif +#endif + + +/* Convert a single-byte character to a wide character. */ +#if @GNULIB_BTOWC@ +# if @REPLACE_BTOWC@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef btowc +# define btowc rpl_btowc +# endif +_GL_FUNCDECL_RPL (btowc, wint_t, (int c)); +_GL_CXXALIAS_RPL (btowc, wint_t, (int c)); +# else +# if !@HAVE_BTOWC@ +_GL_FUNCDECL_SYS (btowc, wint_t, (int c)); +# endif +_GL_CXXALIAS_SYS (btowc, wint_t, (int c)); +# endif +_GL_CXXALIASWARN (btowc); +#elif defined GNULIB_POSIXCHECK +# undef btowc +# if HAVE_RAW_DECL_BTOWC +_GL_WARN_ON_USE (btowc, "btowc is unportable - " + "use gnulib module btowc for portability"); +# endif +#endif + + +/* Convert a wide character to a single-byte character. */ +#if @GNULIB_WCTOB@ +# if @REPLACE_WCTOB@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef wctob +# define wctob rpl_wctob +# endif +_GL_FUNCDECL_RPL (wctob, int, (wint_t wc)); +_GL_CXXALIAS_RPL (wctob, int, (wint_t wc)); +# else +# if !defined wctob && !@HAVE_DECL_WCTOB@ +/* wctob is provided by gnulib, or wctob exists but is not declared. */ +_GL_FUNCDECL_SYS (wctob, int, (wint_t wc)); +# endif +_GL_CXXALIAS_SYS (wctob, int, (wint_t wc)); +# endif +_GL_CXXALIASWARN (wctob); +#elif defined GNULIB_POSIXCHECK +# undef wctob +# if HAVE_RAW_DECL_WCTOB +_GL_WARN_ON_USE (wctob, "wctob is unportable - " + "use gnulib module wctob for portability"); +# endif +#endif + + +/* Test whether *PS is in the initial state. */ +#if @GNULIB_MBSINIT@ +# if @REPLACE_MBSINIT@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef mbsinit +# define mbsinit rpl_mbsinit +# endif +_GL_FUNCDECL_RPL (mbsinit, int, (const mbstate_t *ps)); +_GL_CXXALIAS_RPL (mbsinit, int, (const mbstate_t *ps)); +# else +# if !@HAVE_MBSINIT@ +_GL_FUNCDECL_SYS (mbsinit, int, (const mbstate_t *ps)); +# endif +_GL_CXXALIAS_SYS (mbsinit, int, (const mbstate_t *ps)); +# endif +_GL_CXXALIASWARN (mbsinit); +#elif defined GNULIB_POSIXCHECK +# undef mbsinit +# if HAVE_RAW_DECL_MBSINIT +_GL_WARN_ON_USE (mbsinit, "mbsinit is unportable - " + "use gnulib module mbsinit for portability"); +# endif +#endif + + +/* Convert a multibyte character to a wide character. */ +#if @GNULIB_MBRTOWC@ +# if @REPLACE_MBRTOWC@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef mbrtowc +# define mbrtowc rpl_mbrtowc +# endif +_GL_FUNCDECL_RPL (mbrtowc, size_t, + (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)); +_GL_CXXALIAS_RPL (mbrtowc, size_t, + (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)); +# else +# if !@HAVE_MBRTOWC@ +_GL_FUNCDECL_SYS (mbrtowc, size_t, + (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)); +# endif +_GL_CXXALIAS_SYS (mbrtowc, size_t, + (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)); +# endif +_GL_CXXALIASWARN (mbrtowc); +#elif defined GNULIB_POSIXCHECK +# undef mbrtowc +# if HAVE_RAW_DECL_MBRTOWC +_GL_WARN_ON_USE (mbrtowc, "mbrtowc is unportable - " + "use gnulib module mbrtowc for portability"); +# endif +#endif + + +/* Recognize a multibyte character. */ +#if @GNULIB_MBRLEN@ +# if @REPLACE_MBRLEN@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef mbrlen +# define mbrlen rpl_mbrlen +# endif +_GL_FUNCDECL_RPL (mbrlen, size_t, (const char *s, size_t n, mbstate_t *ps)); +_GL_CXXALIAS_RPL (mbrlen, size_t, (const char *s, size_t n, mbstate_t *ps)); +# else +# if !@HAVE_MBRLEN@ +_GL_FUNCDECL_SYS (mbrlen, size_t, (const char *s, size_t n, mbstate_t *ps)); +# endif +_GL_CXXALIAS_SYS (mbrlen, size_t, (const char *s, size_t n, mbstate_t *ps)); +# endif +_GL_CXXALIASWARN (mbrlen); +#elif defined GNULIB_POSIXCHECK +# undef mbrlen +# if HAVE_RAW_DECL_MBRLEN +_GL_WARN_ON_USE (mbrlen, "mbrlen is unportable - " + "use gnulib module mbrlen for portability"); +# endif +#endif + + +/* Convert a string to a wide string. */ +#if @GNULIB_MBSRTOWCS@ +# if @REPLACE_MBSRTOWCS@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef mbsrtowcs +# define mbsrtowcs rpl_mbsrtowcs +# endif +_GL_FUNCDECL_RPL (mbsrtowcs, size_t, + (wchar_t *dest, const char **srcp, size_t len, mbstate_t *ps) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (mbsrtowcs, size_t, + (wchar_t *dest, const char **srcp, size_t len, + mbstate_t *ps)); +# else +# if !@HAVE_MBSRTOWCS@ +_GL_FUNCDECL_SYS (mbsrtowcs, size_t, + (wchar_t *dest, const char **srcp, size_t len, mbstate_t *ps) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (mbsrtowcs, size_t, + (wchar_t *dest, const char **srcp, size_t len, + mbstate_t *ps)); +# endif +_GL_CXXALIASWARN (mbsrtowcs); +#elif defined GNULIB_POSIXCHECK +# undef mbsrtowcs +# if HAVE_RAW_DECL_MBSRTOWCS +_GL_WARN_ON_USE (mbsrtowcs, "mbsrtowcs is unportable - " + "use gnulib module mbsrtowcs for portability"); +# endif +#endif + + +/* Convert a string to a wide string. */ +#if @GNULIB_MBSNRTOWCS@ +# if @REPLACE_MBSNRTOWCS@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef mbsnrtowcs +# define mbsnrtowcs rpl_mbsnrtowcs +# endif +_GL_FUNCDECL_RPL (mbsnrtowcs, size_t, + (wchar_t *dest, const char **srcp, size_t srclen, size_t len, + mbstate_t *ps) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (mbsnrtowcs, size_t, + (wchar_t *dest, const char **srcp, size_t srclen, size_t len, + mbstate_t *ps)); +# else +# if !@HAVE_MBSNRTOWCS@ +_GL_FUNCDECL_SYS (mbsnrtowcs, size_t, + (wchar_t *dest, const char **srcp, size_t srclen, size_t len, + mbstate_t *ps) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (mbsnrtowcs, size_t, + (wchar_t *dest, const char **srcp, size_t srclen, size_t len, + mbstate_t *ps)); +# endif +_GL_CXXALIASWARN (mbsnrtowcs); +#elif defined GNULIB_POSIXCHECK +# undef mbsnrtowcs +# if HAVE_RAW_DECL_MBSNRTOWCS +_GL_WARN_ON_USE (mbsnrtowcs, "mbsnrtowcs is unportable - " + "use gnulib module mbsnrtowcs for portability"); +# endif +#endif + + +/* Convert a wide character to a multibyte character. */ +#if @GNULIB_WCRTOMB@ +# if @REPLACE_WCRTOMB@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef wcrtomb +# define wcrtomb rpl_wcrtomb +# endif +_GL_FUNCDECL_RPL (wcrtomb, size_t, (char *s, wchar_t wc, mbstate_t *ps)); +_GL_CXXALIAS_RPL (wcrtomb, size_t, (char *s, wchar_t wc, mbstate_t *ps)); +# else +# if !@HAVE_WCRTOMB@ +_GL_FUNCDECL_SYS (wcrtomb, size_t, (char *s, wchar_t wc, mbstate_t *ps)); +# endif +_GL_CXXALIAS_SYS (wcrtomb, size_t, (char *s, wchar_t wc, mbstate_t *ps)); +# endif +_GL_CXXALIASWARN (wcrtomb); +#elif defined GNULIB_POSIXCHECK +# undef wcrtomb +# if HAVE_RAW_DECL_WCRTOMB +_GL_WARN_ON_USE (wcrtomb, "wcrtomb is unportable - " + "use gnulib module wcrtomb for portability"); +# endif +#endif + + +/* Convert a wide string to a string. */ +#if @GNULIB_WCSRTOMBS@ +# if @REPLACE_WCSRTOMBS@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef wcsrtombs +# define wcsrtombs rpl_wcsrtombs +# endif +_GL_FUNCDECL_RPL (wcsrtombs, size_t, + (char *dest, const wchar_t **srcp, size_t len, mbstate_t *ps) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (wcsrtombs, size_t, + (char *dest, const wchar_t **srcp, size_t len, + mbstate_t *ps)); +# else +# if !@HAVE_WCSRTOMBS@ +_GL_FUNCDECL_SYS (wcsrtombs, size_t, + (char *dest, const wchar_t **srcp, size_t len, mbstate_t *ps) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (wcsrtombs, size_t, + (char *dest, const wchar_t **srcp, size_t len, + mbstate_t *ps)); +# endif +_GL_CXXALIASWARN (wcsrtombs); +#elif defined GNULIB_POSIXCHECK +# undef wcsrtombs +# if HAVE_RAW_DECL_WCSRTOMBS +_GL_WARN_ON_USE (wcsrtombs, "wcsrtombs is unportable - " + "use gnulib module wcsrtombs for portability"); +# endif +#endif + + +/* Convert a wide string to a string. */ +#if @GNULIB_WCSNRTOMBS@ +# if @REPLACE_WCSNRTOMBS@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef wcsnrtombs +# define wcsnrtombs rpl_wcsnrtombs +# endif +_GL_FUNCDECL_RPL (wcsnrtombs, size_t, + (char *dest, const wchar_t **srcp, size_t srclen, size_t len, + mbstate_t *ps) + _GL_ARG_NONNULL ((2))); +_GL_CXXALIAS_RPL (wcsnrtombs, size_t, + (char *dest, const wchar_t **srcp, size_t srclen, size_t len, + mbstate_t *ps)); +# else +# if !@HAVE_WCSNRTOMBS@ +_GL_FUNCDECL_SYS (wcsnrtombs, size_t, + (char *dest, const wchar_t **srcp, size_t srclen, size_t len, + mbstate_t *ps) + _GL_ARG_NONNULL ((2))); +# endif +_GL_CXXALIAS_SYS (wcsnrtombs, size_t, + (char *dest, const wchar_t **srcp, size_t srclen, size_t len, + mbstate_t *ps)); +# endif +_GL_CXXALIASWARN (wcsnrtombs); +#elif defined GNULIB_POSIXCHECK +# undef wcsnrtombs +# if HAVE_RAW_DECL_WCSNRTOMBS +_GL_WARN_ON_USE (wcsnrtombs, "wcsnrtombs is unportable - " + "use gnulib module wcsnrtombs for portability"); +# endif +#endif + + +/* Return the number of screen columns needed for WC. */ +#if @GNULIB_WCWIDTH@ +# if @REPLACE_WCWIDTH@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef wcwidth +# define wcwidth rpl_wcwidth +# endif +_GL_FUNCDECL_RPL (wcwidth, int, (wchar_t)); +_GL_CXXALIAS_RPL (wcwidth, int, (wchar_t)); +# else +# if !@HAVE_DECL_WCWIDTH@ +/* wcwidth exists but is not declared. */ +_GL_FUNCDECL_SYS (wcwidth, int, (wchar_t)); +# endif +_GL_CXXALIAS_SYS (wcwidth, int, (wchar_t)); +# endif +_GL_CXXALIASWARN (wcwidth); +#elif defined GNULIB_POSIXCHECK +# undef wcwidth +# if HAVE_RAW_DECL_WCWIDTH +_GL_WARN_ON_USE (wcwidth, "wcwidth is unportable - " + "use gnulib module wcwidth for portability"); +# endif +#endif + + +/* Search N wide characters of S for C. */ +#if @GNULIB_WMEMCHR@ +# if !@HAVE_WMEMCHR@ +_GL_FUNCDECL_SYS (wmemchr, wchar_t *, (const wchar_t *s, wchar_t c, size_t n)); +# endif + /* On some systems, this function is defined as an overloaded function: + extern "C++" { + const wchar_t * std::wmemchr (const wchar_t *, wchar_t, size_t); + wchar_t * std::wmemchr (wchar_t *, wchar_t, size_t); + } */ +_GL_CXXALIAS_SYS_CAST2 (wmemchr, + wchar_t *, (const wchar_t *, wchar_t, size_t), + const wchar_t *, (const wchar_t *, wchar_t, size_t)); +# if ((__GLIBC__ == 2 && __GLIBC_MINOR__ >= 10) && !defined __UCLIBC__) \ + && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4)) +_GL_CXXALIASWARN1 (wmemchr, wchar_t *, (wchar_t *s, wchar_t c, size_t n)); +_GL_CXXALIASWARN1 (wmemchr, const wchar_t *, + (const wchar_t *s, wchar_t c, size_t n)); +# else +_GL_CXXALIASWARN (wmemchr); +# endif +#elif defined GNULIB_POSIXCHECK +# undef wmemchr +# if HAVE_RAW_DECL_WMEMCHR +_GL_WARN_ON_USE (wmemchr, "wmemchr is unportable - " + "use gnulib module wmemchr for portability"); +# endif +#endif + + +/* Compare N wide characters of S1 and S2. */ +#if @GNULIB_WMEMCMP@ +# if !@HAVE_WMEMCMP@ +_GL_FUNCDECL_SYS (wmemcmp, int, + (const wchar_t *s1, const wchar_t *s2, size_t n)); +# endif +_GL_CXXALIAS_SYS (wmemcmp, int, + (const wchar_t *s1, const wchar_t *s2, size_t n)); +_GL_CXXALIASWARN (wmemcmp); +#elif defined GNULIB_POSIXCHECK +# undef wmemcmp +# if HAVE_RAW_DECL_WMEMCMP +_GL_WARN_ON_USE (wmemcmp, "wmemcmp is unportable - " + "use gnulib module wmemcmp for portability"); +# endif +#endif + + +/* Copy N wide characters of SRC to DEST. */ +#if @GNULIB_WMEMCPY@ +# if !@HAVE_WMEMCPY@ +_GL_FUNCDECL_SYS (wmemcpy, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +# endif +_GL_CXXALIAS_SYS (wmemcpy, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +_GL_CXXALIASWARN (wmemcpy); +#elif defined GNULIB_POSIXCHECK +# undef wmemcpy +# if HAVE_RAW_DECL_WMEMCPY +_GL_WARN_ON_USE (wmemcpy, "wmemcpy is unportable - " + "use gnulib module wmemcpy for portability"); +# endif +#endif + + +/* Copy N wide characters of SRC to DEST, guaranteeing correct behavior for + overlapping memory areas. */ +#if @GNULIB_WMEMMOVE@ +# if !@HAVE_WMEMMOVE@ +_GL_FUNCDECL_SYS (wmemmove, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +# endif +_GL_CXXALIAS_SYS (wmemmove, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +_GL_CXXALIASWARN (wmemmove); +#elif defined GNULIB_POSIXCHECK +# undef wmemmove +# if HAVE_RAW_DECL_WMEMMOVE +_GL_WARN_ON_USE (wmemmove, "wmemmove is unportable - " + "use gnulib module wmemmove for portability"); +# endif +#endif + + +/* Set N wide characters of S to C. */ +#if @GNULIB_WMEMSET@ +# if !@HAVE_WMEMSET@ +_GL_FUNCDECL_SYS (wmemset, wchar_t *, (wchar_t *s, wchar_t c, size_t n)); +# endif +_GL_CXXALIAS_SYS (wmemset, wchar_t *, (wchar_t *s, wchar_t c, size_t n)); +_GL_CXXALIASWARN (wmemset); +#elif defined GNULIB_POSIXCHECK +# undef wmemset +# if HAVE_RAW_DECL_WMEMSET +_GL_WARN_ON_USE (wmemset, "wmemset is unportable - " + "use gnulib module wmemset for portability"); +# endif +#endif + + +/* Return the number of wide characters in S. */ +#if @GNULIB_WCSLEN@ +# if !@HAVE_WCSLEN@ +_GL_FUNCDECL_SYS (wcslen, size_t, (const wchar_t *s)); +# endif +_GL_CXXALIAS_SYS (wcslen, size_t, (const wchar_t *s)); +_GL_CXXALIASWARN (wcslen); +#elif defined GNULIB_POSIXCHECK +# undef wcslen +# if HAVE_RAW_DECL_WCSLEN +_GL_WARN_ON_USE (wcslen, "wcslen is unportable - " + "use gnulib module wcslen for portability"); +# endif +#endif + + +/* Return the number of wide characters in S, but at most MAXLEN. */ +#if @GNULIB_WCSNLEN@ +# if !@HAVE_WCSNLEN@ +_GL_FUNCDECL_SYS (wcsnlen, size_t, (const wchar_t *s, size_t maxlen)); +# endif +_GL_CXXALIAS_SYS (wcsnlen, size_t, (const wchar_t *s, size_t maxlen)); +_GL_CXXALIASWARN (wcsnlen); +#elif defined GNULIB_POSIXCHECK +# undef wcsnlen +# if HAVE_RAW_DECL_WCSNLEN +_GL_WARN_ON_USE (wcsnlen, "wcsnlen is unportable - " + "use gnulib module wcsnlen for portability"); +# endif +#endif + + +/* Copy SRC to DEST. */ +#if @GNULIB_WCSCPY@ +# if !@HAVE_WCSCPY@ +_GL_FUNCDECL_SYS (wcscpy, wchar_t *, (wchar_t *dest, const wchar_t *src)); +# endif +_GL_CXXALIAS_SYS (wcscpy, wchar_t *, (wchar_t *dest, const wchar_t *src)); +_GL_CXXALIASWARN (wcscpy); +#elif defined GNULIB_POSIXCHECK +# undef wcscpy +# if HAVE_RAW_DECL_WCSCPY +_GL_WARN_ON_USE (wcscpy, "wcscpy is unportable - " + "use gnulib module wcscpy for portability"); +# endif +#endif + + +/* Copy SRC to DEST, returning the address of the terminating L'\0' in DEST. */ +#if @GNULIB_WCPCPY@ +# if !@HAVE_WCPCPY@ +_GL_FUNCDECL_SYS (wcpcpy, wchar_t *, (wchar_t *dest, const wchar_t *src)); +# endif +_GL_CXXALIAS_SYS (wcpcpy, wchar_t *, (wchar_t *dest, const wchar_t *src)); +_GL_CXXALIASWARN (wcpcpy); +#elif defined GNULIB_POSIXCHECK +# undef wcpcpy +# if HAVE_RAW_DECL_WCPCPY +_GL_WARN_ON_USE (wcpcpy, "wcpcpy is unportable - " + "use gnulib module wcpcpy for portability"); +# endif +#endif + + +/* Copy no more than N wide characters of SRC to DEST. */ +#if @GNULIB_WCSNCPY@ +# if !@HAVE_WCSNCPY@ +_GL_FUNCDECL_SYS (wcsncpy, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +# endif +_GL_CXXALIAS_SYS (wcsncpy, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +_GL_CXXALIASWARN (wcsncpy); +#elif defined GNULIB_POSIXCHECK +# undef wcsncpy +# if HAVE_RAW_DECL_WCSNCPY +_GL_WARN_ON_USE (wcsncpy, "wcsncpy is unportable - " + "use gnulib module wcsncpy for portability"); +# endif +#endif + + +/* Copy no more than N characters of SRC to DEST, returning the address of + the last character written into DEST. */ +#if @GNULIB_WCPNCPY@ +# if !@HAVE_WCPNCPY@ +_GL_FUNCDECL_SYS (wcpncpy, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +# endif +_GL_CXXALIAS_SYS (wcpncpy, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +_GL_CXXALIASWARN (wcpncpy); +#elif defined GNULIB_POSIXCHECK +# undef wcpncpy +# if HAVE_RAW_DECL_WCPNCPY +_GL_WARN_ON_USE (wcpncpy, "wcpncpy is unportable - " + "use gnulib module wcpncpy for portability"); +# endif +#endif + + +/* Append SRC onto DEST. */ +#if @GNULIB_WCSCAT@ +# if !@HAVE_WCSCAT@ +_GL_FUNCDECL_SYS (wcscat, wchar_t *, (wchar_t *dest, const wchar_t *src)); +# endif +_GL_CXXALIAS_SYS (wcscat, wchar_t *, (wchar_t *dest, const wchar_t *src)); +_GL_CXXALIASWARN (wcscat); +#elif defined GNULIB_POSIXCHECK +# undef wcscat +# if HAVE_RAW_DECL_WCSCAT +_GL_WARN_ON_USE (wcscat, "wcscat is unportable - " + "use gnulib module wcscat for portability"); +# endif +#endif + + +/* Append no more than N wide characters of SRC onto DEST. */ +#if @GNULIB_WCSNCAT@ +# if !@HAVE_WCSNCAT@ +_GL_FUNCDECL_SYS (wcsncat, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +# endif +_GL_CXXALIAS_SYS (wcsncat, wchar_t *, + (wchar_t *dest, const wchar_t *src, size_t n)); +_GL_CXXALIASWARN (wcsncat); +#elif defined GNULIB_POSIXCHECK +# undef wcsncat +# if HAVE_RAW_DECL_WCSNCAT +_GL_WARN_ON_USE (wcsncat, "wcsncat is unportable - " + "use gnulib module wcsncat for portability"); +# endif +#endif + + +/* Compare S1 and S2. */ +#if @GNULIB_WCSCMP@ +# if !@HAVE_WCSCMP@ +_GL_FUNCDECL_SYS (wcscmp, int, (const wchar_t *s1, const wchar_t *s2)); +# endif +_GL_CXXALIAS_SYS (wcscmp, int, (const wchar_t *s1, const wchar_t *s2)); +_GL_CXXALIASWARN (wcscmp); +#elif defined GNULIB_POSIXCHECK +# undef wcscmp +# if HAVE_RAW_DECL_WCSCMP +_GL_WARN_ON_USE (wcscmp, "wcscmp is unportable - " + "use gnulib module wcscmp for portability"); +# endif +#endif + + +/* Compare no more than N wide characters of S1 and S2. */ +#if @GNULIB_WCSNCMP@ +# if !@HAVE_WCSNCMP@ +_GL_FUNCDECL_SYS (wcsncmp, int, + (const wchar_t *s1, const wchar_t *s2, size_t n)); +# endif +_GL_CXXALIAS_SYS (wcsncmp, int, + (const wchar_t *s1, const wchar_t *s2, size_t n)); +_GL_CXXALIASWARN (wcsncmp); +#elif defined GNULIB_POSIXCHECK +# undef wcsncmp +# if HAVE_RAW_DECL_WCSNCMP +_GL_WARN_ON_USE (wcsncmp, "wcsncmp is unportable - " + "use gnulib module wcsncmp for portability"); +# endif +#endif + + +/* Compare S1 and S2, ignoring case. */ +#if @GNULIB_WCSCASECMP@ +# if !@HAVE_WCSCASECMP@ +_GL_FUNCDECL_SYS (wcscasecmp, int, (const wchar_t *s1, const wchar_t *s2)); +# endif +_GL_CXXALIAS_SYS (wcscasecmp, int, (const wchar_t *s1, const wchar_t *s2)); +_GL_CXXALIASWARN (wcscasecmp); +#elif defined GNULIB_POSIXCHECK +# undef wcscasecmp +# if HAVE_RAW_DECL_WCSCASECMP +_GL_WARN_ON_USE (wcscasecmp, "wcscasecmp is unportable - " + "use gnulib module wcscasecmp for portability"); +# endif +#endif + + +/* Compare no more than N chars of S1 and S2, ignoring case. */ +#if @GNULIB_WCSNCASECMP@ +# if !@HAVE_WCSNCASECMP@ +_GL_FUNCDECL_SYS (wcsncasecmp, int, + (const wchar_t *s1, const wchar_t *s2, size_t n)); +# endif +_GL_CXXALIAS_SYS (wcsncasecmp, int, + (const wchar_t *s1, const wchar_t *s2, size_t n)); +_GL_CXXALIASWARN (wcsncasecmp); +#elif defined GNULIB_POSIXCHECK +# undef wcsncasecmp +# if HAVE_RAW_DECL_WCSNCASECMP +_GL_WARN_ON_USE (wcsncasecmp, "wcsncasecmp is unportable - " + "use gnulib module wcsncasecmp for portability"); +# endif +#endif + + +/* Compare S1 and S2, both interpreted as appropriate to the LC_COLLATE + category of the current locale. */ +#if @GNULIB_WCSCOLL@ +# if !@HAVE_WCSCOLL@ +_GL_FUNCDECL_SYS (wcscoll, int, (const wchar_t *s1, const wchar_t *s2)); +# endif +_GL_CXXALIAS_SYS (wcscoll, int, (const wchar_t *s1, const wchar_t *s2)); +_GL_CXXALIASWARN (wcscoll); +#elif defined GNULIB_POSIXCHECK +# undef wcscoll +# if HAVE_RAW_DECL_WCSCOLL +_GL_WARN_ON_USE (wcscoll, "wcscoll is unportable - " + "use gnulib module wcscoll for portability"); +# endif +#endif + + +/* Transform S2 into array pointed to by S1 such that if wcscmp is applied + to two transformed strings the result is the as applying 'wcscoll' to the + original strings. */ +#if @GNULIB_WCSXFRM@ +# if !@HAVE_WCSXFRM@ +_GL_FUNCDECL_SYS (wcsxfrm, size_t, (wchar_t *s1, const wchar_t *s2, size_t n)); +# endif +_GL_CXXALIAS_SYS (wcsxfrm, size_t, (wchar_t *s1, const wchar_t *s2, size_t n)); +_GL_CXXALIASWARN (wcsxfrm); +#elif defined GNULIB_POSIXCHECK +# undef wcsxfrm +# if HAVE_RAW_DECL_WCSXFRM +_GL_WARN_ON_USE (wcsxfrm, "wcsxfrm is unportable - " + "use gnulib module wcsxfrm for portability"); +# endif +#endif + + +/* Duplicate S, returning an identical malloc'd string. */ +#if @GNULIB_WCSDUP@ +# if !@HAVE_WCSDUP@ +_GL_FUNCDECL_SYS (wcsdup, wchar_t *, (const wchar_t *s)); +# endif +_GL_CXXALIAS_SYS (wcsdup, wchar_t *, (const wchar_t *s)); +_GL_CXXALIASWARN (wcsdup); +#elif defined GNULIB_POSIXCHECK +# undef wcsdup +# if HAVE_RAW_DECL_WCSDUP +_GL_WARN_ON_USE (wcsdup, "wcsdup is unportable - " + "use gnulib module wcsdup for portability"); +# endif +#endif + + +/* Find the first occurrence of WC in WCS. */ +#if @GNULIB_WCSCHR@ +# if !@HAVE_WCSCHR@ +_GL_FUNCDECL_SYS (wcschr, wchar_t *, (const wchar_t *wcs, wchar_t wc)); +# endif + /* On some systems, this function is defined as an overloaded function: + extern "C++" { + const wchar_t * std::wcschr (const wchar_t *, wchar_t); + wchar_t * std::wcschr (wchar_t *, wchar_t); + } */ +_GL_CXXALIAS_SYS_CAST2 (wcschr, + wchar_t *, (const wchar_t *, wchar_t), + const wchar_t *, (const wchar_t *, wchar_t)); +# if ((__GLIBC__ == 2 && __GLIBC_MINOR__ >= 10) && !defined __UCLIBC__) \ + && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4)) +_GL_CXXALIASWARN1 (wcschr, wchar_t *, (wchar_t *wcs, wchar_t wc)); +_GL_CXXALIASWARN1 (wcschr, const wchar_t *, (const wchar_t *wcs, wchar_t wc)); +# else +_GL_CXXALIASWARN (wcschr); +# endif +#elif defined GNULIB_POSIXCHECK +# undef wcschr +# if HAVE_RAW_DECL_WCSCHR +_GL_WARN_ON_USE (wcschr, "wcschr is unportable - " + "use gnulib module wcschr for portability"); +# endif +#endif + + +/* Find the last occurrence of WC in WCS. */ +#if @GNULIB_WCSRCHR@ +# if !@HAVE_WCSRCHR@ +_GL_FUNCDECL_SYS (wcsrchr, wchar_t *, (const wchar_t *wcs, wchar_t wc)); +# endif + /* On some systems, this function is defined as an overloaded function: + extern "C++" { + const wchar_t * std::wcsrchr (const wchar_t *, wchar_t); + wchar_t * std::wcsrchr (wchar_t *, wchar_t); + } */ +_GL_CXXALIAS_SYS_CAST2 (wcsrchr, + wchar_t *, (const wchar_t *, wchar_t), + const wchar_t *, (const wchar_t *, wchar_t)); +# if ((__GLIBC__ == 2 && __GLIBC_MINOR__ >= 10) && !defined __UCLIBC__) \ + && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4)) +_GL_CXXALIASWARN1 (wcsrchr, wchar_t *, (wchar_t *wcs, wchar_t wc)); +_GL_CXXALIASWARN1 (wcsrchr, const wchar_t *, (const wchar_t *wcs, wchar_t wc)); +# else +_GL_CXXALIASWARN (wcsrchr); +# endif +#elif defined GNULIB_POSIXCHECK +# undef wcsrchr +# if HAVE_RAW_DECL_WCSRCHR +_GL_WARN_ON_USE (wcsrchr, "wcsrchr is unportable - " + "use gnulib module wcsrchr for portability"); +# endif +#endif + + +/* Return the length of the initial segmet of WCS which consists entirely + of wide characters not in REJECT. */ +#if @GNULIB_WCSCSPN@ +# if !@HAVE_WCSCSPN@ +_GL_FUNCDECL_SYS (wcscspn, size_t, (const wchar_t *wcs, const wchar_t *reject)); +# endif +_GL_CXXALIAS_SYS (wcscspn, size_t, (const wchar_t *wcs, const wchar_t *reject)); +_GL_CXXALIASWARN (wcscspn); +#elif defined GNULIB_POSIXCHECK +# undef wcscspn +# if HAVE_RAW_DECL_WCSCSPN +_GL_WARN_ON_USE (wcscspn, "wcscspn is unportable - " + "use gnulib module wcscspn for portability"); +# endif +#endif + + +/* Return the length of the initial segmet of WCS which consists entirely + of wide characters in ACCEPT. */ +#if @GNULIB_WCSSPN@ +# if !@HAVE_WCSSPN@ +_GL_FUNCDECL_SYS (wcsspn, size_t, (const wchar_t *wcs, const wchar_t *accept)); +# endif +_GL_CXXALIAS_SYS (wcsspn, size_t, (const wchar_t *wcs, const wchar_t *accept)); +_GL_CXXALIASWARN (wcsspn); +#elif defined GNULIB_POSIXCHECK +# undef wcsspn +# if HAVE_RAW_DECL_WCSSPN +_GL_WARN_ON_USE (wcsspn, "wcsspn is unportable - " + "use gnulib module wcsspn for portability"); +# endif +#endif + + +/* Find the first occurrence in WCS of any character in ACCEPT. */ +#if @GNULIB_WCSPBRK@ +# if !@HAVE_WCSPBRK@ +_GL_FUNCDECL_SYS (wcspbrk, wchar_t *, + (const wchar_t *wcs, const wchar_t *accept)); +# endif + /* On some systems, this function is defined as an overloaded function: + extern "C++" { + const wchar_t * std::wcspbrk (const wchar_t *, const wchar_t *); + wchar_t * std::wcspbrk (wchar_t *, const wchar_t *); + } */ +_GL_CXXALIAS_SYS_CAST2 (wcspbrk, + wchar_t *, (const wchar_t *, const wchar_t *), + const wchar_t *, (const wchar_t *, const wchar_t *)); +# if ((__GLIBC__ == 2 && __GLIBC_MINOR__ >= 10) && !defined __UCLIBC__) \ + && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4)) +_GL_CXXALIASWARN1 (wcspbrk, wchar_t *, + (wchar_t *wcs, const wchar_t *accept)); +_GL_CXXALIASWARN1 (wcspbrk, const wchar_t *, + (const wchar_t *wcs, const wchar_t *accept)); +# else +_GL_CXXALIASWARN (wcspbrk); +# endif +#elif defined GNULIB_POSIXCHECK +# undef wcspbrk +# if HAVE_RAW_DECL_WCSPBRK +_GL_WARN_ON_USE (wcspbrk, "wcspbrk is unportable - " + "use gnulib module wcspbrk for portability"); +# endif +#endif + + +/* Find the first occurrence of NEEDLE in HAYSTACK. */ +#if @GNULIB_WCSSTR@ +# if !@HAVE_WCSSTR@ +_GL_FUNCDECL_SYS (wcsstr, wchar_t *, + (const wchar_t *haystack, const wchar_t *needle)); +# endif + /* On some systems, this function is defined as an overloaded function: + extern "C++" { + const wchar_t * std::wcsstr (const wchar_t *, const wchar_t *); + wchar_t * std::wcsstr (wchar_t *, const wchar_t *); + } */ +_GL_CXXALIAS_SYS_CAST2 (wcsstr, + wchar_t *, (const wchar_t *, const wchar_t *), + const wchar_t *, (const wchar_t *, const wchar_t *)); +# if ((__GLIBC__ == 2 && __GLIBC_MINOR__ >= 10) && !defined __UCLIBC__) \ + && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4)) +_GL_CXXALIASWARN1 (wcsstr, wchar_t *, + (wchar_t *haystack, const wchar_t *needle)); +_GL_CXXALIASWARN1 (wcsstr, const wchar_t *, + (const wchar_t *haystack, const wchar_t *needle)); +# else +_GL_CXXALIASWARN (wcsstr); +# endif +#elif defined GNULIB_POSIXCHECK +# undef wcsstr +# if HAVE_RAW_DECL_WCSSTR +_GL_WARN_ON_USE (wcsstr, "wcsstr is unportable - " + "use gnulib module wcsstr for portability"); +# endif +#endif + + +/* Divide WCS into tokens separated by characters in DELIM. */ +#if @GNULIB_WCSTOK@ +# if !@HAVE_WCSTOK@ +_GL_FUNCDECL_SYS (wcstok, wchar_t *, + (wchar_t *wcs, const wchar_t *delim, wchar_t **ptr)); +# endif +_GL_CXXALIAS_SYS (wcstok, wchar_t *, + (wchar_t *wcs, const wchar_t *delim, wchar_t **ptr)); +_GL_CXXALIASWARN (wcstok); +#elif defined GNULIB_POSIXCHECK +# undef wcstok +# if HAVE_RAW_DECL_WCSTOK +_GL_WARN_ON_USE (wcstok, "wcstok is unportable - " + "use gnulib module wcstok for portability"); +# endif +#endif + + +/* Determine number of column positions required for first N wide + characters (or fewer if S ends before this) in S. */ +#if @GNULIB_WCSWIDTH@ +# if @REPLACE_WCSWIDTH@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# undef wcswidth +# define wcswidth rpl_wcswidth +# endif +_GL_FUNCDECL_RPL (wcswidth, int, (const wchar_t *s, size_t n)); +_GL_CXXALIAS_RPL (wcswidth, int, (const wchar_t *s, size_t n)); +# else +# if !@HAVE_WCSWIDTH@ +_GL_FUNCDECL_SYS (wcswidth, int, (const wchar_t *s, size_t n)); +# endif +_GL_CXXALIAS_SYS (wcswidth, int, (const wchar_t *s, size_t n)); +# endif +_GL_CXXALIASWARN (wcswidth); +#elif defined GNULIB_POSIXCHECK +# undef wcswidth +# if HAVE_RAW_DECL_WCSWIDTH +_GL_WARN_ON_USE (wcswidth, "wcswidth is unportable - " + "use gnulib module wcswidth for portability"); +# endif +#endif + + +#endif /* _@GUARD_PREFIX@_WCHAR_H */ +#endif /* _@GUARD_PREFIX@_WCHAR_H */ +#endif diff --git a/lib/wcrtomb.c b/lib/wcrtomb.c new file mode 100644 index 0000000..9e6494e --- /dev/null +++ b/lib/wcrtomb.c @@ -0,0 +1,53 @@ +/* Convert wide character to multibyte character. + Copyright (C) 2008-2012 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2008. + + This program 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. + + This program 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 this program. If not, see <http://www.gnu.org/licenses/>. */ + +#include <config.h> + +/* Specification. */ +#include <wchar.h> + +#include <errno.h> +#include <stdlib.h> + + +size_t +wcrtomb (char *s, wchar_t wc, mbstate_t *ps) +{ + /* This implementation of wcrtomb on top of wctomb() supports only + stateless encodings. ps must be in the initial state. */ + if (ps != NULL && !mbsinit (ps)) + { + errno = EINVAL; + return (size_t)(-1); + } + + if (s == NULL) + /* We know the NUL wide character corresponds to the NUL character. */ + return 1; + else + { + int ret = wctomb (s, wc); + + if (ret >= 0) + return ret; + else + { + errno = EILSEQ; + return (size_t)(-1); + } + } +} diff --git a/lib/wctype.in.h b/lib/wctype.in.h new file mode 100644 index 0000000..aa018b1 --- /dev/null +++ b/lib/wctype.in.h @@ -0,0 +1,498 @@ +/* A substitute for ISO C99 <wctype.h>, for platforms that lack it. + + Copyright (C) 2006-2012 Free Software Foundation, Inc. + + This program 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, or (at your option) + any later version. + + This program 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 this program; if not, see <http://www.gnu.org/licenses/>. */ + +/* Written by Bruno Haible and Paul Eggert. */ + +/* + * ISO C 99 <wctype.h> for platforms that lack it. + * <http://www.opengroup.org/susv3xbd/wctype.h.html> + * + * iswctype, towctrans, towlower, towupper, wctrans, wctype, + * wctrans_t, and wctype_t are not yet implemented. + */ + +#ifndef _@GUARD_PREFIX@_WCTYPE_H + +#if __GNUC__ >= 3 +@PRAGMA_SYSTEM_HEADER@ +#endif +@PRAGMA_COLUMNS@ + +#if @HAVE_WINT_T@ +/* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. + 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> +#endif + +/* Include the original <wctype.h> if it exists. + BeOS 5 has the functions but no <wctype.h>. */ +/* The include_next requires a split double-inclusion guard. */ +#if @HAVE_WCTYPE_H@ +# @INCLUDE_NEXT@ @NEXT_WCTYPE_H@ +#endif + +#ifndef _@GUARD_PREFIX@_WCTYPE_H +#define _@GUARD_PREFIX@_WCTYPE_H + +/* The definitions of _GL_FUNCDECL_RPL etc. are copied here. */ + +/* The definition of _GL_WARN_ON_USE is copied here. */ + +/* Solaris 2.6 <wctype.h> includes <widec.h> which includes <euc.h> which + #defines a number of identifiers in the application namespace. Revert + these #defines. */ +#ifdef __sun +# undef multibyte +# undef eucw1 +# undef eucw2 +# undef eucw3 +# undef scrw1 +# undef scrw2 +# undef scrw3 +#endif + +/* Define wint_t and WEOF. (Also done in wchar.in.h.) */ +#if !@HAVE_WINT_T@ && !defined wint_t +# define wint_t int +# ifndef WEOF +# define WEOF -1 +# endif +#else +/* MSVC defines wint_t as 'unsigned short' in <crtdefs.h>. + This is too small: ISO C 99 section 7.24.1.(2) says that wint_t must be + "unchanged by default argument promotions". Override it. */ +# if defined _MSC_VER +# if !GNULIB_defined_wint_t +# include <crtdefs.h> +typedef unsigned int rpl_wint_t; +# undef wint_t +# define wint_t rpl_wint_t +# define GNULIB_defined_wint_t 1 +# endif +# endif +# ifndef WEOF +# define WEOF ((wint_t) -1) +# endif +#endif + + +#if !GNULIB_defined_wctype_functions + +/* FreeBSD 4.4 to 4.11 has <wctype.h> but lacks the functions. + Linux libc5 has <wctype.h> and the functions but they are broken. + Assume all 11 functions (all isw* except iswblank) are implemented the + same way, or not at all. */ +# if ! @HAVE_ISWCNTRL@ || @REPLACE_ISWCNTRL@ + +/* IRIX 5.3 has macros but no functions, its isw* macros refer to an + undefined variable _ctmp_ and to <ctype.h> macros like _P, and they + refer to system functions like _iswctype that are not in the + standard C library. Rather than try to get ancient buggy + implementations like this to work, just disable them. */ +# undef iswalnum +# undef iswalpha +# undef iswblank +# undef iswcntrl +# undef iswdigit +# undef iswgraph +# undef iswlower +# undef iswprint +# undef iswpunct +# undef iswspace +# undef iswupper +# undef iswxdigit +# undef towlower +# undef towupper + +/* Linux libc5 has <wctype.h> and the functions but they are broken. */ +# if @REPLACE_ISWCNTRL@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define iswalnum rpl_iswalnum +# define iswalpha rpl_iswalpha +# define iswblank rpl_iswblank +# define iswcntrl rpl_iswcntrl +# define iswdigit rpl_iswdigit +# define iswgraph rpl_iswgraph +# define iswlower rpl_iswlower +# define iswprint rpl_iswprint +# define iswpunct rpl_iswpunct +# define iswspace rpl_iswspace +# define iswupper rpl_iswupper +# define iswxdigit rpl_iswxdigit +# endif +# endif +# if @REPLACE_TOWLOWER@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define towlower rpl_towlower +# define towupper rpl_towupper +# endif +# endif + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswalnum +# else +iswalnum +# endif + (wint_t wc) +{ + return ((wc >= '0' && wc <= '9') + || ((wc & ~0x20) >= 'A' && (wc & ~0x20) <= 'Z')); +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswalpha +# else +iswalpha +# endif + (wint_t wc) +{ + return (wc & ~0x20) >= 'A' && (wc & ~0x20) <= 'Z'; +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswblank +# else +iswblank +# endif + (wint_t wc) +{ + return wc == ' ' || wc == '\t'; +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswcntrl +# else +iswcntrl +# endif + (wint_t wc) +{ + return (wc & ~0x1f) == 0 || wc == 0x7f; +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswdigit +# else +iswdigit +# endif + (wint_t wc) +{ + return wc >= '0' && wc <= '9'; +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswgraph +# else +iswgraph +# endif + (wint_t wc) +{ + return wc >= '!' && wc <= '~'; +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswlower +# else +iswlower +# endif + (wint_t wc) +{ + return wc >= 'a' && wc <= 'z'; +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswprint +# else +iswprint +# endif + (wint_t wc) +{ + return wc >= ' ' && wc <= '~'; +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswpunct +# else +iswpunct +# endif + (wint_t wc) +{ + return (wc >= '!' && wc <= '~' + && !((wc >= '0' && wc <= '9') + || ((wc & ~0x20) >= 'A' && (wc & ~0x20) <= 'Z'))); +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswspace +# else +iswspace +# endif + (wint_t wc) +{ + return (wc == ' ' || wc == '\t' + || wc == '\n' || wc == '\v' || wc == '\f' || wc == '\r'); +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswupper +# else +iswupper +# endif + (wint_t wc) +{ + return wc >= 'A' && wc <= 'Z'; +} + +static inline int +# if @REPLACE_ISWCNTRL@ +rpl_iswxdigit +# else +iswxdigit +# endif + (wint_t wc) +{ + return ((wc >= '0' && wc <= '9') + || ((wc & ~0x20) >= 'A' && (wc & ~0x20) <= 'F')); +} + +static inline wint_t +# if @REPLACE_TOWLOWER@ +rpl_towlower +# else +towlower +# endif + (wint_t wc) +{ + return (wc >= 'A' && wc <= 'Z' ? wc - 'A' + 'a' : wc); +} + +static inline wint_t +# if @REPLACE_TOWLOWER@ +rpl_towupper +# else +towupper +# endif + (wint_t wc) +{ + return (wc >= 'a' && wc <= 'z' ? wc - 'a' + 'A' : wc); +} + +# elif @GNULIB_ISWBLANK@ && (! @HAVE_ISWBLANK@ || @REPLACE_ISWBLANK@) +/* Only the iswblank function is missing. */ + +# if @REPLACE_ISWBLANK@ +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define iswblank rpl_iswblank +# endif +_GL_FUNCDECL_RPL (iswblank, int, (wint_t wc)); +# else +_GL_FUNCDECL_SYS (iswblank, int, (wint_t wc)); +# endif + +# endif + +# if defined __MINGW32__ + +/* On native Windows, wchar_t is uint16_t, and wint_t is uint32_t. + The functions towlower and towupper are implemented in the MSVCRT library + to take a wchar_t argument and return a wchar_t result. mingw declares + these functions to take a wint_t argument and return a wint_t result. + This means that: + 1. When the user passes an argument outside the range 0x0000..0xFFFF, the + function will look only at the lower 16 bits. This is allowed according + to POSIX. + 2. The return value is returned in the lower 16 bits of the result register. + The upper 16 bits are random: whatever happened to be in that part of the + result register. We need to fix this by adding a zero-extend from + wchar_t to wint_t after the call. */ + +static inline wint_t +rpl_towlower (wint_t wc) +{ + return (wint_t) (wchar_t) towlower (wc); +} +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define towlower rpl_towlower +# endif + +static inline wint_t +rpl_towupper (wint_t wc) +{ + return (wint_t) (wchar_t) towupper (wc); +} +# if !(defined __cplusplus && defined GNULIB_NAMESPACE) +# define towupper rpl_towupper +# endif + +# endif /* __MINGW32__ */ + +# define GNULIB_defined_wctype_functions 1 +#endif + +#if @REPLACE_ISWCNTRL@ +_GL_CXXALIAS_RPL (iswalnum, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswalpha, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswcntrl, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswdigit, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswgraph, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswlower, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswprint, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswpunct, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswspace, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswupper, int, (wint_t wc)); +_GL_CXXALIAS_RPL (iswxdigit, int, (wint_t wc)); +#else +_GL_CXXALIAS_SYS (iswalnum, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswalpha, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswcntrl, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswdigit, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswgraph, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswlower, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswprint, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswpunct, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswspace, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswupper, int, (wint_t wc)); +_GL_CXXALIAS_SYS (iswxdigit, int, (wint_t wc)); +#endif +_GL_CXXALIASWARN (iswalnum); +_GL_CXXALIASWARN (iswalpha); +_GL_CXXALIASWARN (iswcntrl); +_GL_CXXALIASWARN (iswdigit); +_GL_CXXALIASWARN (iswgraph); +_GL_CXXALIASWARN (iswlower); +_GL_CXXALIASWARN (iswprint); +_GL_CXXALIASWARN (iswpunct); +_GL_CXXALIASWARN (iswspace); +_GL_CXXALIASWARN (iswupper); +_GL_CXXALIASWARN (iswxdigit); + +#if @GNULIB_ISWBLANK@ +# if @REPLACE_ISWCNTRL@ || @REPLACE_ISWBLANK@ +_GL_CXXALIAS_RPL (iswblank, int, (wint_t wc)); +# else +_GL_CXXALIAS_SYS (iswblank, int, (wint_t wc)); +# endif +_GL_CXXALIASWARN (iswblank); +#endif + +#if !@HAVE_WCTYPE_T@ +# if !GNULIB_defined_wctype_t +typedef void * wctype_t; +# define GNULIB_defined_wctype_t 1 +# endif +#endif + +/* Get a descriptor for a wide character property. */ +#if @GNULIB_WCTYPE@ +# if !@HAVE_WCTYPE_T@ +_GL_FUNCDECL_SYS (wctype, wctype_t, (const char *name)); +# endif +_GL_CXXALIAS_SYS (wctype, wctype_t, (const char *name)); +_GL_CXXALIASWARN (wctype); +#elif defined GNULIB_POSIXCHECK +# undef wctype +# if HAVE_RAW_DECL_WCTYPE +_GL_WARN_ON_USE (wctype, "wctype is unportable - " + "use gnulib module wctype for portability"); +# endif +#endif + +/* Test whether a wide character has a given property. + The argument WC must be either a wchar_t value or WEOF. + The argument DESC must have been returned by the wctype() function. */ +#if @GNULIB_ISWCTYPE@ +# if !@HAVE_WCTYPE_T@ +_GL_FUNCDECL_SYS (iswctype, int, (wint_t wc, wctype_t desc)); +# endif +_GL_CXXALIAS_SYS (iswctype, int, (wint_t wc, wctype_t desc)); +_GL_CXXALIASWARN (iswctype); +#elif defined GNULIB_POSIXCHECK +# undef iswctype +# if HAVE_RAW_DECL_ISWCTYPE +_GL_WARN_ON_USE (iswctype, "iswctype is unportable - " + "use gnulib module iswctype for portability"); +# endif +#endif + +#if @REPLACE_TOWLOWER@ || defined __MINGW32__ +_GL_CXXALIAS_RPL (towlower, wint_t, (wint_t wc)); +_GL_CXXALIAS_RPL (towupper, wint_t, (wint_t wc)); +#else +_GL_CXXALIAS_SYS (towlower, wint_t, (wint_t wc)); +_GL_CXXALIAS_SYS (towupper, wint_t, (wint_t wc)); +#endif +_GL_CXXALIASWARN (towlower); +_GL_CXXALIASWARN (towupper); + +#if !@HAVE_WCTRANS_T@ +# if !GNULIB_defined_wctrans_t +typedef void * wctrans_t; +# define GNULIB_defined_wctrans_t 1 +# endif +#endif + +/* Get a descriptor for a wide character case conversion. */ +#if @GNULIB_WCTRANS@ +# if !@HAVE_WCTRANS_T@ +_GL_FUNCDECL_SYS (wctrans, wctrans_t, (const char *name)); +# endif +_GL_CXXALIAS_SYS (wctrans, wctrans_t, (const char *name)); +_GL_CXXALIASWARN (wctrans); +#elif defined GNULIB_POSIXCHECK +# undef wctrans +# if HAVE_RAW_DECL_WCTRANS +_GL_WARN_ON_USE (wctrans, "wctrans is unportable - " + "use gnulib module wctrans for portability"); +# endif +#endif + +/* Perform a given case conversion on a wide character. + The argument WC must be either a wchar_t value or WEOF. + The argument DESC must have been returned by the wctrans() function. */ +#if @GNULIB_TOWCTRANS@ +# if !@HAVE_WCTRANS_T@ +_GL_FUNCDECL_SYS (towctrans, wint_t, (wint_t wc, wctrans_t desc)); +# endif +_GL_CXXALIAS_SYS (towctrans, wint_t, (wint_t wc, wctrans_t desc)); +_GL_CXXALIASWARN (towctrans); +#elif defined GNULIB_POSIXCHECK +# undef towctrans +# if HAVE_RAW_DECL_TOWCTRANS +_GL_WARN_ON_USE (towctrans, "towctrans is unportable - " + "use gnulib module towctrans for portability"); +# endif +#endif + + +#endif /* _@GUARD_PREFIX@_WCTYPE_H */ +#endif /* _@GUARD_PREFIX@_WCTYPE_H */ |